The crippling homunculus: inefficient executive control and the persistence of intrusive memories

THE CRIPPLING HOMUNCULUS Inefficient executive control and the persistence of intrusive memories

ISBN: 9789036737258 Author: Johan Rokus Leendert Verwoerd Cover illustration: Image of an artistic representation of trauma (www.mentalhealth.about.com/library/mh/anx/blrelive.htm) Printed by: Offsetdrukkerij Ridderprint B.V., Ridderkerk © 2009, J.R.L. Verwoerd

RIJKSUNIVERSITEIT GRONINGEN

THE CRIPPLING HOMUNCULUS Inefficient executive control and the persistence of intrusive memories

Proefschrift ter verkrijging van het doctoraat in de Gedrags- en Maatschappijwetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op donderdag 26 februari 2009 om 14:45 uur door Johan Rokus Leendert Verwoerd geboren op 5 april 1975 te Woerden

Promotor: Prof. dr. P. J. de Jong Copromotor: Dr. J. P. Wessel Beoordelingscommissie: Prof. dr. A. Arntz Prof. dr. R. de Jong Prof. dr. R. de Raedt

Contents Chapter 1 General Introduction ............................................................................................................7

Chapter 2 Distractibility and individual differences in the experience of intrusive memories ..........27

Chapter 3 Individual Differences in Experiencing Intrusive Memories: The Role of the Ability to Resist Proactive Interference .............................................................................................39

Chapter 4 Low Attentional Control and the Development of Intrusive Memories Following a Laboratory Stressor............................................................................................................61

Chapter 5 Pre-Stressor Interference Control and Intrusive Memories ...............................................75

Chapter 6 Preferential Processing of Visual Trauma-Film Reminders Predicts Subsequent Intrusive Memories ...........................................................................................................................95

Chapter 7 Summary and discussion..................................................................................................111 References ...................................................................................................................................123 Samenvatting ..............................................................................................................................135 Dankwoord .................................................................................................................................147 Curriculum Vitae .......................................................................................................................151

v

vi

Chapter 1 General introduction

-7-

Chapter 1

‘about two years ago, I was dully walking along a Sydney street, worrying about nothing in particular, when – boom! I was small. I was turning the corner past the tankstand and putting my hand out to push open the back door of the Ocean Grove house. It was made of diagonal latticed slats painted dark green and rough to the touch. I nearly keeled over with the vividness of that door. It came from nowhere. It hadn’t entered my mind for over forty years.’ - Helen Garner An old black and white photo, stuck at one corner to the bottom of the drawer. I ease it out carefully, not to tear it. Its shiny surface is covered in tiny cracks, but the picture’s still quite clear. As I gaze at it, my heart starts beating painfully. All those faces, once so intimately, so intensely known. - Phil Salmon

Memory can sometimes present itself like a great mystery. Returning to childhood within a fraction of a second, sitting on the couch with your first love, lost nostalgic moments affected by the passage of time suddenly return with great clarity and physical reliving. This class of involuntary memories possesses a kind of intrinsic beauty for the rememberer which inspired the great novelist and memory-explorer Marcel Proust to formulate his famous Petite Madeleine cookie example. Not surprisingly, these memories are also known as aesthetic memories (Berntsen, 2007). A completely different class of involuntary remembering involves the stressful intrusions after the experience of a traumatic event. According to the most recent version of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; APA, 1994), a traumatic event is defined as an experience where a person “experienced, witnessed, or was confronted with an event or events that involved actual or threatened death or serious injury, or a threat to the physical integrity of self or others” with a subjective response of “intense fear, helplessness or horror” (APA, 1994, pp. 427-428). These horrific experiences will often result in stressful memories more likely to be described as poison instead of joy. For people who develop a posttraumatic stress disorder (PTSD) after trauma, the repeated experiences of intrusive memories are felt as a heavy burden which could deregulate everyday functioning (e.g., work, family) for decades. Prominent examples are cases of holocaust survivors,

-8-

Vietnam veterans and people confronted with sexual abuse in their childhood who remain stuck in the past and become prisoners of their own persistent memories (cf. Schacter, 2001). The psychological symptoms of PTSD comprise: (i) recurrent intrusive distressing recollections of the traumatic event, (ii) avoidance of trauma-related stimuli and/or a general numbing of emotional responsiveness; and (iii) increased levels of arousal (Steel & Holmes, 2007; APA, 1994). Ample evidence has shown that traumatic events appear to be relatively common experiences. For example, the National Comorbidity Survey (NCS) under more than 5800 American citizens aged 15-54 showed prevalence rates of 60.7% of male and 51.2% of female respondents having experienced at least one DSM-III-R traumatic event in their lifetime (Kessler et al., 1995). Another study using the more liberal DSM-IV stressor definition uncovered rates of trauma exposure of 89.6% (Breslau et al., 1998, cited in Duke & Vasterling, 2005). Finally, in a study among 900 US college students, the prevalence rate of having experienced a potentially traumatic event was found to be around 67 % (Bernat, Ronfeldt, Calhoun, & Arias, 1998). In spite of this high prevalence of exposure to traumatic stressors, relatively few people exposed to trauma will subsequently develop PTSD. For example, among adults living in metropolitan Detroit, 89.6% reported exposure to DSM traumatic stressors, yet only 13% of the women and 6.2% of the men had developed PTSD (Breslau, Davis, Andreski & Peterson, 1991, cited in McNally, Bryant & Ehlers, 2003). This indicates that most trauma survivors are well able to reduce the occurrence of stressful intrusive memories in the first months after trauma exposure. Cognitive models of PTSD have based their ideas regarding dysfunctional processing in response to a traumatic event on theories in the experimental memory and attention literature. The underlying idea is that the wealth of knowledge about basic memory and attention processes that has been acquired in the past decades may help explaining the nature of trauma-related intrusions after the experience of a traumatic event. Prevailing models (e.g., Ehlers and Clark, 2000; Brewin, Power & Dalgleish, 1996) have explained persistent intrusions in such a way that dysfunctional information processing during trauma would result in a weakly integrated representation of the traumatic event in the autobiographical knowledge base (cf. Conway & Pleydell-Pearce, 2000). That is, Conway and colleagues (Conway & Pleydell-Pearce, 2000; Conway, Singer, & Tagini, 2004) suggested that individual autobiographical memories are embedded in higher level structures of life themes (e.g., childhood, work, marriage) and general events ( e.g., holiday in Spain, conference in

-9-

Chapter 1

Finland) which together form an integrated autobiographical knowledge base. Weakened integration of the trauma memory in this knowledge base may facilitate the triggering of intrusive memories after trauma and possibly results in the inability to obtain control over these stressful experiences. Although these models are informative in describing what may go wrong when people are confronted with traumatic events, it remains an open question why some individuals show a strong tendency towards dysfunctional information processing in response to trauma but others do not. A large body of work on risk factors has not yet resolved this more specific issue (see Brewin, Andrews, & Valentine, 2000; Ozer, Best, Lipsey, & Weiss, 2003 for reviews). This crucial why-question points to the possible role of innate or early acquired individual differences in basic memory and attention processes. In the experimental memory literature, the importance of individual differences in explaining human cognition and behavior has now been widely acknowledged (e.g., Unsworth & Engle, 2007). Theories of working memory have included a homunculus-like supervisory component that may be responsible for differential performance on a wide range of cognitive tasks and behaviors. A collection of theoretical models has given different headers to this component and its functions, such as central executive (Baddeley, 1996), supervisory attentional control (Atkinson & Shiffrin, 1968), attentional control (Unsworth & Engle, 2007) and executive control (Friedman et al., 2008). The central premise of these theories is that people are born with or grow up in early life with differential resources in executive/attentional ability (Friedman & Miyake, 2008; Engle, 1995). This raises the questions about the implications of individual differences in this basic ability for functioning in everyday life, especially, when one is being confronted with intrusive trauma-reminders. According to working memory models, a well-developed executive ability supports active maintenance of goal-directed information and resistance or inhibition of irrelevant information (Unsworth & Engle, 2007). Thus, executive ability supports flexible control over the contents of consciousness. This would be crucial at times when the risk of being overwhelmed with emotional information during and after a traumatic experience grows. Following this line of reasoning, it can be speculated that a deficit in this basic ability would set people at risk for the persistence of intrusive memories after trauma. A stepwise investigation of this putative prospective relationship between pre morbid executive ability and intrusive memories will be the major theme of the present thesis.

- 10 -

Because ethical considerations prohibit the exposure of people to real traumatic events, the main questions related to pre morbid cognitive vulnerability and intrusive memories are examined with the methodology of presenting non-selected participants with trauma films as an experimental analogue of real-life trauma. The central idea behind this methodology is that analogue intrusions elicited by trauma-films and intrusive memories after trauma represent qualitatively similar kinds of memory that only differ in intensity. That is, psychopathological intrusions can be viewed as an extension of a continuum from our common, everyday intrusions (see Holmes & Bourne, 2008). This general introduction is build op as follows. Before introducing the main research questions presented in the empirical studies, I will first discuss the concepts that are critical for the present research project. Following a short definition of the contents and qualities of intrusive memories in the clinical literature, I will discuss how two prominent information-processing accounts of PTSD explain the development and maintenance of intrusive memories: the cognitive model of Ehlers and Clark (2000) and dual representation theory (Brewin, Power & Dalgleish, 1996). Next, relevant models of executive control and working memory are reviewed. In the experimental literature, there is some consensus that executive control may not be conceptualized as a unitary construct, but that it consists of different sub-functions, each with their own involvement in regulating particular aspects of goal-directed behavior. This will result in a specification of the earlier suggested relationship between pre morbid executive control and intrusive memories. Following this review, I will introduce a model depicturing underlying mechanisms for persistent intrusive memories after trauma. Based on this model, the general outline of this thesis is presented with a short overview of the main research questions addressed in the subsequent empirical chapters.

- 11 -

Chapter 1

1.1. How do cognitive models of PTSD explain image-based intrusive memories? Intrusive memories can be defined as vivid sensory-perceptual memories of earlier stressful or traumatic events. For example, Ehlers et al. (2004) describe the case of a patient who kept seeing headlights coming towards him (as he had seen them shortly before his head-on car crash) (see also Ehlers & Steil, 1995; Ehlers et al., 2002). These unwanted memories enter consciousness involuntary and are often accompanied with a strong sense of reliving of the original emotions. Intrusive memories can be experienced in different modalities, but visual intrusions are most common (Ehlers et al., 2004). Intrusive memories can be differentiated from intrusive thoughts which are more likely to be experienced as verbal rumination (e.g., Ehlers et al., 2004). Although both intrusive memories and thoughts might be explained by similar underlying mechanisms, the focus of the present thesis will be on image-based memories. Several cognitive models have been put forward to address the core phenomena of PTSD. These models vary widely in their level of abstraction and number of representational elements or components that describe these phenomena (see Dalgleish, 2004 for a review). For example, higher level schema-based models try to deal with the integration of the subjective meaning of traumatic events into pre-existing knowledge structures (i.e., schemas) of the self and the world. In contrast, other models are more trauma-centered and try to explain the (pathological) nature of the trauma memory and its relationship with the persistence of trauma-related symptoms such as intrusive memories. Important to mention here is the clinical observation of a compelling difference between intentional/generative retrieval and direct/associative retrieval of the trauma memory. Intentional retrieval consists of a strategic and effortful search process for the recall of knowledge in long term memory (e.g., Conway & Pleydell-Pearce, 2000). For example, when you meet a person you recognize as someone you met regularly in college, it will often require an effortful strategic search process to come up with sensory-perceptual details of the specific moments you shared together. In contrast, direct retrieval is an associative and more automatic retrieval process in which there is direct match between a retrieval cue and the content of a specific representation in long term memory. In the context of trauma, intentional recall of the trauma memory is often fragmented and poorly organized with an additional difficulty of recalling the exact temporal order of events. On the other hand, direct retrieval of the trauma memory

- 12 -

results in intrusive reliving experiences of very specific moments during trauma that involuntary enter awareness in a very vivid and emotional way (Ehlers & Clark, 2000). In the following, I will discuss two prevailing trauma-centric cognitive models that have specifically focused on the imbalance between the degree of intentional and direct retrieval of the trauma memory in explaining persistent intrusive memories and PTSD. I will start with Ehlers and Clark’s (2000) cognitive model for persistent PTSD. This model, primarily intended to provide a theoretical context for the development of a new cognitive behavioral treatment for PTSD (Ehlers et al., 2003; Dalgleish, 2004), gives a detailed account of how dysfunctional encoding of the trauma memory may lead to a processing bias for cues that resemble the earlier traumatic event. This processing bias may then – among other things – be responsible for the unbidden occurrence (i.e., direct retrieval) of stressful intrusive memories after trauma. The second model is dual representation theory (DRT) of PTSD as developed by Brewin and colleagues (Brewin et al., 1996; Brewin, 2001). DRT has its main focus on the representation of the traumatic event in long term memory. The model accounts for the imbalance between the degree of intentional and direct retrieval after trauma by arguing that direct retrieval of image-based intrusive memories and intentional retrieval of autobiographical knowledge are supported by different memory systems in long term memory. Persistent intrusive memories are the result of a relatively large amount of information stored in the memory system supporting direct retrieval of the trauma memory.

Ehlers & Clark’s (2000) Cognitive Model of PTSD Central to Ehlers and Clark’s (2000) cognitive model for persistent PTSD is a sense of serious current threat. That is, in contrast with people who recover from trauma within a relatively short period of time, for people with persistent PTSD the world remains a dangerous place in which normal functioning is impossible. This appraisal of current threat is intimately linked with reliving experiences in the form of intrusive memories, hyperarousal symptoms and strong emotions that apparently come out of the blue. Intrusive memories are interpreted as a sign of an incomplete integration of the trauma memory into the general structure of autobiographical memory which supports direct retrieval of sensory-perceptual details. The model proposes that this lack of integration is the result of dysfunctional processing during the encoding of the traumatic event. In this respect, conceptually-driven

- 13 -

Chapter 1

(meaning-based) processing is differentiated from data-driven processing (i.e., processing sensory impressions). Conceptual processing is thought to support intentional retrieval of the trauma due to integration of trauma-related information with other autobiographical information whereas data-driven processing supports re-experiencing symptoms. The idea is that people with a tendency to encode the trauma in a sensory/perceptual way will show enhanced perceptual priming for stimuli with a temporal association with the traumatic event. Enhanced perceptual priming can be conceptualized as increased activation of sensory fragments of the trauma in long-term memory. This perceptual priming will support a biased processing of trauma-reminders in the external environment and the subsequent emotional reliving (via direct retrieval) in the form of distressing intrusive memories.

Brewin, Power & Dalgleish’ (1996) Dual representation theory Brewin et al.’s (1996; see Brewin, 2001) dual representation theory shares assumptions about encoding with the Ehlers and Clark (2000) model, but makes the additional claim that the representation of the trauma in long-term memory is distributed over two different memory systems. First, intentional retrieval of the traumatic event is supported by a Verbally Accessible Memory system (VAM). VAM memories have in common with normal autobiographical memories that they are accompanied with autonoetic awareness (a feeling of remembering; Tulving, 2002) and can be communicated. Second, involuntary intrusive memories are supported by a different Situational Accessible Memory system (SAM). SAM is a perceptually-based memory system that can be accessed by situational cues with a temporal association with the traumatic event. This memory system echoes earlier proposals for a perceptual representation system (PRS, Tulving & Schacter) in the literature on implicit memory. SAM-based intrusions lack contextual integration in autobiographical memory and are often accompanied with a strong sense of stressful reliving. Furthermore, this memory system is not accessible for voluntary retrieval/introspection. Brewin et al. (1996) argue that individual differences in the tendency to encode the traumatic event in a sensory/perceptual way will predict the amount of information stored in the SAM (for direct retrieval) and VAM systems (for intentional retrieval). Considering this, a relatively large amount of information stored in SAM would result in a relatively high frequency of image-based intrusive memories. The crucial point here is that recovery after trauma heavily depends on the ability to transport information stored in the SAM system into the VAM system. A more complete

- 14 -

VAM system will result in an integrated trauma memory which will ease intentional retrieval and significantly reduce the probability of intrusive re-experiencing. Both Ehlers & Clark’s (2000) cognitive model and DRT (Brewin et al., 1996) highlight aspects of dysfunctional information processing (i.e., encoding, storage) in patients with persistent PTSD. However, these models are less able of explaining individual differences in dysfunctional/healthy processing during the first weeks after a traumatic event. For example, it remains unclear which underlying mechanisms might help trauma survivors to transport information from the SAM to the VAM system during natural recovery. Similarly, it is unclear which underlying mechanisms are responsible for a gradual reduction in perceptual priming, intrusive memories and eventually, the sense of current threat in the aftermath of a traumatic event. In the present dissertation I propose a model that describes a role for individual differences in pre-trauma executive control in explaining the difference between healthy processing towards natural recovery or persistent pathology as described by both clinical models. Before presenting a more detailed account of this model, it is important to introduce some key concepts in the experimental memory literature and summarize relevant earlier findings in the neuropsychological literature.

1.2. Theories of working memory and executive control Working memory as a dynamic system instead of a passive store. According to recent conceptualizations, working memory is described as a set of temporary active LTM representations under supervision of an executive attentional controller. This attentional controller manipulates working memory contents in line with currently activated goals (Kane et al., 2007). This dynamic conception of active memory has important consequences for explaining individual differences. Important in this respect is the construct of working memory capacity (WMC). WMC has been defined as the ability to actively maintain or recover access to goal-relevant information in a wide range of contexts that also trigger irrelevant interfering responses/memories/behaviors. This use of the term capacity strongly differs from the view of primary memory as a passive store with a limited number of items or chucks (7 + 2; Miller, 1956, cited in Kane et al., 2007). WMC as ability depends on the amount of executive resources available for goal-directed behavior. Thus, WMC is just as important in the retention of a single representation of a goal as it is in determining how

- 15 -

Chapter 1

many representations can be maintained. WMC does not refer to memory, but to how executive attention is used to maintain or suppress information. WMC is usually measured with complex span tasks (e.g., Operation Span; Conway et al., 2005) that have been found to predict performance on a wide range of intelligent behaviors. These WM-span tasks present participants with the traditional memory span demand to immediately recall short lists of unrelated stimuli. Additionally and critically, these tasks also consist of a secondary processing component (to prevent rehearsal) through which the maintenance of memoranda is challenged. Operation span (e.g., Conway et al., 2005) for example requires subjects to read aloud and verify arithmetic equations in combination with the instruction to memorize unrelated words following the equations (e.g., IS 6 + 5 = 11? Yes/NO Ball). Empirical work has shown relationships between WM-span tasks and standard measures of intelligence (Kane et al., 2004), active goal maintenance during Stroop performance (“name the color, ignore the word”; Kane & Engle, 2003) and flexible target responding in an attentional orienting task (Unsworth & Engle, 2004). More importantly, Brewin and Smart (2005) showed a negative relationship between performance on a measure of WMC (i.e., Operation span) and the frequency of personally relevant intrusive thoughts using a thought suppression paradigm (Wenzlaff & Wegner, 2000). In general, these studies on WMC provide accumulating evidence for the prominent role of executive control in explaining individual differences on a wide range of every day behaviors. However, the literature on WMC does not speak about the possible existence of specialized executive abilities that would be independently involved in different aspects of cognition and behavior.

Executive control as a multi-component construct A separate literature (e.g., Friedman & Miyake, 2004) has used latent variable analyses to address the question of the unity and diversity of executive functions. The general idea behind these studies is that executive control can be defined as a group of semi-independent abilities, each responsible for a different aspect of behavior. For example, Miyake et al. (2000) divided executive functioning into three correlated but separable abilities: Shifting between task sets, Updating of working memory contents and Inhibition of prepotent response tendencies. Furthermore, these abilities were found to be separately related to performance on neuropsychological tasks of frontal functioning such as the Wisconsin Card Sorting test (WCST) and the Tower of Hanoi. Another division of executive functioning

- 16 -

which may be of particular importance for the present purposes is the distinction between inhibition at the response level and inhibition at the cognitive level. Friedman and Miyake (2004) examined the relationships between executive measures of inhibition of prepotent responses (Response Inhibition; e.g., the Stroop, stop-signal and antisaccade tasks; see also Miyake et al., 2000) and executive measures of resistance to proactive interference or cognitive inhibition (resistance to PI; indexed by list-learning paradigms). Response inhibition may be involved in preventing the natural but socially inappropriate tendency to say something embarrassing about a colleague’s new purple dress. As an example of resistance to PI, consider the first weeks after purchasing a new mobile phone. During this period, it would be difficult to access the new number in long term memory because the extensively used old number will persistently intrude into conscious awareness. Flexible access to the memory representation of the new number would require a well-developed ability to resist unwanted interference from the old number. This executive ability of resistance to PI seems to play an important role in explaining individual differences on complex measures of WMC (e.g., Kane & Engle, 2000; Rosen & Engle,1997, 1998). This could mean that part of the variance of well-known indices of WMC may be attributed to the specific ability to resist or inhibit interference from working memory. Considering this, in the study of Friedman and Miyake (2004) resistance to PI (but not response inhibition), showed a relationship with everyday self-reported intrusive thoughts. As trauma-related intrusive memories may be seen as a profound example of experiencing unwanted interference in real life, a deficit in the general ability to resist PI in working memory may set people at risk for persistent intrusive re-experiencing and PTSD. A large body of work in the neuropsychological literature has compared PTSD patients and trauma-exposed controls on a wide range of cognitive measures. Several studies which focused specifically on indices of proactive and retroactive interference, found recall deficits in PTSD patients when initial learning was taken into account (Uddo, 1993; Vasterling et al., 1998, 2001; Yehuda et al., 1995; see also Vasterling & Brailey, 2005). Moreover, Vasterling et al. (1998) found that the tendency to make intrusion errors on cognitive interference tasks was positively correlated with re-experiencing symptoms. These clinical results could be informative about underlying pathogenic mechanisms such as resistance to PI. However, they leave undecided whether neuropsychological abnormalities are a consequence of PTSD or represent a pre morbid vulnerability factor (e.g., Vasterling &

- 17 -

Chapter 1

Brailey, 2005). The studies in the present thesis were intended as a first step to solve this critical issue.

Inhibition theory Until now, I have discussed the role of executive control in the service of the ongoing processing of active goals in working memory. It is unclear however what would be the consequences of exerting control over material deemed as unwanted or irrelevant. When irrelevant/unwanted material such intrusive memories no longer interfere with the goals in working memory has it become less accessible, or less available? Inhibition theory of executive control deals with this important question by taking the activation metaphor (i.e., the idea that memories can become activated or de-activated) one step further. Inhibition theory as formulated by Anderson and coworkers (Anderson & Spellman, 1995; Levy & Anderson, 2002; 2008) proposes that a healthy functioning memory system needs an active ability to forget as well as the well-known ability to remember. At this point, it is important to note that there are two conceptualizations of the construct of cognitive inhibition. First, in its weaker sense, cognitive inhibition is described as an executive control function or ability to prevent irrelevant information from entering working memory (e.g., resistance to PI) (see Hasher et al., 1999; Lustig et al., 2001). Second, in a stronger sense, cognitive inhibition is described as the consequence of an activation reducing suppression mechanism. In the latter description, inhibition reflects a reduced activation level of the memory representations of earlier interfering material (see Anderson & Spellman, 1995). The central premise behind this idea is that a target memory can be made (temporarily) less available by suppression of its representation below a certain baseline state of activity in LTM. In the literature, two case examples of memory situations are described in which LTM (below baseline) inhibition would be involved: The need for selection during retrieval and the need to stop retrieval itself. As an example of selection during retrieval, imagine a situation in which a person’s goal is to retrieve the name of an old classmate after seeing an old high school picture. Although this retrieval cue seems quite specific, it will often not be specific enough to prevent the activation of several memory traces with potential candidate names. Inhibition theory states that in this situation, inhibitory mechanisms (i.e., executive abilities) are recruited to inhibit competing traces, resulting in the selection of the appropriate name

- 18 -

into consciousness awareness. In contrast, the inhibited competing traces are temporarily less available due to a below baseline activation level. This inhibition process during selective retrieval is widely known as “retrieval induced forgetting” (Anderson & Spellman, 1995). The need to stop retrieval arises when someone is confronted with a cue or reminders and does not want an associated memory come to mind. In case of trauma-reminders in the external environment, this would mean that repeated attempts to stop retrieval of the overwhelming intrusions would result in a gradual decrease in activation of the representation of the trauma memory (see Levy & Anderson, 2008 for a suggested link between inhibition in its stronger sense and trauma-related intrusive memories). In the past decade, empirical studies have tried to test the ideas behind the need for selection and the need to stop retrieval. Standard methodologies such as the retrieval practice paradigm for selective retrieval (Anderson, Bjork, & Bjork, 1994; Anderson & Spellman, 1995) and the think-no-think task (Anderson & Green, 2001) for stopping retrieval have been developed to test the crucial claim of below baseline activation of LTM representations. Although the ideas presented by inhibition theory may sound intriguing from the perspective of a memory theorist, they are in apparent contrast with some basic assumptions in the clinical literature. For example, in the clinical psychology literature, not talking about and trying to forget a traumatic experience is a first step towards development of psychopathology such as PTSD (i.e., avoidance; APA, 1994). One solution to these inconsistent views may be found in the proposal that motivated attempts at avoidance after a traumatic event are the expression of an inability to exert executive (inhibitory) control (see Levy & Anderson, 2008). Although this sounds plausible, it may be quite challenging to conduct an empirical test of this hypothesis in the context of real life intrusive memories. More importantly, before addressing questions about the role of inhibition in a stronger sense, it is important to establish whether there is, indeed, a link between executive control and image-based intrusive memories. Hence the main focus of the present thesis will be on the prospective relationship between pre-morbid executive control (i.e., resistance to PI) and the frequency of intrusive memories after a stressful or traumatic event.

- 19 -

Chapter 1

1.3. Direct retrieval of intrusive memories after trauma: underlying mechanisms and individual differences Studies investigating intrusive memories will almost by definition use subjective indices such as questionnaires, structured interviews and diaries. This may give information about several aspects of intrusive remembering such as the frequency over time, experienced modalities (e.g., visual, tactile, auditory) and unhealthy appraisals (“I am going mad”, “the world is no longer safe” [e.g., Ehlers et al., 2004]). However, a relatively unexplored aspect of intrusive re-experiencing is the retrieval process itself. What is the exact process by which retrieval cues are (automatically) selected and how does this cue interact with the memory contents of the intrusions? Increased knowledge of the mechanisms supporting the retrieval process of intrusive memories may lead to a better understanding how risk factors such as executive control may contribute to the persistence of these memories. Earlier in this introduction, I briefly introduced a model on the role of executive control in explaining natural recovery/persistent symptoms after a traumatic event. A more detailed account of this model – which heavily relies on an integration of different ideas presented in cognitive theories of PTSD - will be given below. Central to the retrieval of intrusive memories is the direct mapping of the cue and the memory contents. Simple cues in the external environment may be powerful enough to directly trigger stressful contents without the cue elaboration process needed for the intentional recall of everyday autobiographical memories (Conway & Pleydell-Pearce, 2000). In an attempt to explain why seemingly unimportant cues can obtain such a powerful intrinsic meaning, cognitive models of PTSD have used insights from the literature on implicit memory (priming; e.g., Tulving & Schacter, 1990), episodic memory (encoding specificity, transfer-appropriate processing; e.g., Tulving, 2002) and autobiographical memory (direct retrieval, weakly integrated representations; e.g., Conway & Pleydell-Pearce, 2000). Considering this, it has been argued that the expression of an implicit perceptual representation system (PRS) in long-term memory (i.e., perceptual priming [Tulving & Schacter, 1990]) might explain the maintenance of intrusive memories (Ehlers et al., 2006; Ehlers & Clark, 2000). This enhanced perceptual priming hypothesis (e.g., Ehlers et al., 2006; Michael & Ehlers, 2007) states that a primary focus on sensory (i.e., visual) details during trauma encoding, would give rise to the preferential processing (due to enhanced activation in

- 20 -

PRS) of those environmental stimuli that bear a strong perceptual resemblance to the actual traumatic situation. This preferential processing of perceptual trauma-reminders would then increase the probability that the trauma memory would intrude in conscious awareness due to direct retrieval (Conway & Pleydell-Pearce, 2000) of fragments of the sensory-perceptual representation. Taken together, the foregoing discussion indicates that sensory reliving of a traumatic situation triggered involuntary by cues reminiscent of those present during the experience is mediated by a different (i.e., more perceptual) memory system (i.e., SAM; Brewin et al., 1996) than the retrieval of normal verbally accessible memories (i.e., VAM; Brewin et al., 1996). In terms of the model depicted in figure 1, well-developed executive control (i.e., resistance to PI) existing prior to the exposure of a traumatic event allows gradual control over the processing/attentional bias for perceptual trauma reminders (i.e., retrieval cues) in the external environment (i.e., Ehlers & Clark, 2000). This gradual control may counter the direct retrieval of stressful intrusive memories in SAM and support the formation of more healthy associations with the trauma reminders in VAM (i.e., Brewin et al., 1996; see Brewin 2001 for a slightly different account of the effects of exposure therapy). In contrast, inefficient executive control may set people at risk for a prolonged engagement with perceptual reminders in the post-trauma environment. This inability to disengage attention from the stressful aspects of the trauma reminders (e.g., “danger”) may maintain imagery (SAM) reliving of the trauma and prevent the formation of more healthy VAM associations to the trauma cue.

- 21 -

Chapter 1

Figure 1. Model explaining the gradual decrease of intrusive memories

Pre trauma

Post trauma

• Sight of a car Trauma reminder

Executive control

• Feeling of… • Smell of… • Sound of..

Resistance to PI

+

VAM SAM

Note. SAM = situational accesible memory; VAM = verbally accesible memory; thick black arrow = strongly associated; thin dashed arrow = weakly associated; dark grey oval (SAM) = highly activated; = inhibitory connection; + = supports the formation of new associations.

-

1.4. Aim and outline of the present dissertation The central aim of the present dissertation is to explore the role of pre-trauma executive control on the persistence of image-based intrusive memories after trauma. Based on the foregoing discussion of the experimental memory literature, it is proposed that this relationship might be restricted to the specific executive ability of resistance to proactive interference. A second aim of the present thesis is to explore the mechanisms by which executive control may counter the retrieval process of intrusive memories after trauma (see the model presented in figure 1). The idea is that individual differences in the (in)ability to disengage attention from perceptual reminders after trauma predicts the frequency of

- 22 -

intrusive re-experiencing. This ability to disengage attention is thought to depend on the amount of executive resources existing prior to experiencing a traumatic event. Based on this model, the empirical chapters of the present thesis focus on three target questions: 1. Is there a prospective relationship between pre-“trauma” executive control and the frequency of subsequent “trauma”-related intrusive memories? 2. If so, is this relationship based on a relative (in)ability to resist PI from working memory? 3. Is there a relationship between the ability to disengage attention of visual traumareminders and the frequency of subsequent image-based intrusive memories?

The first two chapters explore the relationship between executive control and intrusive memories related to a stressful life event in a retrospective design. As our main hypothesis of a specific relationship between resistance to PI and intrusive memories has not been tested before, we used these cross-sectional examinations as stepping stone for testing the prospective model depicted in figure 1. The main goal of the study described in chapter 2 was to test the relationship between self-reported distractibility (as an index of executive control) and the frequency of recent intrusive memories in a large sample of undergraduates (N= 400). In chapter 3, we tested the idea that persistent intrusive memories can be best explained by inefficient executive control at the cognitive level. Non-selected participants filled in a standard index of intrusive memories (i.e., IES; Horowitz et al., 1979) and completed objective measures of executive control at the cognitive level (termed “resistance to proactive interference”; i.e., paired associates list-learning) and objective measures of executive control at the response level (termed “response inhibition”; i.e., Stroop, Random Number Generation). It was hypothesized that intrusive memories would be primarily related to individual differences in the ability to resist PI. Yet, no relationship was expected with indices of response inhibition. Based on the ideas presented by inhibition theory, it was also examined whether individual differences in the ability to reduce the accessibility/availability of irrelevant material (i.e., an effect of well-developed ability to resist PI) would be related to the frequency of intrusive remembering.

- 23 -

Chapter 1

The last three chapters of this dissertation use a prospective design with a trauma film as an analogue stressor for eliciting intrusive memories. Before discussing details of the studies presented in these chapters, I will present an overview of the ins and outs of the trauma film paradigm (e.g., Holmes & Bourne, 2008).

The use of trauma films in analogue settings An important conclusion from early studies using analogue stressors to induce intrusions was that the tendency to experience intrusive memories was common in the population at large and expected to occur following mild as well as severe stress events (see Holmes & Bourne, 2008 for a historical review).This earlier work has given input to more recent developments of the trauma-film paradigm (e.g., Holmes et al., 2004; Holmes & Bourne, 2008) in which participants are presented with a short film (8-12 min) depicting traumatic events (e.g., scenes of a traffic accident; Holmes et al., 2004; Holmes & Steel, 2004; Stuart, Holmes & Brewin, 2006 ) or horrific movie scenes (e.g., Kindt & van den Hout, 2003). During the film, participants could either be asked to view the film as they would naturally do or to adopt a specific processing task or instruction manipulation. After film viewing, participants complete some post-film control indices (e.g., film-related distress) and are subsequently asked to record any spontaneously occurring intrusive memories of the trauma film in a oneweek diary. After one week, participants return to the lab to submit their diaries and to complete some follow up measures. As noted earlier, the trauma-film paradigm supports experimental control over trauma which has the opportunity to study the role of potential pre morbid risk factors such as executive control on persistent intrusive memories in an analogue setting. The remaining chapters were designed as follows. The prospective studies in chapter 4 and 5 addressed the first two target questions more specifically. These questions were based on the model’s underlying assumption of a link between pre-trauma resistance to PI and SAM-based intrusions. In chapter 4, we explored the relationship between pre-film self-report executive/attentional control and the frequency of intrusive memories in the four days after the presentation of a stressful film. Furthermore, we also tested whether the diary method would be preferable above retrospective ratings of intrusive memories via a questionnaire. The main goal of chapter 5 was to examine the

- 24 -

relationship between a performance-based measure of resistance to PI (as a specific executive ability) and image-based intrusive memories related to an earlier presented stressful film. These memories were recorded in a one-week diary and via a questionnaire (i.e., a modified IES) at a follow-up session after 7 days. Finally, chapter 6 was devoted to the second theme and third question in this dissertation: control over the involuntary retrieval of intrusive memories after a stressful event. In terms of the model: we examined the association between the perceptual cue and SAM-based intrusive memories. It was tested whether an inability to disengage attention from visual reminders of an earlier presented stressful film would predict subsequent intrusive memories recorded in a one-week diary. Biased processing of visual film reminders was explored in a single target rapid serial visual presentation paradigm (RSVP; Most et al., 2005). The degree of interference by these film-reminders during the detection of neutral targets was used as an index of the (in)ability to disengage attention. To further support the role of attentional disengagement in the prediction of image-based intrusive memories, we also included a self-report measure of attentional control that was presented before the presentation of the stressful film. This made it possible to undertake a preliminary investigation of the model’s link between pre-trauma executive control and biased processing of perceptual trauma reminders.

- 25 -

Chapter 1

- 26 -

Chapter 2 Distractibility and Individual Differences in the Experience of Involuntary Memories J. Verwoerd, I. Wessel. Personality and Individual Differences (2007: 42, 325-334) The present study explored the idea that the tendency to experience intrusive memories might be associated with relatively weak cognitive control in general as indexed by the general propensity to become distracted by irrelevant information. A sample of undergraduate students (N = 413) filled in self-report measures of involuntary memories, distractibility, depression and repressive coping. The results showed a significant relation between involuntary memories and distractibility, independent of both trait depression and repressive coping, indicating a general vulnerability factor. As cognitive control may be sensitive to circadian variation, time of day effects in the experience of intrusions were also explored. No significant relation emerged. A proposed relationship between deficient inhibitory mechanisms of working memory and vulnerability for developing and maintaining intrusive memories after experiencing a stressful event is also discussed.

- 27 -

Chapter 2

1. Introduction Intrusive memories are involuntary recollections of specific events in the past that are hard to control (American Psychiatric Association, 1994). These memories are the hallmark symptom of posttraumatic stress disorder (PTSD), but they are also found together with other pathologies like depression (Brewin, 1998; Kuyken & Brewin, 1994). Apart from memories of trauma, involuntary memories of past events occur also in the normal population (e.g., Berntsen & Hall, 2004; Brewin, Christodoulides & Hutchinson, 1996). Intrusive traumatic memories and involuntary memories in the normal population have in common that they are primarily visual and are mainly triggered by specific cues in the external physical environment, but also by internal cues (present only in thoughts). Although involuntary memories are a normal everyday phenomenon, it appears that people differ in the frequency of experiencing these memories (Horowitz, Wilner & Alvarez, 1979). There is some evidence that differences in working memory capacity may help to explain inter-individual differences in experiencing involuntary memories. For example, Klein and Boals (2001) found that the frequency of experiencing involuntary memories and attempts to avoid such memories are relatively high in individuals who are relatively low in working memory capacity (WMC). WMC is the ability to keep information quickly retrievable and usable under conditions in which there is interference from information that is strongly elicited by task context but that nevertheless would lead to inappropriate responding on a current task (Engle, 2002). The assumption is that it depends on the ability to keep irrelevant information out of working memory (Rosen & Engle, 1998). In line with this, it has been shown (Brewin & Beaton, 2002; Brewin & Smart, 2005) that individuals who are relatively high in working memory capacity are better able to suppress unwanted thoughts. Perhaps, a general inability to keep distracting information out of awareness could be seen as a personality trait which is also responsible for inter-individual differences in involuntary memories. The main purpose of the present study is to investigate this possibility using self-report measures of involuntary memories and cognitive failures. The degree to which people report cognitive failures has been assumed to be an index of the inability to keep irrelevant information out of awareness (Friedman & Miyake, 2004). Earlier research showed that depression is associated with both increased distractibility and the occurrence of involuntary

- 28 -

intrusions (Ellis, 1990; 1991). Therefore, in this study we also measured trait depression (Spielberger, 1995) to see if any relation between distractibility and involuntary memories would not be an artifact of variations in the tendency to react with high levels of negative feelings in a wide variety of contexts. In addition, the present study explored a number of variables that may shed light on the nature of the association between weak cognitive control and the occurrence of involuntary memories. The first variable was repressive coping (Weinberger, 1990). People with a repressive coping style appear to be particularly adept at using protective strategies to avoid self-threatening material from entering awareness (Myers & Derakshan, 2004). Thus, it is possible that having a repressive coping style lowers individuals’ distractibility, thereby moderating the relation between distractibility and involuntary memories. Second, the role of valence was explored. Since distractibility was hypothesized to be a general and stable vulnerability trait, it was expected that valence of reported memories would not play a moderating role in the relationship between distractibility and involuntary memories. Third, earlier research (Hasher, Zacks & May, 1999; Yoon, May, & Hasher, 2000) found a relation between times of day of preferring habitual and intellectual activities and peak moments of circadian variation. Following this, Meccaci, Righi & Rocchetti (2004) found that self-reported cognitive failures were most noticeable at off-peak times of circadian rhythms. In this study, we explored whether reports of involuntary memories followed a similar pattern. In sum, in this study cognitive failures were predicted to be associated with involuntary memories. This relation was expected to be independent of trait depression, coping style and valence. It was also predicted that off-peak times of circadian rhythms would go together with reporting the highest frequency of intrusions. Finally, it was assumed that a general inability to keep distracting information out of awareness would be responsible for interindividual differences in involuntary memories

2. Method 2.1. Participants A sample of 488 undergraduates completed several questionnaires as part of a larger test battery at the beginning of their first year. Because of a strict time schedule, not all of the participants completed all the questionnaires. Therefore, in this study only participants who

- 29 -

Chapter 2

completed the five most important questionnaires (CFQ, IES, STDS_State/Trait, IMQ, and MEQ) were used for further analyses, leaving a total of 413 valid cases. Additionally, the subsidiary analysis with repressive coping was done over the 358 participants who completed both the anxiety and defensiveness measures used for constructing this scale. Mean age of the participants (95 men, 285 women and 33 with unreported gender) was 20.4 years (range: 1851).

2.2. Materials The Cognitive Failures Questionnaire (CFQ; Broadbent, Cooper, FitzGerald & Parkes, 1982) is a 25 item questionnaire, related to everyday slips and errors in the different cognitive domains of perception, memory and motor functions, experienced over the last six months. Total scores on the CFQ range from 0 to 100 and the response format uses a 5-point Likert scale (0 = never, 4 = always). The CFQ had good internal consistency (Cronbach’s α = .86). The Impact of Events Scale (IES; Horowitz et al, 1979) is a 15-item self-report measure of involuntary memories of some specific emotionally evoking event (range: 0-75). Items are scored on a four-point scale (0 = not at all, 5 = often) with item-loadings of 0, 1, 3, 5. The internal consistency of the IES was high (Cronbach’s α = .93). The Involuntary Memory Questionnaire (IMQ) was constructed for the present study and measures characteristics of the most prominent involuntary memory in the past week. Following Brewin, Christodoulides & Hutchinson (1996), involuntary memories were defined as: “suddenly appearing memories about specific events where you were personally involved. These memories can be positive or negative, they are difficult to control and they can hinder you by your daily tasks.” The questionnaire consisted of the following items: First, participants were asked to describe briefly the content of their most prominent involuntary memory in the past week. Next, they rated the frequency of their memory, together with how positive, how negative, and how intense they experienced their memory on 100 mm Visual Analog Scales (VAS) (range: 0 = not at all, 100 = extreme). Finally participants were asked during which period(s) of the day the intrusions were most noticeable to them. On a timescale with 30-minute intervals (range: 0h – 24h) they reported the length (e.g. 5:00 – 8:00 a.m.) of every single period they had experienced involuntary memories. They were also asked to describe their activities in that period(s).

- 30 -

The Morningness-Eveningness Questionnaire (MEQ; Horne & Östberg, 1976) consists of 19 items related to time of the day preferences for practicing physical and mental activities and subjective alertness after rising and before going to bed. People can get scores in the range from extreme-eveningness (16) to extreme-morningness (86). The MEQ has good internal consistency (Cronbach’s α = .83). In order to assess repressive coping, scores on the Trait scale of the State-Trait Anxiety Inventory (STAI; Spielberger & Vagg, 1984) and the Social Desirability Scale (SDS; Crowne & Marlowe, 1960) were combined. The STAI consists of 20 items scored on 4-point scales, range 20-80 as a measure of anxiety. The SDS consists of 33 dichotomous items (1 = true, 0 = not true) as a measure of defensiveness. Following Mendiola (2002), a repressor score was calculated by subtracting the SDS from the STAI. Next, the difference score was again subtracted from the highest conceivable positive score. This resulted in an ascending repressor score (range: 0-36). The State-Trait Depression Scales (STDS; Spielberger, 1995) were used as indicators of State (STDS-S) and Trait (STDS-T) depression in the normal population. Each subscale consists of 10 items scored on 4-point scales, range: 10-40. The state items are used to measure participants’ reports on how they feel “at the moment” by rating the intensity of their feelings. The trait items measured how participants “generally” feel (Spielberger, 1995).

2.3. Statistical analyses The data were analyzed using Pearson’s product-moment correlations and hierarchical multiple regression (HMR) techniques. The first purpose of the regression analyses was to investigate whether an association between cognitive failures and involuntary memories was moderated by trait and state depression or by repressive coping. In these analyses the IES was used as the dependent variable. The second purpose was to investigate whether characteristics such as valence would moderate the association between cognitive failures and the frequency of the most prominent intrusive memory in the past week. These analyses used the IMQ frequency as the dependent variable. A measure of valence was calculated by subtracting the two valence ratings of the IMQ (positive minus negative). Following this, 100 points were added to the difference score, resulting in a valence score (range: 0 = extremely negative memory - 200 = extremely positive memory). Prior to conducting the analyses, interaction terms were computed as follows. First, for the IES-analyses the repressor score, the CFQ total

- 31 -

Chapter 2

score and the STDS subscale scores were transformed into z-scores. Next, the transformed repressor score and STDS subscale scores were multiplied with the CFQ total score, resulting in three different interaction terms. Second, for the IMQ-analyses a similar multiplication was conducted with the CFQ total scores and both valence and time since event. The regression analyses were conducted in a similar fashion. The first step included the main predictors, whereas the second step entered the interaction terms. The first two analyses used the IES total scores as the dependent variable. The first analysis looked at the possible moderating effects of state and trait depression. Therefore, the two scales of the STDS entered the model in the first step together with the CFQ total scores. The interaction terms between the CFQ and the STDS subscales entered the model in the second step. Second, a similar analysis with the IES looked at repressive coping as a moderating variable. The repressor score and the CFQ total scores were included in the first step and the interaction term between the repressor score and the CFQ total score entered the model in the second step. As for the analysis with the IMQ frequency scores as the dependent variable, the analysis with valence was conducted in a similar way.

3. Results

3.1. Descriptive statistics and contents of reported involuntary memories Mean, standard deviation and range of the scales and questionnaires used in this study are shown in Table 1. Table 2 shows that from a total of 413 participants who filled in the IMQ, only 200 gave a description of their most prominent involuntary memory in the past week. It appeared that valence ratings of those participants differed slightly from participants who did not describe a specific involuntary memory, t(412) = 2.02, p< .05. That is, the first group reported slightly more positive involuntary memories. Frequency of reported memories was also higher for the memory group in comparison with the group who reported no memory, t(412) = 3.56, p < .01. Intensity ratings did not differ between the two groups, t(412) = -0.64, p = .52. The contents of the reported memories were compared with items used to measure life stress on the Life Experience Scale (LES; Johnson, Sarason, & Siegel, 1979). It appeared that 20 % of the memories resembled LES categories of negative events (e.g., death of a close family member, serious illness of a close family member, death of a spouse, divorce of parent, broken relationship and trouble with study/ employer) and 11 % to

- 32 -

LES positive events (e.g. reconciliation with boyfriend/girlfriend, major change in social activities, major change in closeness of family members). However, 69 % of the reported memories could not be classified with the LES (e.g., trouble with a car, dating, going on holiday). Pearson product-moment correlations between all measures are given in Table 3. Because of the large number of individual correlations, a Bonferroni correction was applied, resulting in an alpha of .001. As predicted, the results showed a significant correlation between the CFQ and IES, r(412) = .36, p < .001. The correlation between the two measures of involuntary memories (IES and IMQ-freq) appeared to be rather low, r (412) = .23., p< .001. Table 1. Mean, Standard Deviation (SD) and Range of the scales used in this study N

Mean

SD

Range

MEQ

413

47.23

8.77

22-77

CFQ

413

41.61

11.75

10-77

IES

413

20.42

17.16

0-66

STDS-S

413

22.09

4.01

10-30

STDS-T

413

24.23

3.91

10-32

IMQ frequency

413

50.00

25.36

0-100

IMQ Valence

413

103.63

66.03

0-200

Repressor score

358

20.27

6.48

0-36

Note. MEQ = Morningness-Eveningness Questionnaire; CFQ = Cognitive Failures Questionnaire; IES = Impact of Events Scale; STDS-S = State-Trait Depression Scales: State; STDS-T = State-Trait Depression Scales: Trait; IMQ frequency = frequency ratings Intrusive Memory Questionnaire; IMQ Valence = valence ratings Intrusive Memory Questionnaire.

- 33 -

Chapter 2

Table 2. Number of participants who described and did not describe an involuntary memory plus means and standard deviations (SD) of Valence, Intensity and Frequency of those memories (IMQ) Memory description Number (%) Valence Intensity Frequency (SD) (SD) (SD) Yes 200 108.71* 60.00 54.49* (48.5) (67.45) (23.21) (24.63) No 213 95.32* 58.53 45.67* (51.5) (63.71) (22.78) (25.61) * p < 0.05 3.2. The effect of depression on the relation between the CFQ and IES The results of the HMR analysis with the IES as dependent variable and STDS-T , STDS-S and the CFQ total scores as predictors in the first step accounted for a significant 19% of variance in reported involuntary memories, F(3, 409) = 32.31, p < .001. Independent of each other, both the Trait depression scale and CFQ scores were positively related with the IES: CFQ, β = 0.29, t(412) = 6.31, p < .001; STDS-T, β = 0.18, t(412) = 2.76, p < .01. State depression appeared to be only marginally related to the IES, STDS-S, β = 0.10, t(412) = 2.20, p = .06. The two interaction terms between cognitive failures and the depression scales entered the model in the second step and added no extra variance, ∆R² = .008, F-change (2, 407) = 2.12, p = .28.

3.3. The influence of repressive coping on involuntary remembering (IES) Including the repressor scores and CFQ total scores in the first step of the regression model resulted in a significant amount of explained variance, R² = .19, F (2, 355) = 42.68, p < .001. There was a negative association between repressive coping and involuntary remembering, β= -.21, t(357) = -4.25, p < .001. The CFQ total scores were positively related to involuntary remembering, β = .35, t(357) = 7.11, p < .001. Finally, the interaction between the CFQ total score and the repressor score did not reach significance, ∆R² = .00, F-change (1,354) = 0.19, p = .89. Repressive coping did not moderate the relationship between distractibility and involuntary memories.

- 34 -

3.4. Involuntary memories of one particular event The HMR analysis with IMQ-freq as the dependent variable and valence and CFQ total scores as predictors in the first step resulted in 4% of explained variance in the prediction of IMQ-freq, F (1, 410) = 8.75, p < .001. There was a positive relationship between CFQ total scores and IMQ-freq, β = 0.17, t(412) = 3.30, p < .01, and a negative relation between valence and IMQ-freq, β = -0.13, t(412) = -2.50, p < .01. Including the interaction terms between valence and cognitive failures in the second step did not add any extra explained variance in IMQ-freq, ∆R² = .00, F-change (1,409) = .04, p = .85. Table 3 P-M correlations between the scales and questionnaires used in this study Scales 1 2 3 4 5 6 7 _ 1. CFQ .36* .21* .28* -.20* .15 0.15 2. IES _ .26* .30* -.28* .10 0.23* 3. STDS-State _ .57* -.35* .14 0.03 4. STDS-Trait _ -.49* .19 0.06 5. Repressive _ -.17 -.09 score 6. Valence _ -.10 7. IMQ-frequency _ Note. CFQ = Cognitive Failures Questionnaire; IES = Impact of Events Scale; STDS-State = State-Trait Depression Scales: State; STDS-Trait = State-Trait Depression Scales: Trait; Valence = Valence ratings Intrusive Memory Questionnaire; IMQ-frequency = frequency of memory on Intrusive Memory Questionnaire. *p<0.001.

3.5. The relation between period(s) of experiencing intrusions (IMQ) and circadian typology (MEQ). In order to explore time of day effects, we looked at the connection between MEQ scores and the times at which participants reported to have experienced the highest frequency of intrusions. Of the 413 participants, only 134 gave an indication of the period(s) of the day at which the occurrence of intrusions was most noticeable to them. These 134 participants were subsequently divided into two groups. The first group consisted of those participants who reported to have experienced the highest frequency of intrusions especially in the morning (between 5:00 a.m. and 12:00 p.m., N= 18) and the second group consisted of participants who reported to have experienced the highest frequency of intrusions especially - 35 -

Chapter 2

in the evening (between 18:00 p.m. and 24:00 a.m., N=82), leaving 100 participants for further analyses. The rationale for constructing two groups was that the evening intrusion group was expected to show a tendency to be classified as “morning-type” and the morning intrusion group was expected to show a tendency to be classified as “evening-type”. The participants who reported the highest frequency of intrusions especially in the afternoon were not expected to show any relation with MEQ total scores and were not used for further analysis. MEQ scores of both groups were compared using an independent t-test. It appeared that the MEQ-scores in the morning-intrusion group (M = 48.95, SD = 8.67) did not differ significantly from the evening-intrusion group (M = 47.95, SD = 9.17) in reported MEQ scores, t(99) = 0.42. p = .67. Thus, the time of the day that people reported to have experienced the highest frequency of intrusions was not related to time of the day preferences for performing habitual activities.

4. Discussion The main results of the present study showed that reports of involuntary memories were significantly related to cognitive failures reported in the different cognitive domains of perception, memory and motor functions and that this relationship was independent of trait depression and repressive coping. The results of the self-report measures were in line with those of earlier studies (Brewin & Smart, 2005; Klein & Boals, 2001) showing that measures of WMC were related with individual differences in reported involuntary thoughts and memories. Two possible accounts may be given as an explanation for these findings. The first account states that the ability to focus attention would prevent irrelevant information from entering awareness (Engle, 2002). That is, the ability to keep task-related material in an active state would prevent involuntary memories from entering awareness. The second account is concerned with the functioning of inhibitory mechanisms. Friedman and Miyake (2004) found that both individual differences in self-reported cognitive failures as well as performance on WMC tasks were related to the functioning of inhibitory mechanisms. That is, people who report many cognitive failures and who score low on measures of WMC may be especially weak at suppressing task-irrelevant information Other evidence for an inhibition account of cognitive control comes from studies associating performance on experimental cognitive tasks (Hasher et al., 1999) or self-reported cognitive failures (Meccaci et al, 2004)

- 36 -

with circadian variation. Off-peak moments of circadian rhythms have been related to deficient inhibitory functioning (Hasher et al., 1999; Yoon et al.. 2000). In our study, no relation was found between the period(s) of the day of reporting most intrusions and circadian variation. This could mean that inhibitory mechanisms play no role in the prevention of involuntary memories, or that the questionnaires used in the present study were not sensitive enough to find a link between intrusions and inhibition. This brings us to some methodological issues. First, considering the finding of no relationship between time of day of reported involuntary memories and off-peak times of circadian rhythms, it appeared that only a minority of the sample was able to report about their intrusions. Thus, our non-significant results could be due to power problems, but it is also possible that retrospectively reporting about period(s) of most intrusions is not a valid way to investigate lack of inhibitory control at off-peak times. It is unclear, for instance, whether the reason for reporting intrusive memories especially in the evening (n = 82) was related to inhibitory failures at off-peak times or simply due to the fact that participants (morning and evening types) were better able to remember intrusions at moments when they were not distracted by daily activities. Second, the results showed that the correlation between the IES and the IMQ (both were intended to measure involuntary memories) was unexpectedly low. Since they shared only a very little amount of variance, it is possible that the questionnaires triggered different kinds of memories. A closer look at the contents showed that on the IES, participants responded to statements about an event which had high impact on their lives, so it is possible that this questionnaire would direct them to use strategies for retrieving more specific memories (events located in time and place [see Williams & Broadbent, 1986]). In contrast, the IMQ asked participants about one specific involuntary memory which could be positive or negative. The results showed that about 70% of the reported memories on the IMQ were not related to events reported on the Life Experience Scale (LES; Johnson et al, 1979), and were much more general in nature (e.g., holiday, dating). Third, the present study used a cross-sectional design, so no predictions about the causal relationship between weak cognitive control and involuntary memories could be made. It would be interesting to further investigate the direction of the relationship between cognitive control and involuntary memories using experimentally induced intrusions and tasks specifically designed to measure cognitive inhibition.

- 37 -

Chapter 2

One possible interpretation of the results may be that a relatively weak control in general (e.g., distractibility) is involved in developing involuntary memories after having experienced a stressful event. Other studies have supported this idea. For example, Broadbent et al. (1982) argue that the CFQ measures vulnerability to externally imposed stress, and a twin study of Boomsma (1998) showed that around 50 percent of the variance in CFQ scores may be attributed to genetic differences between individuals. Additionally, a general vulnerability trait would predict no difference in the experience of positive or negative memories. In line with this, the results of the current study showed no effect of valence on the relation between distractibility and involuntary memories. Perhaps then, pre morbid individual differences in distractibility play a role in development of psychopathology after trauma. In the first months after experiencing a trauma, some people are able to get control over the overwhelming intrusions and will not develop PTSD, while other people will develop psychopathology and will stay highly vulnerable for cues triggering involuntary memories (Rothbaum, Foa, Riggs, Murdock & Walsh, 1992). In line with this, Ginzburg, Solomon and Bleich (2002) showed that people with a repressive coping style are less vulnerable to develop PTSD after having experienced a Myocardial Infarction (MI). Interestingly, the results of the present study showed that repressive coping and distractibility were independently related to the frequency of reported involuntary memories. Probably, people with a repressive coping style have developed protective mechanisms against experiencing personally negative material like involuntary memories. Taken together, the present findings are in accordance with the idea that high levels of distractibility are a vulnerability factor for maintaining involuntary memories. In addition, the results suggest that a repressive coping style protects people from experiencing involuntary memories, albeit not via lowering individuals’ distractibility. Future experimental research is necessary to investigate the (causal) role of inhibitory mechanism in the development and maintenance of involuntary memories.

- 38 -

Chapter 3 Individual Differences in Experiencing Intrusive Memories: The Role of the Ability to Resist Proactive Interference J. Verwoerd, I. Wessel, P.J. de Jong Journal of Behavior Therapy and Experimental Psychiatry (in press)

This study explored whether a relatively poor ability to resist or inhibit interference from irrelevant information in working memory is associated with experiencing undesirable intrusive memories. Non-selected participants (N=91) completed a self-report measure of intrusive memories, and carried out experimental tasks intended to measure two different types of inhibition: resistance to proactive interference and response inhibition (i.e., the ability to prevent automatically triggered responses). The results showed a significant relationship between inhibition at the cognitive level (i.e., resistance to proactive interference) and the frequency of intrusive memories (especially in the group of female participants) whereas no such relationship with measures of response inhibition emerged. These findings are consistent with the idea that deficient inhibitory control reflects a vulnerability factor for experiencing intrusive memories. Implications for research investigating risk factors for the development of posttraumatic stress disorder (PTSD) are discussed.

- 39 -

Chapter 3

1. Introduction There is evidence that most people will be exposed to at least one extremely threatening situation during the course of their lives (Ozer, Best, Lipsey, & Weiss, 2003). In a study among 900 US college students, the prevalence rate of having experienced a potentially traumatic event was found to be around 67 % (Bernat, Ronfeldt, Calhoun, & Arias, 1998). There is, however, a great variety in how people deal with these highly aversive experiences. In the immediate aftermath of a traumatic event, most people report elevated levels of psychopathology, but only a minority of them will eventually develop chronic PTSD (McNally, Bryant & Ehlers, 2003). The core symptoms of PTSD consist of persistent avoidance of stimuli associated with the trauma, increased arousal levels (e.g., hyper vigilance, exaggerated startle response) and intrusive re-experiencing, including recurring images, flashbacks, nightmares and distress when confronted with reminders of the traumatic event (American Psychiatric Association, 2000). Considering this, one of the intriguing puzzles surrounding research on PTSD is identifying the underlying cognitive mechanisms that could set people at risk for the maintenance of trauma-related symptoms. Knowledge of the underlying mechanisms could help explaining why and how wellknown pre-trauma risk factors (see Brewin, Andrews & Valentine, 2000; Ozer et al., 2003 for reviews) are involved in pathological responses such as intrusive memories. Cognitive theories of PTSD clearly describe the development of intrusive symptoms due to problems with encoding (e.g., perceptual processing [Ehlers & Clark, 2000]), storage (Situational Accessible Memories [SAM, Brewin, Dalgleish & Joseph, 1996]) and retrieval (Cue-driven retrieval due to strong associative learning [Ehlers & Clark, 2000]) of the traumatic event. However, the question of how individual differences in posttraumatic psychopathology may be explained by pre-trauma information-processing properties that act as vulnerability factors remains largely unanswered.. For example, there is convincing evidence that relatively low pre-morbid intelligence is an important predictor of chronic PTSD symptoms (e.g., Kaplan et al., 2002; Macklin et al., 1998; McNally, 2003), but little is known about the underlying cognitive mechanisms relating intelligence to pathological responses to trauma. Interestingly, recent developments in the experimental memory literature suggest that specific cognitive deficits existing prior to the occurrence of a traumatic event may hamper natural recovery of intrusive re-experiencing that is part of the common posttraumatic - 40 -

response. For example, there is some evidence suggesting that individual differences in working memory capacity (WMC) might play a role in the maintenance of intrusive memories. It has been proposed that WMC reflects a domain-general capability to control attention, which is particularly important in situations involving proactive interference or conflict between competing response tendencies (Engle, 2002). In two studies, Brewin and colleagues (Brewin & Beaton, 2002; Brewin & Smart, 2005) showed a positive relationship between WMC and the ability to block neutral or personally relevant thoughts from entering working memory. It has been suggested that individual differences on indices of WMC (e.g., Operation Span; Turner & Engle, 1989) reflect the capability to actively inhibit interference from events stored in long term memory (Kane & Engle, 2000; Lustig, Hasher, & May, 2001; Rosen & Engle, 1998). In this light, the question arises whether deficient inhibitory control may function as a specific vulnerability factor for the persistence of intrusive memories in the aftermath of a traumatic event. Preliminary evidence for this idea comes from a study of Klein and Boals (2001). In this study employing healthy undergraduate students, the frequency of experiencing intrusive and avoidance symptoms related to a potentially traumatic event was relatively high in people with relatively low WMC (Klein & Boals, 2001). In addition, an earlier study found that people who described themselves as highly distractible (which might be taken as an analogue to performance on WMC tasks) also reported a relatively high frequency of intrusive memories (Verwoerd & Wessel, 2007). The preceding studies have provided indirect evidence for the idea that relatively weak inhibitory control over the contents of working memory might be responsible for individual differences in the experience of unwanted intrusive memories. To further explore this issue, the present retrospective study focused on the relationship between deficient inhibitory control and the frequency of experiencing intrusive memories in an unselected student sample. It has been suggested that inhibition serves different functions (Hasher, Zacks, & May, 1999) and may even consist of two or more independent mechanisms (Friedman & Miyake, 2004). For example, one type of inhibition involves controlling automatically triggered prepotent responses in order to carry out an effortful primary and compatible response. A lack of such response inhibition seems to be involved in dysfunctional impulsive behavior (Nigg, 2000). Alternatively, inhibition may act on a more cognitive level and involve the ability to resist interference from information that was previously relevant to the - 41 -

Chapter 3

task at hand, but has since become irrelevant because of a change in context (Friedman & Miyake, 2004; Hasher et al., 1998). Intrusive memories may be seen as a profound example of experiencing unwanted interference in real life. Therefore, problems with this latter form of inhibition (i.e., resistance to Proactive Interference, PI) may be particularly relevant for explaining posttraumatic intrusive cognition. Interestingly, Friedman and Miyake (2004) found a relationship between a latent variable of resistance to PI and self-reports of general intrusive thoughts. However, it remains to be seen whether a similar pattern of associations will hold for the actual experience of intrusive memories related to a stressful event. The major aim of the present study was to examine whether there is indeed a general relationship between weakened inhibitory control at the cognitive level (i.e., resistance to PI) and the frequency of intrusive memories related to an earlier experienced stressful event. It was expected that a relatively weak ability to resist PI in working memory would be associated with a relatively high frequency of intrusive memories. In contrast, the ability to inhibit automatically triggered responses was not expected to show any relationship with individual differences in intrusive re-experiencing. A subsidiary aim of the current study was to explore the consequences of the process of cognitive inhibition. If a well-developed ability to resist PI prevents unwanted/irrelevant material to intrude into conscious awareness, does this mean that the resisted material has become less accessible for remembering at a later point in time? It has been suggested that keeping earlier studied but now irrelevant material out of working memory results in a reduced activation of the representation of that material in long term memory (Anderson & Spellman, 1995; Rosen & Engle, 1998). Therefore, the present study looked at the ability to resist PI and its supposed consequence, using an AB-AC -AB list-learning paradigm. This paradigm renders two indices that are relevant to the present purpose: a) the number of trials needed in order to learn new responses (AC) in the face of interference from earlier studied (old) material (AB) and b) latencies to respond again with the old (AB) material after having learned these new responses (see Rosen & Engle, 1998). It was expected that showing more PI as reflected by a higher number of trials to learn new responses would predict relatively high scores on a selfreport measure of intrusive memories. By contrast, if indeed the inhibition of unwanted responses has consequences for their later retrievability/accessibility, we expected to find a negative relationship between latencies and intrusive memories. That is, a relatively weak ability to resist PI would be reflected by faster reaction times (higher accessibility) for old

- 42 -

(AB) responses. Thus, we predicted that shorter response latencies would be related to a higher frequency of intrusive memories. In addition, because the clinical literature consistently indicates that female gender increases the probability of developing PTSD following a traumatic experience (Ozer et al., 2003), we explored whether the hypothesized effect of low resistance to PI on the frequency of trauma-related intrusions might be especially pronounced in women. Finally, given the cross-sectional nature of the present study, it is important to rule out alternative explanations for the hypothesized relationship between resistance to PI and intrusive memories. For example, earlier research has shown that both deficient cognitive/inhibitory control and a high frequency of intrusive memories are related to elevated levels of depression (Ellis, 1990; 1991).Therefore, the present study investigated if the proposed relationship between relatively weak inhibitory control and intrusive memories might covary with elevated levels of depression.

2. Method

2.1. Participants A total of 98 first- and second-year psychology students (77 women, 21 men) took part in this study. The age range was 18 to 45 (M = 20.8 years, SD = 4.7). The University of Groningen Behavioral Research Ethical Committee granted ethics approval for this study.

2.2. Questionnaires 2.2.1. Impact of Event Scale The Impact of Events Scale (IES; Horowitz, Wilner & Alvarez, 1979) is a 15-item self-report measure of intrusive memories which assesses intrusive and avoidance symptoms over the past week as a result of some specific emotionally evoking event which happened in the past (range 0-75). In a sample of 66 patients with stress-response syndromes the clinical mean was found to be 39.5 (SD = 17.2) (Horowitz et al., 1979). The IES has assigned itemweights of 0, 1, 3, 5. The internal consistency of the IES was found to be high (Cronbach’s α = .91). In the present study, just before they had to fill in the IES, participants briefly described the most distressing event they ever experienced and reported how long ago the event took place.

- 43 -

Chapter 3

2.2.2. Center for Epidemiologic Studies Depression Scale A Dutch translation of the Center for Epidemiologic Studies Depression Scale (CESD; Bouma, Ranchor, Sanderman & Van Sonderen, 1995) assessed depressive symptoms during the past week. The CES-D contains 20 items and item scores range from 0 (rarely or none of the time) to 3 (most or all of the time). The total score on the CES-D ranges from 0 60. Four items are worded positively and then scored reversed for the total CES-D score. The internal consistency of the CES-D in the current sample was high (Cronbach’s α = .93).

2.3. Experimental Tasks The experimental tasks were programmed in E-Prime Version 1.1 and ran on a Windows 98 computer with a 19-inch CRT color monitor. A microphone and voice key were used to measure verbal RT with millisecond accuracy. Stimuli were presented in uppercase letters (18-point New Courier font) against a black background. 2.3.1. Resistance to PI As an index of the resistance to PI inhibition type we modified the paired-associates learning procedure originally developed by Rosen and Engle (1998). Two lists of cue-target word pairs [see Appendix] were constructed in the following fashion. Twelve independent cue words were chosen from standard association norms in the Dutch language (de Groot, 1980). For each of these cues, two associates were selected as target words for the first (AB) and second list (AC), respectively. The first list (AB) consisted of strongly associated cuetarget pairs. The second list (AC) contained the same cues as the first list, but now accompanied by weakly associated target words (de Groot, 1980 [see appendix]). These lists were used in an AB-AC-AB list-learning paradigm, in which the third list consisted of a second test of the first-list word pairs. Thus, participants studied the same 12 cue words in each of the three lists, but the target words changed from list 1 (e.g., butcher-meat [AB]), to list 2 (butcher-rope [AC]), and then reverted to the original in list 3 (butcher-meat [AB]). In the present study, the procedure of learning the first two lists was based on Rosen and Engle (1998) and consisted of three phases: (1) Study. In the study phase, each pair of words appeared in random order in the center of the computer screen for 2 seconds (e.g., Butcher – Meat). Participants were not required to make any response during this phase and were instructed to study each pair for the subsequent test phase.

- 44 -

(2) Test. In the test phase participants were instructed to say out loud the target word when presented with a cue word until reaching a criterion of three correct responses given to each of the 12 cues. In this second phase, a modified dropout method was used (see Rosen & Engle, 1998), meaning that after three presentations of each of the 12 cue words, only the incorrect responses were retested. After a fixation cross was presented for 500 ms in the middle of the computer screen, a cue word appeared at the same location. The cue words stayed on the screen until the participant’s vocal response triggered the voice key (as measure of reaction time) which made the word disappear (with maximum of 32 seconds). The cuetarget pair was then printed on the screen for 2 seconds after each vocal response, irrespective of whether the participants’ response was correct or incorrect, in order to provide feedback and additional study time on each pair. Just before feedback was given, the experimenter coded the response of the participant (correct, incorrect, microphone error) via a response box. (3) Recency control. In the recency control phase, participants were presented with a final test of each target word to equate on recency of retrieval before advancing to the next list. So, each of the 12 cue words were presented only once in random order and participants were asked to respond with the appropriate target words one more time. In the present study the procedure for learning the third list differed from the study of Rosen and Engle (1998) in that the initial study phase was omitted (i.e., presenting complete word pairs for two seconds each) and participants immediately started with the test phase. A relative difficulty with the retrieval of third-list targets (i.e., increase in RT) was intended to provide an index of the ability to suppress first-list targets during second-list learning. We anticipated that an extra study phase might undermine the task’s sensitivity as a measure of individual differences in retrieval latency for these words on the third list by reducing the consequences of the suppression of the first-list target words during second-list learning. Participants were instructed that they would see the first-list cue words for a second time and that they had to come up with the accompanying first-list target words. In order to prevent confusion in the participants about which targets were to be retrieved on the third list, word pairs from different lists were presented in different colors. The first-list words (AB) were presented in yellow and the second-list words (AC) were presented in blue. The idea was that presentation of the third-list (AB) cues in yellow would act as a reminder that first-list targets should be retrieved.

- 45 -

Chapter 3

Before learning the first list, participants went through a short practice phase which was a short version of the paired-associates task in which they learned a list of 4 digit-letter pairs. The practice was meant to acquaint participants with the three phases of learning each list. 2.3.2. Response inhibition 2.3.2.1. Stroop color-word test. In this experiment, we used a computerized version of the Stroop color-word task as a measure of response inhibition (e.g., Friedman & Miyake, 2004; see also MacLeod, 2005). Stimuli were the words “red”, “blue”, “yellow”, “brown”, and “purple” printed in congruent and incongruent colors plus stimuli which consisted out of strings of asterisks (****) printed in one of the former five colors. On each trial of the Stroop task, participants saw a white fixation point on a black screen for 500 ms, followed by the stimulus (e.g., color-word or asterisks), which remained on the screen until the participants responded, after which the screen remained black for 1,000 ms. Participants had to name the color in which the different stimuli were displayed on the screen as quickly and accurately as possible without letting themselves be distracted by the content of the stimulus. Reaction times (RTs) were picked up by a voice key. Stimuli were presented eight times resulting in a total of 120 trials. These trials were divided into three different conditions: (a) 40 neutral trials which consisted of strings of asterisks matched for length of the color words. (b) 40 congruent trials with color words and ink printed in the same color (e.g., blue printed in blue) (c) 40 incongruent trials with a color word printed in a different color (e.g., blue printed in red). The order of the trials was randomized with the following restrictions: No word or color on a particular trial was related to the word or color on the immediately preceding trial and no condition appeared more than three trials in a row. Before testing, participants received 15 practice trials for voice-key calibration and for getting acquainted with the experimental procedure. Internal consistency was calculated for each of the three conditions separately and was found to be very high (Cronbach’s α between .91 and .94). 2.3.2.2. Random Number Generation task. The Random number generation task (RNG; Towse & Neil, 1998) was included as a measure of the ability to resist non-random (pre-potent) responses (e.g., counting) during random number generation. In the RNG,, participants were asked to generate random sequences of digits (numbers between 1 and 10, 1 and 10 included), for a period of 100

- 46 -

seconds, one digit each second, The rhythm was indicated by a metronome adjusted to 60 bpm. The concept of randomness was explained by asking participants to imagine repeatedly drawing a number out of a hat, reading the number aloud and replacing it before drawing another number. Participants were instructed to keep pace with tones generated by the metronome until they had produced the 100 responses. To get used to the metronome, participants were instructed to just listen and not respond to the first 8 tones. To analyze different types of randomness, we used an online program called RgCalc (Towse & Neil, 1998). This program analyzes the quality of human’s attempts at random generation and produces many different indices that have been commonly used in the analysis of “randomness”.

2.4. Procedure Participants were tested individually. After giving informed consent, they were instructed to fill in the CES-D. Following this, the experimenter gave verbal instructions about the general procedure for the PI task. Next, participants completed the practice phase of learning numbers and letters. After this, the experimenter told them that further instructions would be presented at the computer screen and that they had to remember that good performance included both speed and accuracy. The experimenter coded the participants’ responses in the adjacent room where the responses were clearly audible through an intercom system. After finishing the PI task, the experimenter reentered the room and gave instructions for the RNG task. Participants verbally generated 100 valid responses and the experimenter recorded all the numbers on an answering sheet. Following this, instructions were given for the Stroop task, after which the experimenter left the room to code participants’ responses. Next, participants thought about and were asked to describe the most distressing event they ever experienced, where after they rated recency of the event, event-related distress in the past and event-related distress in the present. Finally, with this event in mind, they filled in the IES and were shortly debriefed about the purpose of the study. The whole procedure took about 60 minutes.

- 47 -

Chapter 3

2.5. Data reduction and statistical analysis 2.5.1 Proactive interference task The method of treating outliers on the interference task was that for every participant, any vocal response time (RT) greater than 2.5 standard deviations above or below the median within each list was replaced by a value that corresponded with a value of exactly 2.5 standard deviations above or below the median. RTs below 300 ms and RTs from random microphone errors were excluded from the analyses. Additionally, for every participant, mean correct RTs were calculated for the first cycle on each of the three lists of 12 word pairs. Five types of responses were coded by the experimenter: correct response, between-list intrusions, within-list intrusions, the omission of a response, and random microphone errors. Two types of dependent variables were derived from the PI task. The main dependent variable was the number of trials needed to reach a criterion of three correct answers for each of the 12 word pairs on the interfering second list. Following Kane and Engle (2000) we chose proportional increase in RT between list 1 and list 3 as the second dependent variable measuring the consequences of resisting PI during 2-list learning. This index was preferred above the use of difference scores between the RT’s of list 3 and list 1, because it is able to capture individual differences in general response latency. Proportional increase in RT between the first- and the third list was calculated by subtracting the mean RT for the response words on the first trial of the first list from the mean RT for the response words on the first trial of the third list and dividing this by first trial list 1 mean RT. 2.5.2. Stroop task The way of treating outliers on the Stroop task was conducted in a similar way as for the interference task. For each of the three conditions (i.e., neutral, congruent and incongruent), RTs greater than 2.5 standard deviations above or below the median were replaced by a value of exactly 2.5 standard deviations above or below the median. In line with the RT measure in the PI-task, the proportional increase in RT between incongruent and neutral trials was used as dependent measure on the Stroop task. 2.5.3. RNG task Before conducting the main analyses for this task, the direction of the scores on several indices of randomness was reversed, such that higher scores would mean better performance. A Principal Component Analysis (PCA) of the present data showed a factor structure that was highly similar to the earlier findings of Towse and Neil (1998) and Miyake

- 48 -

et al., 2000. That is, a single component combined the four indices of randomness (Turning point Index (TPI), Adjacency (A), Random Number Generation (RNG), Runs) that had high loadings on the summarizing factor called Prepotent Associates in the study of Towse and Neil (1998). Both the results of Miyake et al. (2000) and Friedman and Miyake (2004) showed that factor scores from the Prepotent Associates component were significantly related to a latent variable intended to measure Response Inhibition. In line with these earlier findings, the factor scores of the component deemed “Prepotent Associates” were used as dependent variable in the current study. 2.5.4. Statistical analyses The main hypotheses in the present study were tested by using hierarchical multiple regression (HMR) analyses and Pearson product-moment correlations. Three HMR analyses with the IES as dependent variables were conducted. The first analysis was intended to measure the relationship between intrusive memories and resistance to PI. In the first step of this analysis three control variables were included which could possibly covary with the relationship between relatively weak inhibition and intrusive memories: Depressive feelings in the past week, time (in months) since the reported distressing event and gender. The time elapsed (in months) since the distressing event was included as a control, because our investigation of event-related intrusive memories in the normal population was expected to show large variation in recency of the individual intrusion-evoking stressful events. Additionally, in order to control for individual differences in basic learning rate, the number of trials needed to reach three correct responses on the first list of the interference task was also included in the first step. In the second step the number of trials to reach criterion on the second list was included as a measure of the (in)ability to resist PI of earlier learned material. Finally, in the third step, the interaction between gender and resistance to PI was included to further investigate if the hypothesized effect of low resistance to PI would be especially pronounced in the female group. The second HMR analysis tested the relationship between self-reported intrusive memories and the consequence of resisting PI. In the first step of this analysis, the control variables of depression and recency of the reported event were included. In the second step, the proportional increase in RT between the first- and the third list was included as a measure

- 49 -

Chapter 3

of the ability to reduce the accessibility of earlier interfering material. Finally, in the third step, the interaction term with gender was included. The third analysis was used to measure the relationship between intrusive memories and the ability to resist prepotent responses. The first step of this analysis consisted of the control variables, while in the second step the two measures of the hypothesized response inhibition construct (RNG, Stroop) were included. Again, in the third step, the effect of gender was further investigated by including the two interaction terms for the Stroop and RNG measures. For the sake of conciseness, only the analyses in which the measures of inhibitory control explained a significant amount of variance over and above the confounder set are reported in Table 3. To ensure that extreme scores were not unduly influencing the results, we used Mahalanobis d² values for detecting the influence of individual cases on the dependent variable or independent variables in the regression model.

3. Results Data for three participants were excluded from analyses because of undue influence on the β values in the regression model. For two of these participants, data were unreliable due to many microphone errors or knowledge of the experimental manipulation on the interference task. One participant reported clinical levels of depression and intrusive memories. Furthermore, an additional four participants were excluded because they did not report about stressful events of moderate to strong intensity (past impact < 5), leaving 91 participants for further analyses. Descriptive statistics for the scales used in the present study are shown in Table 1. As shown in Table 2, the relatively low number of trials needed to reach three correct responses on the first list of the interference task resulted in a considerable kurtosis. Therefore, for this variable, rank correlations were calculated. None of the other variables used in the present study showed significant deviations in either skewness or kurtosis.

3.1. Descriptive statistics and contents of reported intrusive memories Inspection of the mean impact ratings in Table 1 revealed that participants reported about events from their past that had a high impact on their lives at the time the event took place (i.e., M = 8.43 on a 10-point scale). The contents of the reported memories were compared

- 50 -

with items used to measure life stress on the Life Experience Scale (LES; Johnson, Sarason, & Siegel, 1979). It appeared that 76% of the memories resembled LES categories of negative events (e.g., death of a close family member, serious illness of a close family member, death of a spouse, divorce of parents, broken relationship and trouble with study/employer). Table 1 Mean, Standard Deviation (SD), Minimum and Maximum of the Scales and Experimental Variables in this Study N Mean SD Min Max IES total score 91 14.11 13.48 0 54 IES_Recency

91

58.95

51.45 0

235

CES-D

91

11.53

9.24

0

38

Past impact

91

8.43

1.25

6

10

Number of trials list 1

91

36.36

0.92

36

41

Number of trials list 2

91

39.62

3.45

36

50

Proportional increase in RT Int. task

91

49.80

23.50 11

112

RT 91

23.01

11.06 1.05

45.40

91

-0.01

1.01

3.38

Proportional

increase

in

(Stroop) Prepotent Associates (RNG)

-1.78

Note. IES = Impact of Event Scale; IES_Recency = Time in months since the event described on the IES took place; CES-D = Center for Epidemiologic Studies Depression Scale; Past impact = Distress experienced in relation to the event described on the IES; Number of trials list 1 = number of trials needed to reach 3 correct answers on the interference task (list 1); Number of trials list 2 = number of trials needed to reach 3 correct answers on the interference task (list 2); Proportional increase in RT Int. task = Proportional increase in RT between the first and the third list (interference task); Proportional increase in RT (Stroop) = Proportional Increase in RT between neutral trials and incongruent trials (Stroop); Prepotent Associates (RNG) = factor scores as an index of response inhibition on the Random Number Generation task

- 51 -

Chapter 3

Table 2 Pearson and Spearman’s rank correlations between the variables used in the present study Scales 1 2 3 4 5 6 7 8 1. IES

_

2 .IES_Recency

-.27*

3. Past Impact

.40** -.07

_

4. CES_D

.36** .15

.15

_

5. Number of trials 1

.15ª

-.01ª

-.10ª

.05ª

_

6. Number of trials 2

.32** -.10

-.04

.15

.06ª

_

7. Prop Incr.(PI_task)

.16

.07

.01

.05

-.02ª

.11

_

8. Prop Incr. (Stroop)

-.08

-.03

-.08

.06

-.03ª

-.09

-.03

_

9. Prepot.Ass.(RNG)

.07

-.16

.05

.01

.03ª

.14

.08

-.17

_

Note. N = 91. IES = Impact of Events Scale; IES_Recency = Time in months since the event described on the IES took place; CES-D = Center for Epidemiologic Studies Depression Scale; Number of trials list 1 = number of trials needed to reach 3 correct answers on the first list of the interference task; Number of trials list 2 = number of trials needed to reach 3 correct answers on the second list of the interference task; Prop. Incr. (PI_task) = Proportional Increase in RT between the first and the third list (PI_task); Prop. Incr. Stroop = Proportional Increase in RT between neutral trials and incongruent trials (Stroop); Prepot. Ass. (RNG) = factor scores as an index of response inhibition on the Random Number Generation task. ª Due to strong non-normality of the distribution, Spearman’s rank correlations were calculated for the number of trials list 1 variable. *p < .05 **p < .01.

3.1. Relationship between intrusive memories (IES) and resistance to PI As shown in Table 3, the HMR analysis with the IES as the dependent variable resulted in a significant 21% of explained variance for the four control variables, F (4, 86) = 5,301, p < .001. Significant predictors were depressive symptoms (CES-D) and recency of the distressing event described on the IES (depression, semi-partial r² = 13 %, p < .001; recency of event, semi-partial r² = 9%, p < .05). The number of trials needed to learn the firstlist responses (e.g., baseline for learning paired-associates on the interfering second list) showed no significant relationship with the frequency of intrusive memories, β = 0.06, p = .56.

- 52 -

In the first HMR analysis the number of trials needed to reach a criterion of three correct responses on the second list was included in the second step of the model. The results showed that the number of trials needed added another 5 % of explained variance to the regression model, β = 0.23, p < .05. These findings indicate that people with a reduced ability of suppressing first-list intrusions during second-list learning also reported a relatively high frequency of intrusive memories1. With the inclusion of the interaction term between number of 2-list trials and gender in the third step, a small but significant 5% of explained variance was added to the model, β = -0.22, p < .05. Further exploration of this interaction term revealed that the relationship between number of 2-list trials and intrusive memories was especially pronounced in the female group (10% of unique variance in IES-scores, β = 0.33, p < .01, n = 70). The second HMR analysis used proportional delay in RT between the first- and the third list as predictor in the second step. This step added no significant amount of variance to the model, ∆R² = .02, F (1, 87) = 2.62, p = .11. Furthermore, no additional variance was explained by including the interaction term between the inhibition index and gender in the third step of the model, F (1, 86) = 0.3, p = .58. Thus the present results provide no convincing evidence to sustain the idea that a relatively reduced accessibility of memory representations for earlier interfering material is related to a relatively low frequency of intrusive memories.

3.2. Relationship between intrusive memories and indices of response inhibition The results of the HMR analysis with the IES as dependent variable and the two measures of response inhibition (Stroop, RNG prepotent associates)2 included in the second step showed that no extra variance could be explained by the model, ∆R² = .00, F-change

1

An additional analysis with the number of first-list intrusions spoken out loud as the independent variable (e.g., the number of trials on list 2 also takes possible internal [not spoken aloud] intrusions into account) showed a somewhat reduced but still significant effect, with first-list intrusions adding another 4% of explained variance in IES scores, β = 0.18, p < .05.

2

As shown in Table 2, the two indices supposed to measure the construct of Response Inhibition (RNG, Stroop) did not show a significant association at task level. However, post-hoc analyses showed that when we used the single RNG index with the highest factor loading as single measure of response inhibition (i.e., Adjacency – the ability to prevent a counting response: loading = .93), the correlation with the Stroop measure approached significance (r (91) = .23, p < .05). This may give some indication that both the RNG and the Stroop task do indeed measure the same underlying construct of response inhibition.

- 53 -

Chapter 3

(2,85) = 0.20 , p = .95. Again, no additional variance was explained by including the two interaction terms with gender in the third step, F (1, 83) = 0.30, p = .76. Thus, individual differences on measures of response inhibition did not show any relationship with the frequency of experiencing intrusive memories Table 3 Summary of Hierarchical Regression Analysis with IES scores as dependent variable and the Control Variables and PI (number of trials) as Predictors (N = 91) Model 1: Number of list 2 trials as predictor

B

SE

Beta

0.08 -0.01 0.11 0.13

0.02 0.01 0.46 0.23

0.39** -0.31** 0.02 0.06

CES-D IES_recency

0.08 -0.01

0.02 0.01

0.36** -0.29**

Gender Number of trials list 1 Number of trials list 2

0.10 0.16 0.13

0.44 0.22 0.05

0.02 0.07 0.23*

0.08 -0.01

0.02 0.01

0.34 -0.31**

Predictors Ste p 1

(∆)R ²

DF

F-change

21%

4, 86

5.72**

CES-D IES_recency Gender Number of trials list 1 2

5%

3

1,85

5.81**

5% CES-D IES_recency

1,84

5.66*

Gender -0.08 0.44 -0.02 Number of trials list 1 0.22 0.22 0.09 Number of trials list 2 0.12 0.05 0.21* -0.52 0.22 -0.22* Gender * trials list 2 Note. CES-D = Center for Epidemiologic Studies Depression Scale; IES_Recency = Time in months since the event described on the IES took place; Number of trials list 1 = number of trials needed to reach 3 correct answers on the first list of the interference task; Number of trials list 2 = number of trials needed to reach 3 correct answers on the second list of the interference task. *p < .05 **p < .01.

- 54 -

4. Discussion The main goal of the present study was to investigate the relationship between relatively weak inhibitory control and the frequency of experiencing intrusive memories. The results can be summarized as follows. First, as predicted, a relatively reduced ability to resist PI was associated with a relatively high frequency of intrusive memories related to an aversive event in the past. This relationship was most pronounced in women and still present after controlling for differences in recency of the experienced event and was independent of individual differences in negative mood. Second, individual variation in the frequency of intrusive memories did not show any relationship with performance on measures of response inhibition. In agreement with previous findings using more indirect measures of interference control (e.g., Klein & Boals, 2001; Verwoerd & Wessel, 2007), the results of the present study provide additional and more direct evidence for the idea that individual differences in interference control may explain the sensitivity to experiencing undesirable intrusive memories. Moreover, our finding that the frequent experience of intrusive images and emotions was specifically related to deficient inhibitory control at the cognitive level (i.e., resistance to PI), replicates and extends the findings of Friedman and Miyake (2004). Furthermore, the present study showed that our main finding of a relationship between resistance to PI and intrusive memories could not be explained by elevated levels of depression. Earlier studies showed that both weakened cognitive/inhibitory control and a high frequency of intrusive cognitions are related to elevated levels of depression (Ellis, 1990; 1991). By including a self-report index of depression we countered any confounding effect that a negative mood might have on the relationship between resistance to PI and intrusive memories. Finally, our findings stress the importance of selecting the appropriate individual inhibition measures when conducting applied research exploring deficits in different disorders such as PTSD or ADHD (Nigg, 2000, Pennington & Ozonof, 1996). More generally, our finding of a unique relationship between intrusive memories and the ability to resist PI is also consistent with earlier work indicating that several independent inhibitionrelated functions are involved in different facets of cognition and behavior (Dempster, 1993; Harnishfeger, 1995; Nigg, 2000). In the current study, we also anticipated that repeatedly retrieving second-list responses would result in a relatively decreased accessibility for the interfering first-list responses on later (third-list) retrieval attempts. Moreover, it was hypothesized that people

- 55 -

Chapter 3

with a relatively well-developed ability to reduce the accessibility of this earlier interfering/unwanted information, would show a relatively low frequency of intrusive memories. Yet, in the present study, no evidence emerged to support this idea. There was no relationship between the relative delay in retrieving first-list responses during the third stage and the frequency of intrusive memories. It should be acknowledged, however, that the absence of such a correlation in the present study does not rule out the possibility that the accessibility of earlier unwanted material may nevertheless be involved in individuals suffering from intrusive cognitions. That is, the absence of such a relationship might be due to the methodological characteristics of the present study. For example, in an attempt to increase the sensitivity of the present PI task as an index of individual differences in inhibitory capacity, we removed the extra learning phase before the cued recall of the third list response words. We anticipated that the original procedure of Rosen and Engle (1998) including such an extra phase would reduce the effect of the suppression of the first list, thereby undermining the task’s sensitivity as a measure of individual differences. However, on the basis of the present findings it can not be ruled out that the absence of such an extra learning phase resulted in such a strong interference effect of second-list responses, that individual differences in inhibitory capacity were completely overshadowed (correlated costs and benefits problem, Anderson & Levy, 2007). In addition, it might be that our color manipulation was not sufficiently successful in reducing individual differences in maintaining the task goal in an active state (i.e., “retrieving first-list responses”). If so, individual differences in RT might reflect differences in goal neglect (e.g., Kane & Engle, 2003) rather than inhibitory capacity. Future studies are necessary to more definitely settle these issues. Furthermore, the variance in IES-scores explained by our interference measure in the overall sample may seem rather low and not of clinical significance. Yet, the results of the current study may well have underestimated the true variance in persistent intrusions that could be explained by inhibitory ability in trauma-exposed individuals. The analogue nature and methodology of the present study has the inherent characteristic of only partly representing the situation in the aftermath of a traumatic event. For example in the present study we used a retrospective and subjective measure of major life events. Furthermore, even though these events were very stressful, they would not meet the DSM-IV (APA, 2000) criteria for a traumatic stressor. Following this, for the undergraduates in the current study,

- 56 -

the role of inhibitory control in reducing the impact of intrusive re-experiencing may be of less importance than for trauma-exposed individuals for whom it is of crucial importance to counter the interference of very distressing intrusive memories. The significant interaction between inhibitory capacity and gender that was evident in the present study is consistent with this idea. That is, as women have found to be at increased risk for the development of PTSD after trauma (Brewin et al,, 2000), a well-developed ability to resist PI would be especially important in this subgroup. Future studies should be used to more specifically test this idea. Finally, although the absence of a relationship between response inhibition and the frequency of intrusions was in line with our predictions, it should be acknowledged that this was tested by accepting the null hypothesis (i.e., no effect). This leaves room for alternative explanations for the current differential pattern of findings regarding response inhibition and resistance to PI. For example, one could speculate that the unique relationship between resistance to PI and intrusive memories may reflect differences in task difficulty. Although we cannot rule out this possibility, it has to be noted that our rationale for selecting the indices for response inhibition and resistance to PI was theory driven. We closely followed the ideas of Friedman and Miyake (2004) who used confirmatory factor analysis to show that the color-word Stroop and RNG task had high loadings on a hypothetical construct they called response inhibition. This construct was part of a model describing the existence of several independent inhibitory mechanisms supporting different classes of goal directed behaviors. In selecting the indices used in the present study, we followed Friedman and Miyake’s (2004) idea that behavior can be independently controlled at the cognitive level and the response level. Although our results were in line with this idea, future studies using more sophisticated statistical techniques such as structural equation modeling (SEM) are required to arrive at more solid conclusions concerning the differential involvement of response inhibition and resistance to PI in the generation of undesirable intrusions in memory. As noted earlier, the results of the present study could be important for studies investigating risk factors and PTSD. According to Kindt and Engelhard (2005), it is striking that during the last decade, a plethora of studies has been conducted on discovering risk factors of PTSD (see Brewin et al., 2000 and Ozer et al., 2003 for reviews), while only a minority of studies investigated the underlying pathogenic mechanism. Following this, renewed interest in the inhibition metaphor in cognitive psychology (e.g., Dempster &

- 57 -

Chapter 3

Corkill, 1999; Nigg, 2000) may enrich thinking about inhibition in psychopathology research. For example, Dempster (1999) stressed the importance of inhibitory control in a wide range of intelligent behaviors. Following this, one could argue that a weakened ability to resist interference from working memory might explain the well-established finding that low (premorbid) intelligence is a risk factor for PTSD (e.g., Macklin et al., 1998; McNally, 2003). Considering this, our findings may help explain the clinical observation that in the aftermath of a traumatic event, for most people the overwhelming intrusive memories would subside within one month (see Rothbaum, Foa, Riggs, Murdock & Walsh, 1992), while a minority of them will stay vulnerable and eventually develop PTSD. In conclusion, the present findings are important in that they give evidence for a more direct relationship between individual differences in inhibitory control and the frequency of intrusive memories. Moreover, it was shown that intrusive memories are specifically related to the ability to resist PI from working memory while the ability to resist automatically triggered (prepotent) responses did not show any relationship with the frequency of intrusive memories. Future research is necessary to examine the causal status of resistance to PI in the development and maintenance of involuntary memories. Moreover, it would be interesting for future studies to follow up the idea that apart from individual differences in sensitivity to PI, individual differences in the ability of reducing the accessibility of earlier interfering representations (e.g., Anderson & Spellman, 1995) might be related to the frequency of intrusive memories.

- 58 -

Appendix Lists of Words used in the Proactive Interference Task AB AC Slager – Vlees Slager – Touw (Butcher – Meat) (Butcher – Rope) Huis – Tuin Huis – Bank (House – Garden) (House – Couch) Naald – Draad Naald – Steek (Needle –Thread) (Needle –Stitch) Goud – Zilver Goud – IJzer (Gold – Silver) (Gold – Iron) Schaap – Wol Schaap – Hei (Sheep – Wool) (Sheep – Moor) Mouw – Jas Mouw – Aap (Sleeve – Coat) (Sleeve – Ape) Tabak – Pijp Tabak – Opa (Tobacco – Pipe) (Tobacco – Grandpa) Kaas – Melk Kaas – Ham (Cheese – Milk) (Cheese – Gammon) Bed – Deken Bed – Poten (Bed – Blanket) (Bed – Legs) Kwast – Verf Kwast – Hout (Brush – Paint) (Brush – Wood) Gips – Been Gips – Arm (Plaster – Leg) (Plaster – Arm) Ketel – Pan Ketel – Gas (Kettle – Pan) (Kettle – Gas)

- 59 -

Chapter 4

- 60 -

Chapter 4 Low Attentional Control and the Development of Intrusive Memories Following a Laboratory Stressor J. Verwoerd, P.J. de Jong, I. Wessel Journal of Psychopathology and Behavioral Assessment (2008: 30, 291-297)

This study tested whether low attentional control3 set people at risk for experiencing undesirable intrusions. Participants completed measures of attentional control and neuroticism and subsequently watched an emotional film fragment. In the four days following the presentation of the fragment, half of the participants (n = 17) were asked to keep a diary for the registration of intrusive memories. The other half of the participants (n = 16) only rated the number of intrusions retrospectively during the follow-up session. Low attentional control had independent predicting properties for the development of intrusive symptoms in the diary group. No such relationship was found in the no-diary controls, probably due to the relatively low frequency of intrusive symptoms that was elicited in this group.

3

In this chapter we use the term attentional control to refer to a general executive ability to hold or maintain the attentional spotlight on relevant information in working memory or the outside world (internally and externally focussed attention). This could be accomplished by the inhibition of irrelevant information or the activation of relevant information.

- 61 -

Chapter 4

1. Introduction One of the hallmark symptoms of posttraumatic stress disorder (PTSD) is the repeated occurrence of unwanted intrusive memories of the traumatic event. It appears that most people, after having experienced a traumatic situation, get rid of these overwhelming intrusive memories within one month (see Rothbaum, Foa, Riggs, Murdock & Walsh, 1992), while a minority of them will stay highly vulnerable for cues triggering involuntary memories and will eventually develop PTSD. Considering this, vulnerability for persistent involuntary reliving of former traumatic events could imply a dysfunction in the mechanisms of memory control. Germane to this, a series of recent cross-sectional studies using analogue samples provided evidence indicating that a relatively low level of executive control over the contents of working memory (i.e., working memory capacity [WMC]) is associated with a relatively frequent occurrence of intrusive cognitions (Brewin & Beaton, 2002; Brewin & Smart, 2005; Klein & Boals, 2001). It has been proposed that individual differences in WMC are related to a domain-general capability to control attention¹, which is particularly important in situations involving proactive interference or conflict between competing response tendencies (Engle, 2002). For example, in a study with normal undergraduate students, the frequency of experiencing intrusive memories and attempts to avoid such memories were relatively high in people who scored relatively low on a behavioral index of WMC (i.e., Operation Span; Klein & Boals, 2001). In line with this, it has been shown (Brewin & Beaton, 2002; Brewin & Smart, 2005) that individuals with a relatively high WMC are better able to suppress unwanted thoughts. Finally, using a self-report measure of distractibility (which might be taken as an index of deficient attentional control) it was found that people who described themselves as highly distractible also reported a relatively high frequency of intrusive memories (Verwoerd & Wessel, 2006). These earlier results are in line with the idea that deficient attentional control functions as a general vulnerability factor for development and maintenance of intrusive memories. However, because these studies were all cross-sectional in nature it remains to be seen whether indeed low levels of attention or working memory control set people at risk for experiencing intrusive memories. To further explore this issue the present study was intended as a first step to test the predictive properties of (self-reported) attentional control on the development of experimentally induced intrusive memories using a prospective design.

- 62 -

A common laboratory method of investigating the development of intrusive memories is the use of a stressful film fragment. The stressful film paradigm creates an analogue situation in which response to trauma can be investigated in a controlled way (Holmes, Brewin & Hennessy, 2004). A number of studies investigating information processing abnormalities have used this method to study the development of intrusive memories in the days after watching a highly stressful film (Brewin & Saunders, 2001; Davies & Clark, 1998; Holmes, et al., 2004; Stuart, Holmes & Brewin, 2006). In most of these studies, participants were presented with a diary for the online registration of the frequency and characteristics of their intrusions rather than via an unexpected presentation of retrospective self-report questionnaires at a follow up session. There may be two consequences of using a diary for the registration of intrusive memories. On the one hand, using a diary may undermine the sensitivity of the present design to find a (negative) relationship between attentional control and the frequency of intrusive memories. That is, the frequency of reported diary intrusions may be influenced by the ability to keep the task goal of online diary registration in an active state. This may result in an artificially inflated number of diary intrusions in participants high in attentional control (potentially even reversing the theoretically expected negative correlation between attentional control and intrusive memories). On the other hand, the use of a diary may generally increase the activation level of the stressful memory. Yet, it may well be that the material used in analogue studies lacks sufficient emotional intensity to elicit intrusive memories very easily. If so, an increase in memory activation levels through the use of a diary may be a prerequisite for triggering a sufficient number of intrusive memories in order to reliably detect individual differences in the development of intrusive memories. Therefore, the present study included both a diary group and a no-diary control group to test the influence of using a diary on the number of reported intrusions as well as on the relationship between attentional control and intrusive memories. The second aim of the present study was to investigate the role of neuroticism on intrusive memory development. It has been suggested that people high in the personality trait neuroticism may display distorted cognitive/information processing patterns which predispose them to show strong emotional reactivity and instability when confronted with threatening and stressful life events (Muris, 2006). In line with this, earlier studies using clinical samples found a relationship between neuroticism and PTSD symptom severity (e.g., Morgan, Matthews, & Winton, 1995; Vassiliki & Tarrier, 2001; but see Engelhard, van den

- 63 -

Chapter 4

Hout & Kindt, 2003 for a different interpretation). Furthermore, neuroticism has also been found to be strongly related with attentional control (r’s between -.40 and -.57; e.g., Derryberry & Reed, 2002; Muris, 2006; Muris, de Jong & Engelen, 2004). Therefore, to get a better idea of the unique contribution of attentional control, the current study controlled for any mediating role that elevated levels of neuroticism might have on the relationship between low levels of attentional control and intrusive memories. In sum, the main goal of the present study was to investigate the predictive value of deficient attentional control on the development of intrusive memories related to an earlier (experimentally elicited) aversive event. It was expected that relatively low levels of attentional control would predict relatively high levels of film-related intrusive memories in the four days after watching a stressful film. Additionally, it was expected that the diary method for the registration of intrusions would affect the predictive value of attentional control on self-reported intrusive and avoidance symptoms at follow-up. Finally, the role of neuroticism as a vulnerability factor and possible mediator was explored. It was expected that relationship between attentional control and intrusive memories would still exist after controlling for the shared variance with neuroticism.

- 64 -

2. Method 2.1. Participants Thirty-three undergraduates participated in this study (18 men and 15 women). The mean age was 20.68 years (SD = 2.89, range: 18-47). Participants (contacted by email or phone) were told that they would see a shocking film fragment with possible distressing effects. Following recruitment, they were randomly allocated into either a “diary group” or a “control group”. Seventeen participants (8 men, 9 women) were included in the diary group and sixteen participants were included in the control group (10 men, 6 women). All participants gave their written informed consent to take part. The University of Groningen Behavioral Research Ethical Committee granted ethics approval for this study.

2.2. Stressful film fragment The experimental manipulation used in this study consisted of a 9-min fragment of “Irreversible” produced by Gaspar Noé. This film contains shocking and gruesome images of a brutal murder in a dark nightclub with very explicit and extreme violence. Subjects were not acquainted with the film before seeing the fragment. With respect to possible negative side effects of showing a film with a shocking content, participants were explicitly informed during recruitment and the experimental session prior to viewing the video that the film contained extreme violence and that they could leave the room and terminate the experiment at any point. The fragment was projected on a large (150 x 120 cm) screen in a darkened room.

2.3. Measures 2.3.1. Attention and distress. In order to measure the amount of attention paid to the film fragment, participants rated the percentage of time they had looked away from the screen (0 = not at all - 100 = all of the time). They also reported on a 100 mm Visual Analog Scale (VAS; 0 = not at all -100 = extremely) how much distress they experienced while viewing the fragment. 2.3.2. Attentional control. In this study, we used a Dutch translation of the Attentional Control Scale (ACS, Derryberry & Reed, 2002; Verwoerd, de Jong & Wessel, 2006).The ACS is a 20-item self-report measure which measures attentional focusing (i.e., the ability to focus attention on one task: “When I need to concentrate and solve a problem, I have trouble

- 65 -

Chapter 4

focusing my attention”) and attentional shifting (i.e., the ability to shift attention from one task to another and to engage in dual task behavior: “It is easy for me to alternate between two different tasks”). Scores on the ACS can range from 20 to 80 and the response format uses a 4-point scale (1 = almost never; 2 = sometimes; 3 = often; 4 = always), higher scores indicating good attentional control. The internal consistency of the ACS was found to be acceptable in the present sample (Cronbach’s α = .71). 2.3.3. Neuroticism. Neuroticism was measured with the Five Factor Personality Inventory (FFPI; Hendriks, Hofstee & de Raad, 1999) which is a self-report questionnaire for measuring the five super ordinate factors of temperament. The emotional stability scale of the FFPI consists of 20 items (e.g, “Is afraid to do something wrong”) answered on a five-point scale (1 = does not at all apply to me, 5 = does very much apply to me). For half of the items, the direction was reversed, so that higher scores would indicate higher neuroticism. The internal consistency of the emotional stability scale used in this study was high (Cronbach’s α = .93). 2.3.4. Diary. Participants in the diary group were asked to use an A5-format diary to record any intrusions of the films during the four days following the film (e.g., Brewin & Saunders, 2001; Davies & Clark, 1998; Holmes et al., 2004). They noted each intrusion’s content (“Describe the content of your intrusion.”) and rated the percentage of distress associated with experiencing the intrusion (0 = no distress – 100 = high levels of distress). Elaborate verbal and written instructions were given about the nature of unwanted intrusions and how to keep the diary. As in re-experiencing symptoms in PTSD, intrusions were defined as “spontaneously occurring” (not intentional) memories of the film (see Holmes et al., 2004). Participants were asked to carry the diary with them and to record every intrusive memory related to the film fragment as soon as possible. The main dependent variable was the total number of intrusions reported in the diary. At follow up, participants had to fill in a diary compliance rating. They were asked to rate on a 100 mm VAS scale how good they had been able to keep track of their intrusions in the diary (0 = not at all – 100 = very good). 2.3.5. Impact of Event Scale. A modified Dutch version of the Impact of Event Scale (IES, Horowitz, Wilner & Alvarez, 1979; see van der Ploeg et al., 2004) instructed participants to respond to statements related to the film fragment they had been shown a few days earlier (e.g., images came up spontaneously). The modified scale consisted of 11 items answered on

- 66 -

a five-point scale (range: 0 – 55), with assigned item weights of 0, 1, 3, 5. The internal consistency of this scale was found to be high (Cronbach’s α = .89).

2.4. Procedure Participants were recruited from fellow students and friends of five research assistants who had to finish a second year research practical. In the week following the randomization, all participants filled in the ACS and were informed (by email or phone) about the time and day they were expected to participate in the experiment. Both the diary group and the control group watched the distressing fragment in group sessions on separate hours of the same day at the beginning of the week. Upon arrival at the test room, participants gave informed consent. After this, they were asked to fill in the neuroticism scale and were instructed that they could leave the room at any time. Before starting the distressing fragment, the experimenter gave instructions not to talk with each other and to avoid looking away as much as possible. Directly after viewing the film, participants rated the amount of attention they had paid to the film and how much distress they had experienced during the fragment. After this, participants in the diary group were instructed in the use of the 4-day diary while participants in the control group were only asked to come back after four days. The follow-up was conducted four days after the experimental manipulation. Participants in the diary group delivered their diaries and completed the diary compliance rating. All participants filled in the IES. Before leaving, the experimenter debriefed the participants about the purpose of the study. 2.5. Statistical analyses Hierarchical multiple regression analysis (HMR) was used to examine the main research questions of the predictive value of attentional control and the possible moderating role of using a diary on IES-scores at follow up. In the first step of this analysis, the variable group (1 = diary, 0 = control) was included as a dummy variable, together with the attentional control scale. In the second step of the analysis, the moderating role of using a diary was tested by forcing the normalized product (interaction) of attentional control and group into the equation. In order to avoid further lowering of the participants/predictor ratio of the regression model, additional analyses used partial correlations to examine any mediating effects of neuroticism.

- 67 -

Chapter 4

3. Results One participant reported that she did not understand the questions on the neuroticism scale. Therefore, the analyses with neuroticism were conducted without this participant.

3.1. General results Mean, standard deviation and range of the main variables are shown in Table 1. The two groups (diary and control) did not differ on measures of attentional control or neuroticism and the percent of time looked away from the stressful fragment. Both groups also rated the distressing fragment as moderately emotional. Table 1. General Statistics (Mean, Standard Deviation (SD) and Range) of the Main Variables in the Diary and Control Group Diary (N = 17) Control (N = 16) Mea SD Range Mea SD Range t P n n Attentional 53.0 5.18 43-64 51.8 5.95 46-65 -.64 n.s. control 6 1 Neuroticism 44.4 11.0 26-61 39.8 8.19 30-58 -.94 n.s. 1 4 8 Emotionality 67.3 28.7 6-100 55.8 23.9 11-80 -1.25 n.s. 5 9 1 7 Percent of time 6.59 9.28 0-34 7.50 16.3 0-56 .20 n.s. looked away 1 Follow-up (IES) 11.4 10.9 0-38 4.06 4.42 0-14 3.21 <.01 7 4 Diary intrusions 2.59 1.84 0-7 Diary 64.2 21.8 10-96 compliance 4 4 Note. Emotionality = Emotionality ratings directly after viewing the film fragment; Percent of time looked away = Percent of time participants looked away from the film fragment; IES = Impact of Event Scale.

3.2. The effect of keeping a diary on the relationship between self-reported attentional control and IES-scores at follow-up The HMR analysis with the IES as dependent variable, resulted in a significant 28% of explained variance in first step of the model, F (2, 30) = 5.85, p < .01. Within this first step, the dummy variable group (diary, control), was a strong predictor of the variance in IES - 68 -

scores, β = 0.50, p < .01. Participants who filled in a diary during the four days after viewing the distressing film fragment reported considerably more intrusions at follow-up (M = 11.47, SD = 10.91) than people in the control group (M = 4.06, SD = 4.41). However, the contribution of attentional control did not reach conventional levels of significance, β = -0.24, t (31) = -1.54, p = .14. The second step containing the interaction term between group and attentional control added a significant 9% to explained variance, F-change (1,29) = 4.23, p < .05. To further explore this interaction term, separate Pearson correlations between IES scores and attentional control were computed for the diary and the control group. It appeared that for participants who filled in a diary, self-reported attentional control strongly predicted follow up IES scores [r(17) = -.57, p < .05 ]. In contrast, individual differences in attentional control for the participants in the control group did not have any predictive value for the later reported intrusive memories at follow-up [ r(16) = .05, p = .87].

3.3 Pearson correlations between neuroticism and IES-scores at follow-up As shown in Table 2, the relationship between attentional control and neuroticism appeared to be moderately high [ r(32) = -.42, p < .05 ]. Furthermore, as expected, neuroticism was strongly associated with IES-scores at follow up [ r(32) = .51, p < .01]. In general, there was a moderate to strong tendency for people with relatively high scores on neuroticism to experience a relatively high frequency of intrusive memories. Table 2. Pearson Correlations Between the Main Variables used in the Present Study 1 2 3 4 5 1. Attentional control _ 2. Neuroticism total score -.42* _ 3. Follow-up (IES) -.18 .59** _ 4. Emotionality -.12 .45** .44** _ 5. Diary intrusions (N = 17) -.58* .37 .53* .21 _ 6. Diary compliance (N = 17) -.24 -.32 .41 -.23 .15 Note. IES = Impact of Event Scale; Emotionality = Emotionality ratings directly after viewing the film fragment. *p < .05. **p < .01.

- 69 -

Chapter 4

3.4. The mediating role of neuroticism in explaining the relationships of attentional control with IES-scores and number of intrusive memories recorded in the diary. Since there was no significant association between attentional control and IES scores in the control group, all subsequent analyses are restricted to the diary group. The results showed that, after controlling for neuroticism, there was still a trend towards significance for the correlation between attentional control and IES-scores at followup, [ pr(13) = -.50, p = .058]. Additionally, the unique relationship between attentional control and the number of intrusive memories recorded in the diary was of considerable magnitude and in the same direction, although it failed to reach conventional levels of significance [ pr(13) = -.43, p = .11]. Taken together, these results suggest that weakened attentional control has independent predicting validity (over and above neuroticism) for the frequency of intrusive memories after an experimentally induced stressful situation. However, this predictive relationship was only evident in the group that used a diary as a method for the registration of intrusive memories.

4. Discussion The major purpose of the current study was to explore the predisposing role of low attentional control and neuroticism on the development of intrusive memories in a prospective design. The main results can be summarized as follows. First, in line with previous studies, high levels of neuroticism were associated with a relatively high frequency of undesirable intrusive memories related to an experimentally induced aversive event. Second, for participants in the diary group, attentional control indeed had predictive value for the occurrence of subsequent intrusive and avoidance symptoms. Meanwhile, a similar relationship was absent for the no-diary control group The prognostic relationship between deficient attentional control and the frequency of undesirable intrusive memories in the diary group corroborate the results of prior studies relying on retrospective measures of intrusive memories. For example, earlier retrospective studies related vulnerability for experiencing intrusive memories to self-reported distractibility (Verwoerd & Wessel, 2006), to low working memory capacity (WMC) (Brewin & Beaton, 2002; Brewin & Smart, 2005; Klein & Boals, 2001), and to a reduced ability to resist or inhibit interference from working memory (Verwoerd, Wessel & de Jong, submitted). Interestingly, the present findings give preliminary support our hypothesis that

- 70 -

the predictive relationship between low attentional control and intrusive memory development would be largely independent of individuals’ level of neuroticism. That is, attentional control uniquely predicted 25% of the variance in intrusive memories reported at the follow-up session. Meanwhile, consistent with the notion that neuroticism is a vulnerability factor for developing strong reactivity and instability when confronted with threatening/traumatic events (Muris, 2006), neuroticism was shown to have predictive value for the development of intrusive memories. This pattern of results is in line with earlier analogue studies (Muris, 2006; Muris, de Jong & Engelen, 2004) by showing unique contributions of attentional control and neuroticism in the context of intrusive memory development. Some comments are in order to explain why the relationship between low attentional control and a high frequency of intrusive memories did only present itself in the diary group. The diary group reported almost three times as many IES-intrusions at the follow-up session than the no-diary control group. This finding points to the possibility that keeping a diary for the registration of intrusive memories in analogue designs is a prerequisite to reliably evoke a meaningful number of intrusions in non-clinical individuals. That is, repeatedly thinking about the task goal of keeping a diary may increase the activation level of the stressful film memory and may subsequently trigger intrusive memories. Following this, for people in the diary group, good attentional control would be an indispensable ability to prevent undesirable intrusive memories from entering working memory. In contrast, for participants in the control group, the emotional fragment might not be sufficiently intense to develop a minimum number of intrusive memories. So, in this group, there was no need for executive attention to prevent the experience of undesirable intrusive memories. On a more theoretical level, the question may rise how participants high in selfreported attentional control did prevent intrusive memories from entering working memory. Extending on earlier findings in the context of WMC (e.g., Klein & Boals, 2001; Kane & Engle, 2000), described as a domain-general capability to control attention, a recent study tested the idea that the ability to resist or suppress interference of irrelevant material from working memory might play a role in this. In this study, which used a retrospective design, individuals who reported a relatively high level of intrusive and avoidance symptoms on the IES showed a reduced ability to resist interference of first-list intrusions during second-list learning in an AB-AC-AB list-learning paradigm (Verwoerd, Wessel & de Jong, submitted).

- 71 -

Chapter 4

Thus, a deficient ability to resist or suppress interference from working memory might be the underlying mechanism which sets people low in self-reported attentional control at risk for experiencing undesirable intrusive memories. Germane to this, the current findings of a relationship between low attentional or working memory control and intrusive memories might also contribute to further understanding the common finding of intelligence (IQ) as risk factor for the development and maintenance of PTSD symptoms ( e.g., MacNally & Shin, 1995). Some limitations of the present study need attention. First, the results of the present study relied primarily on the use self-report measures. Regarding the use of the attentional control scale, it might be possible that subjective reports about one’s ability to control attention are not similar to more objective indices of attentional control. Therefore, further research is needed to test the predictive relationship of attentional control on the development of intrusive memories with more objective (experimental) measures. Second, it should be acknowledged that the size of the diary group was rather modest. So it seems important to investigate in future studies whether the main findings of the present study can be replicated in larger samples using extreme groups of high and low attentional control. Third, the selection of participants under fellow students of research assistants should be taken with caution. It is not clear to what extent these participants are representative of undergraduates in general, or the larger population. Future research should make use of more unbiased selection methods. Fourth, one further limitation/implication for future analogue diary studies might be that participants’ strategies to keep track of experimentally elicited intrusive memories may not model the behavior of people in the aftermath of a real traumatic event in a correct way. That is, after a traumatic event, people try to avoid thinking about the trauma as much as possible instead of consciously keeping track of their intrusions. Meanwhile, strategies to avoid thinking about an aversive event may have a similar effect on intrusive remembering as the strategies used to keep the task goal of online diary registration in an active state as participants probably did in the present study. In conclusion, the present findings are important in that they give preliminary evidence for a prospective relationship between attentional control and the development of intrusive memories after the experience of a stressful situation (e.g., film fragment). Furthermore, there was some indication that the relationship between attentional control and intrusive memories was not mediated by elevated levels of neuroticism, preliminary

- 72 -

supporting a unique predictive value. However, the most important findings of the present study were restricted to the participants who filled in a diary for the registration on intrusive memories. Possibly, the use of a diary supported the intrusiveness of the stressful film memory representation which in turn supported the detection of individual differences in the predictive value of attentional control. An important next step would be to replicate the present findings in larger samples using behavioral rather than self report measures of attentional control. Furthermore, the methodology of the present study might also be extended to clinical samples by measuring low attentional control and high neuroticism pretrauma and investigating the subsequent incidence of PTSD.

- 73 -

Chapter 4

- 74 -

Chapter 5

Pre-Stressor Interference Control and Intrusive Memories

J. Verwoerd, I. Wessel, P.J. de Jong, M.M.W. Nieuwenhuis (submitted)

Although intrusive imagery is a common response in the aftermath of a stressful or traumatic event, only a minority of the trauma victims show persistent re-experiencing and related psychopathology. Individual differences in pre trauma executive control have been proposed to play a critical role in here. Therefore, this study investigated whether a relatively poor pre stressor ability to resist proactive interference in working memory might set people at risk for experiencing undesirable intrusive memories after being exposed to a stressful event. Non-selected participants (n = 85) completed a modified version of a widely used test of interference control in working memory (CVLT; Kramer & Delis, 1991) and subsequently watched an emotional film fragment. Following the presentation of the fragment intrusive memories were recorded in an one-week diary and at a follow up session 7 days later. A relatively poor ability to resist proactive interference was related to a relatively high frequency of filmrelated intrusive memories. This relationship was independent of prior trauma exposure, negative mood, neuroticism and gender. These findings are consistent with the idea that a pre morbid deficit in the ability to resist proactive interference reflects a vulnerability factor for experiencing persistent intrusive memories after trauma exposure.

- 75 -

Chapter 5

1. Introduction Although intrusive imagery is a common response in the aftermath of a stressful or traumatic event, only a minority of the trauma victims show persistent re-experiencing and related psychopathology (Rothbaum et al., 1992). Considering this, one of the key issues in research on post-trauma information processing concerns individual differences in vulnerability. Inspired by recent findings in the experimental memory literature, this study focuses on the role that differential pre trauma executive control might play in dysfunctional information processing and persistent intrusive remembering after trauma. Executive control reflects a collection of separable meta-level abilities that regulate other (lower level) cognitive processes in the service of goal-directed behavior. One of these executive abilities concerns response inhibition (RI). RI refers to the ability to inhibit dominant responses in order to carry out a weaker primary and incompatible response and may be involved in dysfunctional impulsive behavior (Friedman & Miyake, 2004; Nigg, 2000). Another critical executive ability can be described as resistance to proactive interference (PI). PI refers to the ability to remove information that is no longer relevant from working memory (Friedman & Miyake, 2004; for other executive abilities see Friedman et al., 2008; Miyake, Friedman, Emerson, Witzki & Howerter, 2000; Hedden & Yoon, 2006). As an example of PI, consider the first weeks after purchasing a new mobile phone. During this period, it would be difficult to access the new number in long term memory because the extensively used old number will persistently intrude into conscious awareness. Flexible access to the memory representation of the new number would require a well-developed ability to resist unwanted interference from the old number. Interestingly, using latent variable analysis, Friedman and Miyake (2004) found a relationship between resistance to PI and self-reported general intrusive thoughts. This raises the question if this specific executive ability of resistance to PI may also contribute to the gradual decrease of intrusive imagebased memories that generally arise in the immediate aftermath of trauma (a prominent example of PI in real life). That is, a deficient ability to resist PI may complicate a rapid decline of these intrusive memories and may thus constitute a vulnerability factor for their persistence. In line with this, a recent cross-sectional study showed that a relatively good ability to resist PI was associated with fewer self-reported intrusive memories related to an earlier experienced stressful life event (Verwoerd, Wessel & de Jong, 2008). This earlier retrospective finding gives some support for the notion that a deficient ability to resist

- 76 -

interference from irrelevant material in working memory may set people at risk for persistent intrusions in the aftermath of a stressful or traumatic event. A critical question that remains is whether these individual differences in executive functioning already exist prior to experiencing trauma. Two lines of evidence have now begun to provide support for the notion that a relatively low ability to resist PI may act as a pre-morbid risk factor. First, the results of a large twin study under more than 500 adolescents showed that individual differences in general executive control as well as in specific executive abilities were almost entirely genetic in origin (Friedman et al., 2008). Second, using a stressful film fragment as an analogue stressor (e.g., Holmes and Bourne, 2008), a recent prospective study showed a relationship between pre-film resistance to PI (measured via AB-AC list-learning) and intrusive cognitions measured 24 hours after the presentation of the film (Wessel, Overwijk, Verwoerd, & de Vrieze, 2008). These findings provide additional support for the proposed role of a reduced ability to resist PI in the persistence of intrusive memories after trauma. However, because both the cross-sectional study of Verwoerd et al. (2008) as well as the prospective study of Wessel et al. (2008) used the same measure (i.e., paired-associates learning) to index resistance to PI, it might still be the case that task-specific factors (rather than resistance to PI) were responsible for earlier found associations with intrusive cognition. Important to mention in this context is the task impurity problem (e.g., Friedman & Miyake, 2004). This impurity problem holds that about 80% of the variance on individual indices of executive control is related to task specific abilities. One method of resolving this problem would be a conceptual replication with the use of a different individual index of the ability to resist interference. Therefore, the first goal of the present study was to replicate the earlier finding of Wessel et al. (2008) using a modified version of the California Verbal Learning Test (CVLT) as an index of resistance to PI. The CVLT is a widely used neuropsychological test to asses multiple aspects of verbal learning and memory (Delis & Kramer, 1991). In the CVLT, participants learn two lists of words (List A and List B) that are made up of different semantic categories (e.g., animals, vegetables). Some categories are similar between lists (shared categories) while others change from the first- to the second list (unshared categories). In line with earlier studies using the CVLT, the present study used the decrease in recall performance (instead of cued recall; e.g., Wessel et al. [2008]) between the first and

- 77 -

Chapter 5

second list shared category words as an index of the ability to resist PI. In addition, intrusive memories were recorded in a one-week diary. This larger time window for the collection of intrusive memories (7 days instead of 24 hours [Wessel et al., 2008]) is more in line with earlier trauma-film studies (Holmes, Brewin & Hennessy, 2004) and was intended to increase the sensitivity of the design to detect the alleged influence of individual differences in persistent intrusive re-experiencing. A second goal of the current study was to investigate the role of neuroticism. Some theoretical models have suggested an interaction between attentional/executive control and neuroticism in the prediction of anxiety (Derryberry & Reed, 2002). That is, people with low levels of attentional/executive control, together with high levels of neuroticism are expected to show an increased risk of developing pathological anxiety. In line with this idea, the second goal of the present study was to investigate if the influence of a weakened ability to resist interference from working memory on the persistence of intrusive remembering is especially pronounced in individuals with high levels of neuroticism. The final target of the present prospective design was to more specifically test the relationship between resistance to PI and image-based intrusive memories. Earlier clinical observations showed that intrusive memories related to stressful or traumatic events usually consist of the unwanted occurrence of sensory fragments of the event, often accompanied with high levels of distress (Hackman, Ehlers, Speckens & Clark, 2004). It has been found further that intrusions in the visual modality are the most common (Ehlers, Hackman, & Michael, 2004). In a recent study, Ehlers et al. (2004) distinguished intrusive memories from non-memory intrusions that may also be experienced as intrusive (Ehlers & Clark, 2000). Non-memory intrusions may include intrusive thoughts which appear in an evaluative verbal form (e.g., rumination; Murray, Ehlers & Mayou, 2002) with no direct relationship to the earlier stressful event. As the earlier study of Wessel et al. (2008) mainly focused on filmrelated intrusive thoughts, a more specific analysis of the frequency and modality of filmrelated memories would increase the external validity of the present analogue design. In line with earlier studies (Wessel et al., 2008; see also Verwoerd, de Jong, & Wessel, 2008) we investigated the prospective link between executive control (i.e., resistance to PI) and intrusions in unselected participants employing a variant of the trauma film paradigm (Holmes & Bourne, 2008). The trauma-film paradigm provides a prospective experimental tool for studying pre- and peri-traumatic risk-factors for the development of

- 78 -

analogue intrusions. Prior to the presentation of a stressful film, participants completed a modified version of the CVLT and a self-report index of neuroticism. After film presentation, participants were asked to record of the frequency and modalities of their intrusions in a oneweek diary.

2. Method

2.1. Participants. Eighty-five undergraduate students participated in this study (20 men and 65 women4; mean age: 20.53 years, age range: 18-50). All participants gave their written informed consent to take part. The University of Groningen Behavioral Research Ethical Committee granted ethics approval for this study. Participants were given course credits for participating.

2.2. Stressful film fragment Participants were shown a 9-min fragment of “Irreversible” produced by Gaspar Noé. The fragment shows a gruesome and brutal murder in a dark nightclub (see Verwoerd, de Jong & Wessel, 2008). The film fragment was projected on a large (150 x 120 cm) screen in a darkened room.

2.3. Resistance to PI As an index of the ability to resist PI from working memory, we used a modified version of an interference subtest of the California Verbal Learning Test (CVLT; Delis, Kramer, Kaplan, & Ober, 1987). Two word lists (A and B) were constructed. List A contained 10 names of vegetables, 10 names of animals, and 10 names of flowers. List B contained 10 new names of vegetables, 10 new names of animals, and 10 names of musical instruments (see appendix). Thus, the lists shared the categories animals and vegetables but they did not share the categories flowers (list A) and musical instruments (list B). The order

4

An original sample of 115 participants took part in an extended version of the present experiment. Twenty participants were non-native speakers of the Dutch language and therefore not used in the analyses reported in the present study. Another 10 participants were excluded for the following reasons: 5 participants did not show up at the follow up session; three participants were excluded because of equipment failure on the CVLT and another 2 participants complained about feeling unwell (i.e., headache, flu) at the time of testing.

- 79 -

Chapter 5

of the lists was counterbalanced between subjects. That is, half of the participants learned list A first (order AB) while the other half learned list B first (order BA). Words on both lists were matched on frequency and typicality of category membership as much as possible (Baayen, Piepenbrock, & Rijn, 1993; Ruts et al., 2004). In the present study, the procedure for learning the two word lists was as follows. First, the 30 words of the first list were presented on the computer screen for 1 second with a 1000 ms inter-stimulus interval. Participants were told that they should try to encode the words to the best of their ability in order to recall them subsequently. The learning phase was followed by a 4 minute free recall test (trial 1). During free recall, participants’ were asked to speak out loud their responses, which were tape recorded for later analyses. Next, encoding and free recall of first list words was repeated once more (trial 2). Participants were encouraged to encode and recall more words than during the first trial. After this, the words of the second list were presented in the same way as in trial 1. The recall phase of this list consisted of 1 trial of 4 minutes verbal recall.

2.4. Neuroticism Neuroticism was measured with the Five Factor Personality Inventory (FFPI; Hendriks, Hofstee & de Raad, 1999) which is a self-report questionnaire for measuring the five super ordinate factors of temperament. The emotional stability scale of the FFPI consists of 20 items (e.g., “Is afraid to do something wrong”) answered on a five-point scale (1 = does not at all apply to me, 5 = does very much apply to me). For half of the items, the direction was reversed, so that higher scores would indicate higher neuroticism. The internal consistency of the emotional stability scale used in this study was high (Cronbach’s α = .93).

2.5. Intrusive memory measures Impact of Movie Scale (IMS). A modified Dutch version of the Impact of Event Scale (IES, Horowitz, Wilner & Alvarez, 1979) instructed participants to respond to statements related to the film fragment they had been shown 7 days earlier (e.g., “Images came up spontaneously”). The IMS consisted of 11 items answered on a five-point scale (range: 0 – 55), with assigned item weights of 0, 1, 3, 5. The internal consistency of this scale was found to be high in the present sample (Cronbach’s α = .90).

- 80 -

Intrusion diary. Participants were asked to use an A5-format diary to record any intrusions of the films during one week following the film (e.g., Davies & Clark, 1998; Holmes, Brewin, & Hennessey, 2004). In their diaries, participants described the content of each intrusion and were asked to specify their unwanted memory as “image”, “thought”, “feeling/emotion” or “something else”. Furthermore, participants also indicated the amount of distress, vividness and level of control they experienced for each intrusive memory in their diaries on 100 mm VAS scales. Elaborate verbal and written instructions were given about the nature of unwanted intrusions and how to keep the diary. Intrusions were defined as “spontaneously occurring” (not intentional) memories of the film (see Holmes et al., 2004). Participants were asked to carry the diary with them and to record every intrusive memory related to the film fragment as soon as possible upon their occurrence. The main dependent variable was the total number of intrusions reported in the diary.

2.6. Control measures Film Exposure and distress. In order to measure the amount of attention paid to the film fragment, participants rated the percentage of time they had looked away from the screen during the most intense part (the murder; 2 min) of the fragment on a 100 mm Visual Analog Scale (0 = not at all - 100 = all of the time). They also rated (VAS; 0 = not at all -100 = extremely) film-related distress. Furthermore, participants were asked if they had seen the stressful fragment before (Yes/No). Prior trauma. The Negative Life Events and Trauma Questionnaire (NLETQ; Morgan & Janoff-Bulman, 1994) was used as a screening instrument to check the frequency of earlier stressful/traumatic life events. These included verbal, physical, and sexual abuse, parental death, serious accidents, serious illness, and natural disasters. The main variable was the number of prior traumatic events with a self-reported intensity above 50 (range: 0-100). Negative mood. In the current study, we used a Dutch translation of the Center for Epidemiologic Studies Depression Scale (CES-D; Bouma, Ranchor, Sanderman & Van Sonderen, 1995). The CES-D contains 20 items addressing depressive symptoms during the past week. Scores for each item range from 0 (rarely or none of the time) to 3 (most or all of the time).Total maximum score on the CES-D is 60. Four items are worded positively and then scored reversed for the total CES-D score. The present study showed that the internal consistency of the CES-D was high (Cronbach’s α = .93).

- 81 -

Chapter 5

2.7. Procedure5 The present study consisted of two sessions separated by one week. Participants were told in advance that they would be presented with a film fragment with emotional content. Upon arriving at the test room, they filled in the FFPI and completed the CVLT. After this, participants were shown with the distressing fragment. They were told that they were free to leave the room and to terminate the experiment at any point. Following the presentation of the fragment, participants completed the film exposure and distress ratings. Following this, the experimenter gave instructions about the diary and the follow-up 7 days later. At followup, participants delivered their diaries, completed the Impact of Movie Scale and filled in the CES-D and the prior trauma screening list.

2.8. Data reduction and statistical analyses In order to normalize distributions, outliers were replaced with a value of three SD’s of the mean for the diary intrusion variable (2 outliers) and the CES-D (2 outliers). After this correction, all of the variables in the present study were within the normal range (skewness and kurtosis approximately between -1 and 1). In order to asses PI on the CVLT, an interference score was calculated as follows. First, a weighted average of shared category members was computed for trial 1 of List 1. This was done because the relative proportion of shared and non-shared category items recalled over two trials may affect the potential for PI during List 2 recall. The number of items recalled on trial 1 was multiplied by the total number of shared items that were recalled on trial 1 and 2; the outcome of this multiplication was divided by the total number of items recalled on trial 1 and 2 resulting in a weighted average score for trial 1 (see Kramer & Delis, 1991). Subsequently, the number of correctly recalled shared category items on list B was subtracted from this weighted average. This interference score could be interpreted as such that higher scores would reflect more interference. Data were analyzed with a series of Hierarchical Multiple Regression (HMR) analyses. Dependent variables were the IMS total score and the number of diary intrusions.

5

Participants also completed 2 measures of attention (Rapid Serial Visual Presentation [RSVP]) which are not reported here.

- 82 -

Predictors were entered in three steps. The first step contained five control variables. To begin with, emotionality ratings of the film-clip and prior trauma exposure (NLETQ) were included. Furthermore, gender was dummy-coded and included because adjusted exploratory t-tests revealed that females scored higher on the IES total score than males, t (69) = -2.42, p < .05. To control for any effects of list order (CVLT), a dummy coded variable for version (AB or BA) was included. Finally, the CES-D was included to control for the effects of a general negative mood in the week after the presentation of the stressful film. The second step contained the CVLT-interference score and the FFPI. In the third step, the product of the normalized CVLT and FFPI scores was included to investigate the interaction between resistance to PI and neuroticism. Additionally, a descriptive analysis of the content and modality of the overall number of diary intrusions was conducted.

3. Results

3.1. Descriptives Descriptive statistics for the variables used in the present study are shown in Table 1. In general, participants rated the film-fragment as moderately stressful and were generally able to prevent looking away during the most threatening moments of the fragment. Furthermore, approximately 62% of the participants had experienced one or more prior traumatic or highly stressful events. Only three participants (3.5%) had seen the film fragment before. Earlier studies (e.g., Kramer & Delis, 1991) that investigated PI with the CVLT showed a between lists decrease in performance for shared trials (normal interference) and a between lists increase in performance increase for unshared trials (release from PI). Overall, the current results were in line with these findings: Recall performance for shared items on list 2 (M = 9.86; SD = 2.56) decreased in comparison with shared items on the first trial of list 1 (M = 10.78; SD = 2.63), t (84) = 3.50, p < .01. On the other hand, recall performance for the non-shared items showed the opposite pattern with better recall performance on list 2 (M = 6.42; SD = 1.70) than on the first trial of list 1 (M = 5.87; SD = 1.46), t (84) = -2.69, p < .01.

- 83 -

Chapter 5

Table 1. Mean, Standard Deviation and Range of the main variables used in the present study. Variable

N

Mean

SD

Min

Max

CVLT shared List1-trial 1 85

10.78

2.63

5

17

CVLT shared List 2

85

9.86

2.56

5

16

CVLT unshared List1- 85

5.87

1.70

2

9

trial 1 CVLT unshared List 2

85

6.42

1.46

3

10

CVLT interference score

85

1.05

2.19

-4.50

6.04

FFPI-neuroticism

85

43.19

9.99

22

79

distress 85

63.09

22.84

0

100

Percent of time looked 85

18.47

31.74

0

100

Film-related (VAS) away IES-total

85

4.55

5.03

0

21

Diary intrusions

85

1.56

1.82

0

7

CES-D

85

7.07

6.93

0

29

Prior trauma

85

1.26

1.53

0

6

CVLT = California Verbal Learning Test; FFPI-neuroticism = Five Factor Personality Inventory: neuroticism scale; IMS-total = Impact of Movie Scale-total score; CES-D = Center for Epidemiologic Studies Depression Scale; Percent of time looked away = percent of time looked away during the most intense moments of the stressful film (2 min); Prior trauma = number of earlier traumatic events with an impact above 50 (scale 0-100). 3.2. Relationship between resistance to PI (CVLT) and Intrusive memories (IMS-total and diary intrusions). As shown in table 2, the first step of the HMR analyses with the IMS-scores as the dependent variable resulted in a borderline significant 11% of explained variance for the five control variables, F (5,79) = 2.04, p = .08. Inclusion of the CVLT interference score and FFPI in the second step resulted in an additional 10% of explained variance in intrusive memories. This result was almost completely based on the significant effect of the CVLT interference score, semi-partial r = .33. p < .05. A relatively low ability to resist interference on the CVLT resulted in a relatively high frequency of film-related intrusive memories in the week after the presentation of the stressful film. The FFPI did not explain any variance in IMS scores. - 84 -

Finally, the inclusion of the interaction term between the CVLT and FFPI in the third step of the model did not result in any extra explained variance in the IMS scores. The HMR analysis with the number of diary intrusions as the dependent variable resulted in a non-significant 5% of explained variance for the control variables in the first step (see Table 2). The second step resulted in a significant contribution in explained variance for the interference score, semi-partial r = .27, p < .05 with no significant effect for the FFPI. The interaction term in the third step of the model did not result in any extra explained variance in the number of diary intrusions.

- 85 -

Chapter 5

Table 2. Summary of Hierarchical Regression Analyses of IMS scores and Dairy intrusions (N = 85) Model 1 IMS scores B

SE

Beta

Prior trauma Gender

0.53 1.28

0.36 1.30

0.16 0.11

Film-related distress

0.02

0.03

0.08

CVLT-list-order CES-D

-1.94 0.01 1.31

1.13 0.08 0.89

-0.19 0.01 0.13

Prior trauma Gender Film-related distress

0.56 1.76 0.03

0.35 1.32 0.03

0.17 0.15 0.14

CVLT-list-order

-3.18

1.16

-0.32**

CES-D FFPI CVLT-interference

-0.07 0.03 0.80

0.09 0.07 0.27

-0.10 0.05 0.35**

0.56 1.78 0.03 -3.18 -0.07 0.03 0.80 -0.07

0.35 1.34 0.03 1.17 0.10 0.07 0.28 0.72

0.17 0.15 0.14 -0.32** -0.10 0.05 0.35** -0.01

B

SE B

Beta

Predictors Ste p 1

2

F-change

11%

5, 79

2.04

0% Prior trauma Gender Film-related distress CVLT-list-order CES-D FFPI CVLT-interference FFPI*CVLT-PIscore Model 2 Diary intrusions Predictors

- 86 -

DF

10%

3

Ste p 1

(∆)R²

2,77

1,76

4.80*

0.12

(∆)R²

DF

F-change

5%

5, 79

0.54

Prior trauma Gender

0.10 0.32

0.14 0.49

0.08 0.07

Film-related distress

0.01

0.01

0.14

CVLT-list-order CES-D

-0.27 -0.02

0.43 0.03

-0.07 -0.08

Prior trauma Gender Film-related distress

0.08 0.27 0.02

0.13 0.49 0.01

0.07 0.06 0.19

CVLT-list-order

-0.63

0.44

-0.17

CES-D FFPI CVLT-PI-score

-0.07 0.05 0.24

-0.04 0.03 0.10

-0.28* 0.25 0.29*

2

3

11%

0%

2, 77

1, 76

5.23**

0.48

Prior trauma 0.08 0.13 0.07 Gender 0.31 0.50 0.07 Film-related distress 0.02 0.01 0.18 CVLT-list-order -0.64 0.44 -0.17 CES-D -0.07 -0.04 -0.26 Neuroticism 0.05 0.03 0.25 CVLT-PI-score 0.22 0.10 0.27* -0.20 0.27 -0.08 FFPI*CVLT-PIscore Note. CVLT = California Verbal Learning Test; FFPI-neuroticism = Five Factor Personality Inventory: neuroticism scale; IMS-total = Impact of Movie Scale-total score; CES-D = Center for Epidemiologic Studies Depression Scale; Prior trauma = number of earlier traumatic events with an impact above 50 (scale 0-100). *p < .05 **p < .01. 3.3. Descriptive analyses of diary intrusions Fifty-two participants (61%) recorded at least one intrusive memory in their diaries. Figure 1 shows how the total number of intrusive memories was distributed over the 7 days following the presentation of the stressful film. Most of the intrusions were experienced during the first day, after which a sharp drop was followed by a more gradual decrease during day 2, 3 and 4. Inspection of the modality of the recorded diary intrusions revealed that 60% percent was experienced as a visual image-based intrusion, 14% was experienced as a feeling or emotion and 20% was experienced as a thought. These results are generally in line with - 87 -

Chapter 5

earlier clinical investigations of the phenomenology and time course of intrusive remembering (Ehlers et al., 2004). Figure 1. Percentages of total intrusions per day distributed over the week following them presentation of the stressful film 45% 40% 35% 30% 25%

Percentage Intrusions

20% 15% 10% 5% 0% Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8

4. Discussion The present study examined whether a habitually poor executive ability existing prior to the experience of a stressful or traumatic event would increase the probability of persistent intrusive images of that stressor afterwards. The idea was that a relatively good pre stressor ability to resist PI in working memory would be crucial for the gradual decrease of intrusive re-experiencing after trauma. The main results can be summarized as follows. First, a relatively low ability to resist interference (as indexed by the CVLT) was associated with/ predicted a relatively high frequency of film-related intrusive memories after the presentation of a stressful film. This finding could not be explained by individual differences in prior trauma or depressive symptoms and was independent of gender and list order effects on the CVLT task. Second, this relationship between resistance to PI and intrusive memories was independent of trait levels of neuroticism. Hence, no evidence was fond for the suggestion that people with a combination of low resistance to PI and high levels of neuroticism would posses an increased risk of persistent intrusions after trauma (e.g., Derryberry & Reed, 2001).

- 88 -

These results point to the prominent role that pre- stressor executive ability might play in explaining post-trauma symptomatology. Third, the current study investigated the frequency and qualities of intrusive re-experiencing in a one week diary. The results showed that the majority of the recorded diary intrusions were experienced in the visual modality. This finding extends the results of Wessel et al. (2008) on intrusive thoughts and concurs with earlier clinical observations involving trauma-related intrusive memories that mainly consisted of brief sensory – mostly visual - fragments of the traumatic experience (Ehlers et al., 2004; Ehlers & Steil, 1995; Ehlers et al., 2002). Furthermore, graphical representation of the film-related intrusions showed a pattern that mirrors the decline in trauma-related intrusions after trauma (Rothbaum et al., 1992). That is, a relatively large part of the intrusions from participants who filled in their diaries (i.e., at least one diary intrusion) were reported within 24 hours after the presentation of the stressful film, after which a sharp decline was followed by a more gradual decrease during the third and fourth day. The present findings replicate and extend earlier prospective results of Wessel et al. (2008) who investigated the relationship between resistance to PI and intrusive thoughts. Using a different paradigm (CVLT) and an independent sample, our conceptual replication provides more consistent evidence for the role of a specific rather than general executive ability in the prediction of intrusive cognition. This specific executive function was the ability to resist PI in working memory. Recent developments in the experimental memory literature stress the importance of investigating separate (instead of general) executive abilities in applied research settings (e.g., response inhibition, resistance to PI; Friedman et al., 2008; Friedman & Miyake, 2004). Failure to specify the precise underlying mechanisms may lead to null- or inconsistent findings in studies investigating the role of executive control in psychopathology. In the context of intrusive cognitions, this might have been the case with studies using working memory capacity (WMC) as a general index of executive control. WMC is defined as the extent to which goal-relevant information is kept in an active state in working memory, despite the interference of irrelevant information (Unsworth & Engle, 2007). There are some analogue studies that have investigated the relationship between WMC and stressor related intrusive cognition. One retrospective study found the predicted negative relationship between WMC and intrusive memories of a stressful personal event (Klein & Boals, 2001). However two recent prospective studies using trauma-films as analogue stressors failed to find the expected association (Nixon et al., 2007; Wessel et al.,

- 89 -

Chapter 5

2008, experiment1). Considering this, the fact that we were able to find a significant relationship between resistance to PI and intrusive cognition in two earlier independent samples (Verwoerd.et al., 2008; Wessel et al., 2008), shows that the selection of indices of specific (instead of general) executive abilities might lead to more consistent results and conclusions in studies investigating the role of executive control in psychopathology. Our findings regarding the role of resistance to PI are generally consistent with previous neuropsychological research comparing PTSD patients and trauma-exposed controls on a wide range of cognitive measures. Several studies which focused specifically on indices of proactive and retroactive interference found recall deficits in PTSD patients when initial learning was taken into account (Uddo, Vasterling, Brailey, & Sutker, 1993; Vasterling, Brailey, Constants, & Sutker, 1998, 2002; Yehuda et al., 1995; see also Vasterling & Brailey, 2005). Moreover, Vasterling et al. (1998) found that the tendency to make intrusion errors on cognitive interference tasks was positively correlated with re-experiencing symptoms. These clinical results may be very informative about underlying pathogenic mechanisms such as resistance to PI. However, it remains unclear whether neuropsychological abnormalities are a consequence of PTSD or represent a pre morbid risk (e.g., Vasterling,2005). Considering this, the current prospective findings support the possibility that these cognitive abnormalities may already exist prior to trauma and may thus function as a pre morbid risk factor for the development and maintenance of PTSD. The present analogue findings may also inform future prospective work under selective groups with an increased risk for trauma exposure (Medical service personnel, soldiers before deployment). That is, cognitive abnormalities may result from a deficit in specific executive ability of resistance to PI which should be taken into account when selecting experimental indices in future clinical studies. There are several issues that deserve further attention. First, the present study used a variant of the trauma-film paradigm to build a laboratory model for individual differences in persistent intrusive re-experiencing (see Holmes & Bourne, 2008 for a review). Studies using analogue stressors and populations are open to the objection that the studied processes are dissimilar to those observed in the aftermath of an actual trauma. Laboratory studies provide the opportunity of investigating the influence of risk factors on intrusions in a prospective design, but at the same time render problems of external validity. Some support for the validity of the current findings can be found in the quality of the recorded diary intrusions, which were generally in line with earlier clinical investigations of intrusive memories after

- 90 -

trauma (Ehlers et al., 2004). However, these laboratory findings surely have to be extended to other methodologies with survivors of actual traumatic events. Second, the current study found no moderating effect of high levels of neuroticism on the prospective relationship between resistance to PI and intrusive memories. This could mean that the combination of a low ability to resist PI from working memory in combination with high levels of neuroticism does not increase the risk for persistent intrusive memories (e.g., Derryberry & Reed, 2001). However, this non-significant interaction may also result from the observation that neuroticism by itself did not have any predictive value on the frequency of the current film-related intrusions. This finding seems to be inconsistent with earlier clinical work showing that high levels of neuroticism would increase the risk of developing PTSD after trauma (e.g., Lauterbach & Vrana, 2001). Possibly, the analogue nature of the present design was responsible for this inconsistency, but it may also be the case that high neuroticism (in contrast with low executive ability) mainly influence intrusionrelated qualities such as distress and experienced control instead of frequency, which was the main dependent variable of the present study. Future analogue and clinical work may further elucidate the precise characteristics of this relationship. Third, the number of recorded diary intrusions in the current study was low. This made it difficult to test the effect of resistance to PI on individual differences in the gradual decrease of intrusive memories during the week following film presentation. In line with theory, one would expect that people with a well-developed ability to resist PI are able to get control over their analogue intrusive experiences within a few days. In contrast, people with a low ability to resist PI would show more persistent intrusive memories distributed over the whole week. Future work may rely on more realistic and intense stimulus material (within ethical limits) that may elicit a larger number of intrusive memories and would resemble reallife trauma more closely (Holmes & Bourne, 2008). In conclusion, the present results indicate that a relatively weak executive ability to resist interference from working memory may be involved in image-based intrusive memories of stressful events. Moreover, our data suggest that this weakened ability already existed prior to the presentation of the stressor. These results support the idea that a pre morbid deficit in the ability to resist PI from working memory may complicate a rapid decline of intrusive memories after trauma and may thus constitute a vulnerability factor for their persistence and subsequent development of PTSD. Future prospective work should test

- 91 -

Chapter 5

if the present relationship would generalize to real traumatic events, including selected groups with an increased risk for trauma-exposure (e.g., soldiers before deployment to a war zone).

- 92 -

Appendix Words used in the CVLT task List A

List B

1

Selderij (Celery) SV

Komkommer (Cucumber) SV

2

Erwten (Peas) SV

Biet (Beet) SV

3

Uien (Onions) SV

Bonen (Beans) SV

4

Rode Kool (Red cabbage) SV

Spinazie (Spinach) SV

5

Paprika (Sweet pepper) SV

Broccoli (Broccoli) SV

6

Asperges (Asparagus) SV

Radijs (Radish) SV

7

Aardappel (Potato) SV

Bloemkool (Cauliflower) SV

8

Courgette (Courgette) SV

Witlof (Chicory) SV

9

Mais (Corn) SV

Aubergine (Eggplant) SV

10

Prei (Leek) SV

Spruiten (Sprout) SV

11

Aap (Monkey) SA

Tijger (Tiger) SA

12

Leeuw (Lion) SA

Vleermuis (Bat) SA

13

Muis (Mouse) SA

Zebra (Zebra) SA

14

Ezel (Donkey) SA

Schaap (Sheep) SA

15

Konijn (Rabbit) SA

Giraffe (Giraffe) Sa

16

Kangoeroe (Kangaroo) SA

Neushoorn (Rhino) SA

17

Nijlpaard (Hippopotamus) SA

Varken (Pig) SA

18

Wolf (Wolf) SA

Hert (Deer) SA

19

Hamster (Hamster) SA

Olifant (Elephant) SA

20

Eekhoorn (Squirrel) SA

IJsbeer (Polar bear) SA

21

Madelief (Daisy) UF

Basgitaar (Bass guitar) UM

22

Paardebloem (Dandelion) UF

Blokfluit (Recorder) UM

23

Lelie (Lily) UF

Triangel (Triangle) UM

24

Iris (Iris) UF

Cello (Cello) UM

25

Geranium (Geranium) UF

Harp (Harp) UM

26

Margriet (Marguerite) UF

Klarinet (Clarinet) UM

27

Hyacint (Hyacinth) UF

Trompet (Trumpet) UM

28

Krokus (Crocus) UF

Accordeon (Accordion) UM

29

Narcis (Daffodil) UF

Hoorn (Horn) UM

30

Orchidee (Orchid) UF

Saxofoon (Saxophone) UM

- 93 -

Chapter 5

SV = Shared category vegetables; SA = Shared category animals; UF = Unshared category flowers; UM = Unshared category musical instruments.

- 94 -

Chapter 6 Preferential Processing of Visual Trauma-Film Reminders Predicts Subsequent Intrusive Memories.

J. Verwoerd, I. Wessel, P.J. de Jong, M.M.W. Nieuwenhuis Cognition and Emotion (in press)

This study used an analogue design to test the hypothesis that preferential processing of visual trauma reminders in the aftermath of a stressful or traumatic event gives rise to subsequent intrusive memories. Shortly after the presentation of a stressful film fragment, participants (n = 36) were asked to detect neutral targets (rotated buildings or nature scenes) in a single target rapid serial visual presentation (RSVP) paradigm. During half of the streams, the target was preceded by a distracter. The distracters consisted of visual images extracted from an earlier presented stressful film (e.g., persons and objects that figured in the film). The degree of interference by these film-reminders predicted subsequent intrusions recorded in a one-week diary. The results provide evidence that a deficient ability to obtain attentional control over perceptual “trauma” reminders during goal-directed behavior may set people at risk for persistent intrusive memories. Implications for research investigating attentional bias and intrusive memories in context of posttraumatic stress disorder (PTSD) are discussed.

- 95 -

Chapter 6

1. Introduction Intrusive memories related to stressful or traumatic events usually consist of unwanted occurrence of sensory fragments of the event, often accompanied with high levels of distress (Hackmann, Ehlers, Speckens & Clark, 2004). It has been found that intrusions in the visual modality are the most common (Ehlers, Hackmann, & Michael, 2004). These unwanted visual images are often associated with PTSD, but are also found to be related with depression (Reynolds & Brewin, 1999). These trans-diagnostic findings would call for pinpointing the mechanism that would be responsible for intrusive memories. A better understanding of the underlying mechanisms might help to explain why some individuals show an increased vulnerability for the maintenance of these memories and related psychopathology (e.g., Rothbaum et al., 1992). Recently, it has been argued that the expression of an implicit perceptual representation system (PRS) in long-term memory (i.e., perceptual priming [Tulving & Schacter, 1990]) might underlie the persistence of intrusive memories (Ehlers et al., 2006; Ehlers & Clark, 2000). This enhanced perceptual priming hypothesis (e.g., Michael & Ehlers, 2007) states that a primary focus on sensory (i.e., visual) details during trauma encoding would result in a relatively strong activation of these features in the PRS. This would in turn give rise to a preferential processing of those environmental stimuli that bear a strong perceptual resemblance to the actual traumatic situation. Prolonged engagement with these trauma-reminders would eventually increase the probability that the trauma memory would emerge in conscious awareness and may thus be responsible for the maintenance of intrusive re-experiencing symptoms over time (e.g., Michael & Ehlers, 2007). From such a perspective, the inability to prevent prolonged engagement with perceptual trauma-reminders might be critically involved in the occurrence and persistence of unwanted reminders. Accordingly, individual differences in the (in)ability to disengage attention from trauma reminders might help answering the critical question of why in some people intrusive memories persist whereas in others the memories readily subside (cf. Pineles, Shipherd, Welch, & Yovel, 2007). Based on the ideas behind the perceptual priming hypothesis, the current study was one of the first in selecting visual stimuli (rather than words) as “trauma” reminders with direct reference to an earlier stressful event. The major aim of this study was to test whether indeed an inability to disengage attention from visual trauma-related stimuli predicts the

- 96 -

maintenance of intrusive memories. We used a “trauma” film fragment as an analogue stressor (e.g., Holmes & Bourne, 2008) and employed stills depicting typical characters and objects taken from the stressful film as visual stimuli in a single target rapid serial visual presentation (RSVP) paradigm. In this task, participants searched for a neutral target embedded in a series of rapidly presented visual images. On particular trials, critical traumafilm-related distracter images (i.e., the stills taken from the film) appeared shortly (+/200ms) before the target (cf. Most, Chun & Widders, 2005). Preferential attention towards these trauma-film related distracter stimuli would interfere with subsequent target identification (cf. Arnell, Killman, & Fijavz, 2007). If an inability to disengage attention from perceptual trauma-reminders contributes to the maintenance of intrusive memories, performance on these RSVP trials would have predictive value for the frequency of subsequently occurring intrusions. Therefore, the present study examined whether RSVP-task performance (i.e., searching for a rotated building) on film trials would predict the frequency of intrusive memories recorded in a one-week diary. We also used a self-report measure of (habitual) attentional control (Derryberry & Reed, 2002). This was done to examine whether preferential processing of film-reminders (i.e., an inability to disengage) would be related to low levels of pre-film attentional control (rather than differences in anxiety/emotionality, see Barnard, Ramponi, & Battye, 2005).

2. Method

2.1. Participants Thirty-nine undergraduate students participated in this study (9 men and 30 women; mean age: 20.54 years, age range: 18-27). All participants gave their written informed consent to take part. The University of Groningen Behavioral Research Ethical Committee granted ethical approval for this study. Participants were paid 10 € for participation.

2.2. Pre-film attentional control A Dutch translation of the Attentional Control Scale (ACS, Derryberry & Reed, 2002; Verwoerd, de Jong & Wessel, 2006) was used. The ACS is a 20-item self-report measure which measures attentional focusing (9 items: “When I need to concentrate and solve a problem, I have trouble focusing my attention” [present sample’s cronbach’s α = .79]) and

- 97 -

Chapter 6

attentional shifting/disengagement (11 items: “When a distracting thought comes to mind, it is easy for me to shift my attention away from it.” [present sample’s cronbach’s α = .67]). The response format of these subscales uses a 4-point scale (1 = almost never; 2 = sometimes; 3 = often; 4 = always), with higher scores indicating better attentional control. Some evidence for good convergent validity of the ACS with a behavioral measure of attentional control was found in an earlier study of Derryberry & Reed (2002), who showed that low scorers on the ACS showed decreased performance on a behavioral index of attentional bias.

2.3. Stressful film fragment Participants were shown a 9-min fragment of “Irreversible” directed by Gaspar Noé. The fragment shows a gruesome and brutal murder in a dark nightclub. The film fragment was projected on a large (150 x 120 cm) screen in a darkened room.

2.4. Intrusive memory measures 2.4.1. Impact of Movie Scale (IMS). A modified Dutch version of the Impact of Event Scale (IES, Horowitz, Wilner & Alvarez, 1979) instructed participants to respond to statements related to the film fragment they had been shown 7 days earlier (e.g., “Images came up spontaneously”). The IMS consisted of 11 items answered on a five-point scale (range: 0 – 55), with assigned item weights of 0, 1, 3, 5. The internal consistency of this scale was found to be high (Cronbach’s α = .90). 2.4.2 .Intrusion diary. Following earlier trauma-film studies (Holmes et al., 2004), participants were asked to use an A5-format diary to record any intrusions of the films during one week following the film. Elaborate verbal and written instructions were given about the nature of unwanted intrusions and how to keep the diary. Intrusions were defined as “spontaneously occurring” (not intentional) memories of the film (see Holmes et al., 2004). Participants were asked to carry the diary with them and to record every intrusive memory related to the film fragment as soon as possible upon their occurrence. The main dependent variable was the total number of intrusions reported in the diary.

- 98 -

2.5. RSVP-task Stimuli6 used in an earlier RSVP study (Most et al., 2005) were color photographs which were drawn from several pools that consisted of 260 upright landscape/architectural scenes, 24 neutral pictures of people and animals, 24 stills from the distressing film, 168 targets (84 architectural photos rotated 90 degrees to the left or right) and 112 filler targets (28 architectural/landscape photos rotated 90 degrees to the left or right). The stimuli (19 cm wide x 19 cm high) were presented on a 75-Hz 19 inch-CRT monitor that was attached to a desktop computer controlled by E-prime running software. The RSVP task consisted of 288 trials. A trial consisted of a stream of 17 images, presented for 120 ms each. Trials were distributed over 4 blocks of 72 trials. Each block contained 24 trials with a neutral distracter, 24 trials with a film distracter and 24 filler trials where no critical distracter appeared. Filler trials were not used for further analyses. The different types of distracters were randomly distributed over each block. The target-stimulus appeared either 2 scenes (lag 2, 240 ms) or 8 scenes (lag 8, 960 ms) after the critical distracter (see figure 1). At the end of each trial, participants were asked to respond with the left and right arrow key to indicate in which direction they thought the target was rotated. Similar to the procedure designed by Most et al. (2005) there were two conditions. In the condition that is critical for the present hypotheses, participants were instructed that the target always consisted of a rotated building (Specific attentional set)7. In the other condition, the target could either be a building or a landscape (Non-specific attentional set). This nonspecific condition was included because Most et al. (2005) found that overall, performance was better in the specific condition (due to the potential contribution of attentional control in this condition). Replication of these earlier findings would attest to the validity of the present experimental procedure. Participants started with a 16-trial training session in order to get used to the procedure. None of the stimuli displayed during the training session were used in the experimental phase.

6

We thank Dr. S. B. Most for kindly providing the original RSVP task and stimuli on which this adaptation was based. 7

For reasons of clarity and space, we have only focused on the results relevant to the present study. A detailed overview of the RSVP-results can be obtained on request from the first author.

- 99 -

Chapter 6

Figure 1. Example of a part of a Lag 2 RSVP trial. Each trial consisted of 17 pictures. The duration of each picture presentation was 120 ms. During a distracter-trial (neutral and film), the target and distracter were embedded in 15 scene (landscape/architectural) pictures. During a filler-trial, no distracter has been presented, but an extra scene picture has been inserted instead.

- 100 -

2.6. Control measures 2.6.1. Film Exposure and distress. In order to measure the amount of self-reported attention paid to the film fragment, participants rated the percentage of time they had looked away from the screen during the most intense part (the murder) of the fragment on a 100 mm Visual Analog Scale (VAS; 0 = not at all - 100 = all of the time). They also rated (VAS; 0 = not at all -100 = extremely) how much distress they experienced while viewing the fragment. Furthermore, participants were asked if they had seen the stressful fragment before (Yes/No). 2.6.2. Neuroticism. Neuroticism was measured with the Five Factor Personality Inventory (FFPI; Hendriks, Hofstee & de Raad, 1999). The emotional stability scale of the FFPI consists of 20 items (e.g., “Is afraid to do something wrong”) answered on a five-point scale (1 = does not at all apply to me, 5 = does very much apply to me). The internal consistency of the emotional stability scale used in this study was high (cronbach’s α = .93). 2.7. Procedure Participants were told in advance that they would see a film fragment with possible distressing contents. Upon arrival at the test room, participants filled in the ACS and FFPI and were presented with the distressing fragment. They were told that they were free to leave the room and to terminate the experiment at any point. Following the presentation of the fragment, participants completed the film exposure and distress ratings. Subsequently, they received verbal instructions about the RSVP task and were asked to complete the task. Following this, the experimenter gave instructions about the diary and the follow-up 7 days later. At follow-up, participants delivered their diaries and completed the IMS. Before leaving, the experimenter debriefed the participants and rewarded them with a monetary compensation.

2.8. Data reduction and statistical analyses Due to positively skewed distributions of the IMS and the number of intrusions recorded in the diary, square-root transformations were conducted on both variables. These transformed distributions approached normality (i.e., skewness and kurtosis between -1 and 1). All relevant analyses are based on these transformed variables but the table presents the untransformed values for interpretation purposes. Three participants were excluded from

- 101 -

Chapter 6

further analyses because they did not rate the film fragment as at least moderately stressful (VAS distress < 50). In order to see whether an inability to disengage attention from visual film-reminders predicts subsequent (diary) intrusions, the lag 2 difference score between incorrect answers on film and neutral distracter trials was calculated for the trials with a specific target (Most et al., 2005). Because neutral stimuli were subtracted from film stimuli, higher scores reflect more difficulty in controlling film reminders. Earlier work (e.g., Most et al., 2005; Arnell et al., 2007) showed that when a distracter stimulus is presented shortly before target appearance (lag 2), target detection will be hampered. This effect was significantly reduced with a longer distance between the distracter and the target (lag 8). In line with these earlier findings, we only used the lag 8 trials as a manipulation check for measuring impaired target detection on lag 2. The main hypotheses were tested by using hierarchical multiple regression (HMR) analyses with the IMS at follow up and the number of diary intrusions as dependent variables. In the first step of these analyses the two control variables of neuroticism (FFPI) and prior film exposure were included. Apart from the selection of participants on their film-related distress (a state measure), we included neuroticism (a trait measure) to control for any effects that increased levels of anxiety (state and trait) might have on the processing of film-related distracters on the RSVP (see Barnard et al. 2005). The second step consisted of the inclusion of the RSVP difference score to investigate the additional contribution of film-related disengagement problems in the explanation of intrusive memories. Subsequently, these analyses were repeated with self-reported attentional control as an additional control variable to investigate its effect on the predictive power of the RSVP difference score. It was expected that the inclusion of attentional control would result in a reduction in explained variance for the RSVP in comparison with the former analyses without attentional control. As the selfreport index of habitual attentional control does not fully represent the hypothesized role of attentional disengagement in RSVP target detection, we did not expect a complete drop in explained variance in this analysis. A final HMR analysis investigated the direct association between attentional control and RSVP performance. After neuroticism and prior film exposure were included in the first step, the additional variance in RSVP performance related to attentional control was tested in the second step by the inclusion of the ACS subscales.

- 102 -

3. Results

3.1. Descriptives Descriptive statistics for the variables used in the present study are shown in Table 1. The remaining 36 participants rated the fragment as highly distressing and were generally able to prevent looking away during the most threatening moments of the fragment. Furthermore, 9 (22%) participants had seen the film fragment before. Inspection of the relevant scatter plots revealed that the current results did not depend on the effects of a few outliers. As the focus subscale of the ACS did not correlate with any of the dependent variables, analyses with this subscale will not be further reported (see Table 2). Table 1. Mean, Standard Deviation and Range of the main variables used in the present study. Variable

N

Mean

SD

Min

Max

ACS_shifting/disengagement 36 28.75

3.58

19

35

ACS_focus

36 21.28

3.91

15

29

FFPI-neuroticism

36 46.13

9.86

30

68

IMS-total

36 7.50

7.87

0

34

Diary intrusions

36 2.00

2.70

0

14

Film-related distress (VAS)

36 80.95

10.92

57

98

Percent of time looked away

36 27.80

37.20

0

100

Specific Film trials lag 2

36 .80 (3%)

1.14

0

5

Specific Neutral trials lag 2

36 1.03 (4%)

.92

0

4

ACS_shifting/disengagement and ACS_focus = Shifting/disengagement and focus subscales of the Attentional Control Scale (ACS); IMS = Impact of Movie Scale; Percent of time looked away = percent of time looked away during the worst moments of the film fragment; Specific Film trials lag 2 = mean number of incorrect target detections on the lag 2 specific film trials; Specific Neutral trials lag 2 = mean number of incorrect target detections on the lag 2 specific neutral trials. 3.2. RSVP-task Accurate responses on the RSVP task consisted of the correct identification of the rotation direction (i.e., left, right) of the targets. To see whether our findings would replicate the earlier results of Most et al. (2005), incorrect responses (i.e., target misses) were entered

- 103 -

Chapter 6

into a 2 (specificity: specific vs. non-specific attentional set) Х 2 (lag: 2 vs. 8) Х 2 (distracter type: film vs. neutral) repeated measures analysis of variance with three within-subject variables. In line with Most and colleague’s (2005), there was both a main effect of lag (errors lag 2 > error lag 8), F(1,38) = 31.90, p < .001, partial η² = .46 and of specificity (errors specific trials < errors non-specific trials), F(1,38) = 21.47, p < .001, partial η² = .36. In addition, the interaction between lag and specificity reached significance, F(1,38) = 18.31, p < .001, partial η² = .33. At lag 2, participants showed a relative increase in performance levels (i.e., fewer errors) on specific trials when compared with the non-specific trials. This difference in performance disappeared at lag 8. In apparent contrast to the earlier findings of Most et al. (2005) we did not find main and interaction effects with distracter (i.e., errors film > errors neutral).

3.3. Prospective relationship between preferential processing of perceptual film-reminders and subsequent intrusive memories As shown in table 2, the first step of the HMR analysis with the transformed IMSscores as the dependent variable resulted in a significant 18% of explained variance for the two control variables. Inspection of the individual predictors showed a significant contribution of neuroticism, semi-partial r = .41, p < .05. In line with predictions, the inclusion of the RSVP difference score in the second step resulted in an additional 22% of explained variance, semi-partial r = .52, p < .01. The HMR analysis with the (transformed) number of diary intrusions as the dependent variable showed a slightly different pattern. In this analysis, the three control variables did not explain significant variance in diary intrusions. However, when the RSVP difference score was included in the second step, an additional 27% of explained variance was added to the model, semi-partial r = .55, p < .05. 3.4. The role of pre-film attentional control on intrusive memories and RSVP performance As shown in Table 2, the HMR analysis with the IMS as dependent variable showed a non-significant trend of attentional shifting/disengagement in the second step, semi-partial r = -.28, p = .08 The inclusion of the RSVP difference score in the third step resulted in an additional 15% of explained variance (first analysis: 22%), semi-partial r = .45, p < .01. An almost similar pattern of results was found with the number of diary intrusions as dependent variable: this time the second step showed a significant effect of shifting/disengagement, semi-partial r = -.38, p < .05. In the third step, the RSVP difference score added another 16%

- 104 -

(first analysis: 27%) of explained variance to the model, semi-partial r = .48, p < .01. Note that overall, the addition of self-reported ACS resulted in a substantial reduction in explained variance (7 % and 11% for IMS and diary, respectively) compared to the initial analyses. Finally, in the HMR analysis with the RSVP difference score as dependent variable, the first step with neuroticism and prior film exposure failed to reach significance, R² = .13, p = .10 (see Table 2). However, the addition of the shifting/disengagement scale in the second step showed a significant relationship with the strength of film-related interference on the RSVP, semi-partial r = -.37, p < .05. Participants relatively low in self-reported attentional control showed relatively strong film-related interference on the RSVP.

- 105 -

Table 2. Summary of Hierarchical Regression Analyses of IMS scores (with and without ACS_shifting), Diary intrusions (with and without ACS_shifting) and the RSVP difference score (N = 36) Dependent Predictors (∆)R DF F-change B SE Beta ² IMS scores Step 1 Prior film exposure Neuroticism

.18

Step 2 Lag 2 score

.22

2, 33

1,32

3.55* -0.57 0.06

0.52 0.02

-0.17 0.40*

0.72

0.21

0.50**

-0.06 0.01

0.33 0.02

-0.03 0.16

0.47

0.13

0.55**

-0.57 0.06

0.52 0.02

-0.17 0.40*

-0.12

0.07

-0.28

11.64**

difference

Diary intrusions Step 1 Prior film exposure Neuroticism

.02

Step 2

.27

Lag 2 score

2, 33

1,32

0.41

12.01**

difference

IMS scores Step 1 Prior film exposure Neuroticism Step 2 ACS_Shifting 106

.18

.08

2, 33

1,32

3.55*

3.12

Step 3 Lag 2 score

.16

1,31

8.20**

difference

0.66

0.23

0.46**

-0.06 0.01

0.33 0.02

-0.03 0.16

-0.10

0.04

-0.40*

0.39

0.14

0.46**

Diary intrusions Step 1 .02 Prior film exposure Neuroticism Step 2 .15 ACS_Shifting Step 3 .16 Lag 2 difference score Lag 2 score

2, 33

1,32 1,31

0.41

5.72* 7.50**

difference

Step 1 .13 2, 33 2.47 Prior film exposure -0.82 0.38 -0.36* Neuroticism 0.01 0.01 0.10 Step 2 .12 1,32 5.19* ACS_Shifting -0.11 0.05 -0.36* Note. ACS = Attentional Control Scale; IMS = Impact of Movie Scale; Lag 2 difference score = number of errors in target detection for RSVP film trials minus number of errors in target detection for RSVP neutral trials. *p < .05 **p < .01.

107

4. Discussion The main finding of the current study showed that preferential processing of perceptual reminders of an earlier presented trauma-film predicted subsequent intrusive memories. In addition, this distracting property of the film-reminders was associated with habitual self-reported attentional control. This pattern of findings is consistent with an earlier trauma-film-study showing that low self-reported attentional control has predictive validity for the occurrence of subsequent diary intrusions (Verwoerd, de Jong & Wessel, in press). The present findings raise the question of how attentional control might contribute to the reduction of real-life trauma memories. One explanation might be that lack of attentional control would result in a prolonged engagement with trauma-reminders, resulting in an increased probability of persistent triggering of intrusive memories. This would be in accordance with earlier clinical findings that PTSD patients have specific problems with attentional disengagement (Pineles et al., 2007). Individual differences in pre-trauma cognitive (i.e., attentional) control might further help explaining individual differences in this increased engagement with perceptual trauma-reminders and thereby in the (non)persistence of intrusive memories. That is, a general, pre-morbid deficiency in executive attention might set people at risk for developing persistent pathology (i.e., attentional bias, intrusive memories) in the aftermath of an actual traumatic event. Accordingly, it has been shown that pre-trauma cognitive control as measured by an AB-AC list-learning paradigm was associated with a relatively frequent occurrence of subsequent intrusive cognitions related to a stressful film (Wessel, Overwijk, Verwoerd & de Vrieze, 2008). The current associations of pre-film self-reported attentional control with biased processing of film reminders and subsequent intrusive memories further supports the role of individual differences in premorbid cognitive control in the prediction of post-trauma symptomatology (see also Verwoerd, de Jong & Wessel, in press). However, future studies should decide whether the relationship between pre-trauma executive ability and the frequency of subsequent intrusive memories is indeed mediated by the (in)ability to disengage attention from perceptual “trauma” reminders. The current results on attentional control and intrusive memories are in line with the notion that perceptual priming explains the triggering of visual intrusions (Micheal & Ehlers, 2007; Ehlers et al., 2006). The theoretical framework behind the perceptual priming hypothesis emphasizes a strong link between the encoding and retrieval context of the trauma memory. That is, a primary focus on sensory (i.e., visual) details during trauma encoding would result in enhanced priming (biased processing) for retrieval cues with a direct 108

perceptual resemblance to the earlier trauma context. In this light, testing hypotheses about preferential processing in the post-trauma environment would be best accomplished with the use of perceptual stimuli. Note however, that earlier studies measuring attentional bias in clinical populations mainly used words as trauma-related stimuli (see Buckley, Blanchard, & Neill, 2000 and Williams, Mathews & MacLeod, 1996 for reviews). Following this, the present design has a clear advantage over other methodologies testing biased processing of trauma-related material. First, the selection of visual film-distracters with direct reference to the earlier presented stressful film-fragment would more closely resemble visual reminders in the external post-trauma environment. Furthermore, the use of a single target RSVP (cf. Most et al., 2005) as a measure of the (in)ability to disengage attention from “trauma” reminders demonstrated that preferential attention towards emotional stimuli could disrupt goal-directed behavior of detecting neutral targets. This process is difficult to capture with a response-time measure such as the modified Stroop-task (e.g., Williams et al., 1996) in which the target and distracter share the same dimension (cf. Most et al., 2005). A first limitation of the present study was the use of an analogue design. Studies using analogue stressors and populations are open to the objection that the studied processes are dissimilar to those observed in the aftermath of an actual trauma. Therefore, these laboratory findings surely have to be extended to other methodologies with survivors of actual traumatic events. Some support for the validity of the current findings can be found in earlier traumafilm studies which have shown to be a valid tool for studying peri- and posttraumatic processing (see Holmes and Bourne, 2008 for a review). In line with this, if only quantitative differences exist between analogue and traumatic symptoms, the present effects could as likely be more (rather than less) enhanced for real-life trauma relative to the current analogue trauma. Secondly, the error rate on the RSVP was quite low which could indicate the existence of floor effects on this measure. Possibly, rotated buildings were generally rather easy to detect in a stream of mainly landscape scenes, irrespective of the presence of a distracter stimulus. Thus, the generally low detection threshold for the rotated target buildings may have attenuated the disruptive effect of the presence of a distracter stimulus. Future studies may increase task difficulty by including targets that would “pop up” less easily from the background (landscape) scenes which may enhance the sensitivity of the task as an index of attentional control.

109

Chapter 6

Third, although the present findings provide a case for the role of deficient attentional control in explaining the present relationship between RSVP performance and intrusive memories, it should be acknowledged that the correlational nature of the current study does not rule out alternative explanations. Therefore, future experimental research should manipulate attentional control to further test the validity of “trauma”-related disengagement problems in explaining persistent intrusive memories Finally, in contrast with earlier studies that used distracters with intrinsic negativity (e.g., mutilated faces, Most et al., 2005; Smith et al., 2006), the present study did not find a main effect of visual distracters related to an earlier stressful film. Although this null finding might seem puzzling at first sight, it has to be noted that we selected stills from highly specific scenes of a stressful film fragment. This will make it almost by definition very difficult to find neutral control stimuli that would match the film stills on a wide range of dimensions (e.g., a systematic difference in perceptual quality was still possible). Therefore, we only used the neutral distracter stimuli as a more general baseline for the construction of our difference score. That is, the lag 2 difference score measured the relative difficulty in the detection of targets after film distracters. In our study, no specific predictions were made about the absolute difference in target misses between film and neutral distracters. In conclusion, the present study provided evidence to suggest that an inability to disengage attention from perceptual trauma-reminders contributes to the frequency of intrusive memories. Future work might examine how pre-trauma cognitive control may influence the degree of preferential processing (i.e., prolonged engagement with) perceptual trauma-reminders. Flexible control over trauma-reminders in the external environment may support goal-directed behavior thereby reducing the probability of re-experiencing unwanted intrusive

110

memories.

Chapter 7

Discussion

111

Chapter 7

The central aim of this dissertation was to examine the prospective relationship between inefficient executive control and persistent intrusive memories. The current research questions are based on the clinical observation that during natural recovery after trauma, most people experience some form of reliving of the earlier traumatic event, but only a minority of them will show persistent intrusive memories over time (e.g., Rothbaum et al., 1992). This observation raises the question of what mechanisms might underlie individual differences in vulnerability/resilience. Considering this, the empirical chapters described in this thesis were devoted to answering this crucial question by examining the predisposing role of executive control. More specifically, it was explored whether persistent intrusive memories may be the result of innate or early acquired individual differences in the specific executive ability of resisting PI in working memory (e.g., cognitive inhibition). The choice for the potential involvement of this specific executive ability was based on recent developments in the experimental memory literature. That is, it has been argued that executive control may not be conceptualized as a unitary construct, but that it consists of a set of related but separable executive functions, each devoted to a specific aspect of goal-directed behavior (e.g., Hedden & Yoon, 2006; Friedman & Miyake, 2004; Friedman et al., 2008). The executive ability of resistance to PI was most likely to be involved in gradually decreasing interference of stressful intrusive images in the aftermath of a traumatic event. The analogue methodology of the trauma-film paradigm (e.g., Holmes & Bourne, 2008) presented in chapter 4, 5 and 6 made it possible to investigate the potential involvement of pre-stressor executive control in a fully prospective design. In earlier clinical studies, this idea of pre-trauma vulnerability could only be tested indirectly, by comparing PTSD patients and trauma-exposed controls (e.g., Vasterling & Brailey, 2005) or by retrospectively collecting archival data about more general ability measures such as IQ (Macklin et al., 1998). Thus, positive outcomes of such studies do not address the question whether specific neuropsychological abnormalities are a consequence of PTSD or represent a risk/resilience factor for PTSD (cf. Vasterling & Brailey, 2005). Considering this, the prospective studies in the current thesis were intended to collect preliminary evidence to resolve the issue about the direction of the relationship for the specific symptom of intrusive memories. A second theme of this dissertation was to examine - from a cognitive perspective factors contributing to the direct retrieval of intrusive memories. As noted in the introduction,

112

increased knowledge of the mechanisms supporting involuntary retrieval of intrusive memories may lead to a better understanding of how risk factors such as executive control may contribute to the persistence of these memories. In earlier studies, it was suggested that enhanced perceptual priming (e.g., Michael & Ehlers, 2007; see also Ehlers & Clark, 2000) for stimuli reminiscent of the earlier traumatic event may be responsible for the (early) preferential processing of trauma reminders and subsequent retrieval of intrusive memories. Generally in line with this earlier work, our model as presented in chapter 1 also emphasizes the perceptual nature of trauma reminders as potential triggers of intrusive memories. Moreover, we proposed an additional role for the involvement of attentional control. That is, it was hypothesized that an inability to disengage attention from perceptual trauma-reminders might be crucially involved in the development of persistent intrusive memories after trauma. In this final chapter, I will provide a summary of the empirical studies presented in chapter 2 to 6, and discuss the results in terms of the model presented in chapter 1. Furthermore, reflections on future work addressing untested aspects of the model are provided. Subsequently, some methodological issues are discussed. I will conclude with clinical implications.

7.1. Summary of empirical findings Prior research has shown that image-based intrusive memories usually found in response to trauma also occur on a regular basis in the normal non-traumatic population (Bywaters et al., 2004). These earlier findings were confirmed in the first analogue study presented in chapter 2. The results of this cross-sectional examination showed a positive association between selfreported distractibility and the frequency of intrusive and avoidance symptoms related to a stressful life event. Furthermore, it was shown that the relationship between distractibility and intrusive memories was still present after controlling for differences in depressive feelings, indicating unique explanatory properties for cognitive ability. Interestingly, we also found a main effect (but no significant interaction with distractibility) for repressive coping. That is, people with a repressive coping style reported fewer recent intrusive memories. This indicates that both a well-developed cognitive ability as well as a motivational tendency to avoid negative affect may protect people from experiencing persistent intrusive memories related to stressful life events.

113

Chapter 7

The cross-sectional study presented in chapter 3 used objective performance-based measures of resistance to PI and response inhibition to explore their relationship with intrusive memories related to a stressful life event. The results were in line with our predictions. That is, after correcting for possible confounding effects of depression and age of the reported stressful life event, a unique relationship between intrusive memories and the ability to resist PI appeared. Moreover, this relationship was especially pronounced in the group of female participants. By contrast, this was not the case for the executive ability of response inhibition. These results are in line with the idea that separate executive abilities may underlie different aspects of (abnormal) behavior (e.g., Nigg, 2000). For example, it has been suggested that a lack of response inhibition might be involved in dysfunctional impulsive behavior (e.g., ADHD; Nigg, 2000; Pennington & Ozonof, 1996). In chapter 3, we also proposed that resistance or inhibition of unwanted material from working memory might have consequences for the accessibility of that unwanted material at a later point in time. It was hypothesized that an inefficient ability to resist PI would result in a relatively increased accessibility for earlier interfering material at a later point in time. More importantly, it was hypothesized that this increased accessibility would be related to a relatively high frequency of intrusive memories of an earlier experienced stressful life event. Inspection of the results showed that this last hypothesis could not be confirmed. No relationship between an index of memory accessibility (i.e., reaction times) and intrusive memories emerged. The study reported in chapter 4 was intended as a first step in exploring the prospective relationship between self-report executive control and intrusive memories. In this study, a trauma film was used as an analogue stressor for eliciting intrusive memories. We further explored whether the instruction to keep an intrusion diary would trigger unforeseen recording strategies. These potential strategies may interfere with the natural course of experiencing intrusive memories. Results showed that for participants who were asked to fill in a diary, a negative relationship between self-reported executive control and intrusive memories at a follow-up session emerged. By contrast, in the group of control participants, no evidence was found for a prospective relationship between executive control and intrusive memories. These results provide preliminary evidence for the idea of inefficient executive control as a pre-trauma vulnerability factor for persistent intrusive memories. Furthermore, the results of our diary manipulation suggest that the instruction to fill in a diary can be a valid method for recording spontaneous intrusive memories related to a stressful film.

114

However, the small sample size of this study (n= 17 in the diary group) calls for careful interpretation of the results. The design of the prospective study presented in chapter 5 integrated several ideas tested in the foregoing chapters. The main goal was to examine whether individual differences in pre-stressor resistance to PI would predict the frequency of image-based intrusive memories one week after the experience of a stressful event (a film fragment). The results were in line with these expectations. That is, low performance on the index of resistance to PI was related to a high frequency of diary intrusions and intrusive memories at a one-week follow-up session. Moreover, results showed that resistance to PI was still able to predict intrusive memories over and above prior trauma exposure, depression and impact of the stressful event (film). These findings provide some evidence for the idea that the specific executive ability of resistance to PI may function as an independent pre-morbid risk factor for persistent intrusive memories after trauma. Important to note, inspection of the modalities of the diary intrusions showed that the majority of the recorded memories were experienced as visual intrusions. This is in line with earlier clinical observation regarding the modality of trauma-related intrusive memories (e.g., Ehlers et al., 2004), thereby providing evidence for the external validity of the diary method. The main goal of the study presented in chapter 6 was to explore cognitive mechanisms underlying the retrieval of intrusive memories after a stressful or traumatic event. Based on earlier work on the role of enhanced perceptual priming in explaining intrusive re-experiencing (e.g., Michael & Ehlers, 2007), we proposed an additional role for attentional control in predicting persistent intrusive memories. It was hypothesized that a relative inability to disengage attention from perceptual “trauma”-reminders in an RSVP paradigm would be associated with a relatively high frequency of subsequent intrusive memories. Results showed a pattern of findings in line with these expectations. That is, performance on the RSVP task significantly predicted the frequency of subsequent intrusions. To further explore whether this relationship would indeed reflect genuine differences in the (in)ability to disengage attention, several other variables were included in the analyses. First, self-report indices of film-related distress and neuroticism were included to control for any effects of increased levels of anxiety (state and trait) on the processing of film-related distracters on the RSVP (see Barnard et al. 2005). Furthermore, a pre-film self-report index of attentional control was included (see chapter 4) to more directly test the involvement of

115

Chapter 7

attentional control in RSVP performance. Inspection of the data showed that the relationship between film-related interference on the RSVP and the frequency of diary intrusions remained significant after controlling for effects of neuroticism and film-related distress. Moreover, a significant association between pre-film self-reported attentional control and RSVP target detection emerged, suggesting that individual differences in the disengagement component of attention played a role in the prevention of processing film-related distracter stimuli.

7.2. Discussion of the results in terms of the intrusion model In this paragraph I will return to the model introduced at the end of chapter 1. Based on an integration of ideas presented in the experimental memory literature and the clinical literature on PTSD, the model provides a schematic overview of underlying mechanisms/processes involved in persistent intrusive memories after trauma. More specifically, it was intended to explain how pre-trauma resistance to PI as a specific executive function may contribute to persistent intrusive memories or natural recovery after trauma. The empirical chapters of this thesis were devoted to test the tenability of several components of this model. First, the results of the studies presented in chapter 2 , 3 (retrospective) and 4 (prospective) were used as steppingstone for the design of the study presented in chapter 5. This study was used to test the supposed prospective link between resistance to PI and SAMbased intrusive memories. The results supported the model’s assumptions and were in line with an earlier prospective investigation of Wessel et al. (2008). That is, a relatively low pre“trauma” ability to resist PI in working memory was associated with a relatively high frequency of SAM-based intrusions (i.e., large dark oval in the model p.15). These analogue findings present important information about the potential role of low resistance to PI as a cause (instead of a consequence) of persistent intrusive memories after an actual traumatic event and may therefore help explaining unresolved issues in earlier clinical studies. As mentioned in the introduction, Vasterling et al. (1998) found a positive relationship between performance on indices of interference control (e.g., commission errors) and re-experiencing symptoms in a group of PTSD patients (see also Vasterling & Brailey, 2005). The study of Vasterling and coworkers (1998) presents evidence for the idea that deficient executive control plays a role in the persistence of intrusive memories. However, it remains undecided whether this deficit was already present prior to the traumatic situation.

116

Second, further inspection of the model shows a strong link (thick black arrow) between perceptual reminders of earlier trauma and the frequency of SAM-based intrusive memories. This link emphasizes the potential importance of biased processing of perceptual trauma reminders in the prediction of image-based intrusive memories. The idea is that prolonged engagement with these reminders would be associated with a high frequency of intrusive memories (large black oval). The results of the analogue RSVP study presented in chapter 6 supported this part of the model. That is, the degree of film-related interference on the RSVP task was related to the frequency of intrusive memories. Furthermore, the majority of these memories were experienced in the visual modality, supporting the idea that intrusive memories are indeed mainly image-based. Together, these findings support the view that an inability to disengage attention from perceptual trauma reminders may set people at risk for the persistent experience of SAM-based intrusive memories after real trauma. Considering the foregoing discussion of results in terms of the intrusion model, it will be clear that several questions remain unanswered. For example, it would be an interesting avenue for future research to further explore whether individual differences in the (prolonged) engagement with perceptual trauma reminders would depend on pre-trauma resistance to PI. One of the most important proposals of the model is that the established association between pre-trauma resistance to PI and SAM-based intrusive memories (see chapter 5) will be (partly) mediated by an individual’s ability to disengage attention from perceptual trauma reminders. This proposal is depicted in the model by an inhibitory connection between resistance to PI and the association between the perceptual cue and SAM-based memories. Future analogue studies may explore this more complex relationship by testing whether film-related interference on the RSVP task would mediate the relationship between performance-based indices of pre-film resistance to PI and SAM-based intrusive memories. The model further shows that pre-trauma resistance to PI may also support the formation of newly acquired VAM association to the trauma cue. This idea was based on Dual Representation Theory (DRT; Brewin et al., 1996; Brewin, 2001) of PTSD which stated that the formation of verbally accessible association to trauma cues would be crucial for recovery after trauma. This ability to form new VAM-based associations after trauma exposure can be expressed by the degree of coherence and or fragmentation of the trauma narrative (cf. Brewin et al., 1996). Thus, a first step to investigate this part of the model

117

Chapter 7

would be to relate performance on indices of (pre-trauma) resistance to PI with subjective ratings of the coherence and fragmentation of the trauma narrative. This will probably require the use of clinical groups because the degree of memory fragmentation after viewing a stressful film might be very low and not useful for exploring individual differences. Followup studies may then further explore whether a more detailed and coherent trauma narrative may be predicted by an individual’s ability to disengage attention from perceptual trauma reminders.

7.3. Methodological issues 7.3.1. Use of analogue groups and trauma films The empirical studies presented in chapter 2 to 6 used student samples to investigate the relationship between inefficient executive control and the (sub-)clinical phenomenon of intrusive memories. Studying the natural occurrence of intrusive memories related to stressful life events or trauma films in a sample of healthy undergraduates may result in floor effects and a restriction of range on self-report measures of individual differences. Similarly, the use of student samples with above average scores on measures of intelligence may result in ceiling effects and restriction of range on indices of executive ability. Although the mean frequency of intrusive memories was generally low in the studies using the trauma-film paradigm, the range of these intrusions, together with the intrusions reported in two retrospective studies presented in chapter 2 and 3 was substantial enough to detect the expected correlations with executive ability. These findings concur with earlier observations that image-based intrusions also occur on a regular basis in the normal nontraumatic population (Bywaters et al., 2004). Regarding individual differences in executive control, the modified versions of the interference tasks used in this dissertation (i.e., AB-ACAB list-learning, CVLT) were difficult enough to create a considerable amount of variance on these measures. Taken together, after inspection of the relevant distributions, there is no reason to believe that methodological artifacts such as restriction of range could have reduced the sensitivity to find significant correlations between cognitive ability and intrusive memories. However, the generally low frequency of elicited diary intrusions in the present studies (chapter 5: m = 1.56; chapter 6: m = 2.00) might cause problems when designing future experimental studies. That is, follow-up studies may want to explore protective effects of certain personality traits. Such studies are designed to show a reduction in the frequency of

118

intrusive memories in a group of people selected on the personality trait in comparison with a non-selected control group. The presently used trauma-film may not be sufficiently intense to elicit a considerable amount of intrusions, needed for investigating difference between groups in future experimental work. A more general point of concern for working with first-year undergraduates is that these students are often obliged to participate in a large number of experimental studies. This may potentially result in a reduction of their motivation to successfully complete every study to the best of their ability. This may be especially problematic in less structured assignments as recording naturally occurring intrusions in a one-week diary. One way to circumvent this problem might be to ask participants to fill in a confidential questionnaire at follow-up with questions about motivation during the whole procedure. Of course, it must be clear to them that their responses would have no consequences for payment or study credits. Participants with low motivation during one or more parts of the experiment could then be excluded from further analyses. A better, but more time-consuming and expensive procedure would be using community samples in future studies. These samples are often better motivated to participate in experimental psychology studies and do often better represent the general population than a sample of university students.

7.3.2. RSVP task In chapter six, we found a significant relationship between film-related interference on the RSVP-task and subsequent (diary) intrusions. Based on earlier RSVP studies (e.g., Most et al., 2005; Smith et al., 2006) and on the significant correlation with an index of self-report attentional control we proposed that good performance on the RSVP task would depend on an individual’s ability to disengage attention from visual film reminders. Although this proposal seems plausible, it has to be noted that the correlational nature of the study presented in chapter six did not definitely rule out alternative explanations. For example, because it was very difficult to match our film distracters with appropriate neutral control stimuli, it is possible that the significant relationship between interference for film distracters and intrusive memories was unrelated to the earlier presentation of a stressful film. That is, the significant correlation with intrusive memories would not depend on film distracters as genuine reminders with acquired emotionality, but on particular aspects (e.g. brightness, color) of these distracters. Without a control condition in which half of the participants would

119

Chapter 7

complete the RSVP task before the stressful film, it is impossible to make strong conclusions about film-related interference and its relationship with intrusive memories. Furthermore, the involvement of attentional control on RSVP performance was not tested experimentally which makes alternative explanations possible for the correlation between film-related interference and intrusive memories. Future studies should manipulate attentional control by comparing RSVP performance in an experimental dual-task condition (e.g., counting backwards in threes during target detection) with performance in a neutral condition without a processing manipulation. If the ability to disengage attention from negative distracters plays a crucial role in the detection of target stimuli on the RSVP, performance in the dual task condition should be worse in comparison with the neutral condition. Another interesting research avenue for future studies is to focus more specifically on the time course of distracter-target interference on the RSVP. Earlier studies compared lags of 200 ms (lag 2) and 800 ms (lag 8) and found interference effects on the shorter lag 2 (errors in target detection after negative distracters > errors in target detection after neutral distracters) and no effect of the presence of a distracter on the longer lag 8. With the selection of better matched neutral distracters in future studies (which is difficult but not impossible), it would be interesting to compare different time windows (e.g. lags of 100, 200, 400 and 600 ms) for measuring film-related interference on the RSVP. In line with earlier findings in studies that used paradigms measuring spatial attention (e.g., dot probe; Bar-Haim et al., 2007), interference effects would also be expected at larger time windows of 400 and 600 ms. Of course, the interest in exploring different time windows on the RSVP also holds for the correlation with intrusive memories, which may also depend on a specific time window between distracter and target.

7.4. Future directions Based on the predictions made by our model, future studies should test whether a premorbid deficit in the ability to resist interference in working memory is also involved in the prediction of intrusive memories after real trauma. For example, a prospective study of soldiers before deployment to a war zone would provide a fruitful research direction for further examining resistance to PI as a vulnerability factor for persistent intrusive memories.

120

In an ideal study, soldiers are tested shortly before deployment and than followed and tested on their symptoms for at least one year after returning from their mission. Furthermore, a commonly held view in the clinical literature is that in the immediate wake of trauma, repeated exposure to the stressful trauma memory may support emotional processing and eventually, recovery. Based on this assumption, early intervention programs have been developed. The aim of these psychological debriefing programs is to focus on the traumatic event and to educate people about post-traumatic stress. Although talking about the traumatic experience might seem helpful at first sight, outcome studies have come up with mixed and, in some respect even disappointing results. That is, some studies show no differences between debriefing programs and no intervention at all, while other studies even suggest that such programs could work counterproductive because they would intensify the very symptoms they try to prevent (Mayou, Ehlers, & Hobbs, 2000). Thus, it is possible that early exposure to trauma memories via debriefing procedures interferes with the normal cognitive processing leading to recovery. In line with this, in a recent monograph, McNally, Bryant, and Ehlers (2003) suggested that massive debriefing programs may underestimate the psychological resilience of trauma survivors and that it would be more helpful to focus attention on individuals whose symptoms have failed to abate within the first weeks posttrauma. These persons may be at the highest risk for developing psychopathology such as PTSD. In the current dissertation, I propose that people with a well-developed executive ability would not need an intervention to support the formation of healthy associations to trauma reminders. Executive control and more specifically resistance to PI may help traumaexposed people to disengage attention from the threatening aspects/associations of trauma reminders which may help them with the formation of integrated verbally accessible associations (see the model presented in chapter 1). Building on these last conclusions, a fruitful direction for future research would be to investigate the possibility to train executive control in people with an increased risk of persistent intrusive memories after trauma. Although this may seem in contrast with our earlier proposal of the genetic origins of executive control, high heritability does not mean that environmental factors cannot and do not affect executive functions (cf. Friedman et al., 2008). According to Friedman et al. (2008): “…heritability is an estimate of the genetic influence on individual differences around a population mean (i.e., the population variance), rather than an estimate of the influences on the mean itself. Thus, environmental factors can

121

Chapter 7

influence a population’s average at the same time that genetic factors influence its variance (Scarr, 1992). The effects of training on executive functions may be analogous.” (Friedman et al., 2008, p. 218). In line with this reasoning, accumulating evidence in the experimental and neuro-imaging literature has shown that executive processes and their underlying brain regions are plastic and adaptive and can be modified by training (e.g., Erickson et al., 2007; Kramer, Larish & Strayer, 1995). For example, some studies have investigated rehabilitation of executive functions in clinical (e.g., ADHD; White & Shaw, 2006) and aging populations (Kramer et al., 1995). Whether these observed training induced changes in behavior and brain activity would transfer to traumatized people with inefficient executive control is an important question that deserves further research.

122

References American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. Washington, DC: American Psychiatric Association, 2000. Anderson, M. C., Bjork, R. A., Bjork, E. L. (1994). Remembering can cause forgetting: Retrieval dynamics in long-term memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20, 1063-1087. Anderson, M. C., & Green, C. (2001). Suppressing unwanted memories by executive control. Nature, 410, 366-369. Anderson, M. C., & Levy, B. J. (2007). Theoretical issues in inhibition: Insights from research on human memory. In Gorfein, D. & MacLeod, C. (Eds.). Inhibition in Cognition (pp. 81-102). Washington, DC: American Psychological Association. Anderson, M. C., & Spellman, B. A. (1995). On the status of inhibitory mechanisms in cognition: Memory retrieval as a modal case. Psychological Review, 102, 68-100. Arnell, K. M., Killman, K. V., & Fijavz, D. (2007). Blinded by emotion: Target misses follow attention capture by arousing distracters in RSVP. Emotion, 7, 465-477. Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory: A proposed system and its control processes. In: K. W. Spence & J. T. Spence. The Psychology of Learning and Motivation: II. Oxford, UK: Academic Press. Baayen, R.H., Piepenbrock, R., & van Rijn, H. 1993. The CELEX lexical data base on CDROM. Philadelphia: Linguistic Data Consortium. Baddeley, A. (1996). Exploring the central executive. The Quarterly Journal of Experimental Psychology, 49, 5-28. Bar-Haim, Y., Lamy, D., Pergamin, L., Bakermans-Kranenburg, M. J., & IJzerdoorn, M. H. (2007). Threat-related attentional bias in anxious and non-anxious individuals: A meta-analytic study. Psychological Bulletin, 133, 1-24. Barnard., P. J., Ramponi, C., & Battye, G. (2005). Anxiety and the deployment of visual attention over time. Visual Cognition, 12, 181-211. Barnier, A.J., Levin, K., & Maher, A. (2004). Suppressing thoughts of past events: Are repressive copers good suppressors? Cognition and Emotion, 18, 513-531. 123

References

Bernat, J. A., Ronfeldt, H. M., Calhoun, K. S., & Arias, I. (1998). Prevalence of traumatic events and peri-traumatic predictors of posttraumatic stress symptoms in a nonclinical sample of college students. Journal of Traumatic Stress, 11, 645-664 Berntsen, D. (2001). Involuntary memories of emotional events: Do memories of trauma’s and extremely happy events differ? Applied Cognitive Psychology, 15, 135-158. Berntsen, D., & Hall, N. M. (2004). The episodic nature of involuntary autobiographical memories. Memory & Cognition, 32, 789-803. Berntsen, D. (2007). Involuntary autobiographical memories: Speculations, findings, and an attempt to integrate them. In J. H. Mace (Ed.), Involuntary Memory (pp. 20-49). Oxford, UK: Blackwell. Boomsma, I. (1998). Genetic analysis of cognitive failures (CFQ): A study of Dutch adolescent twins and their parents. European Journal of Personality, 12, 321-330. Bouma, J., Ranchor, A. V., Sanderman, R., & Van Sonderen, E. (1995). Het meten van symptomen van depressie met de CES-D: Een handleiding [The measurement of symptoms of depression with the CES-D: A manual]. Rijksuniversiteit Groningen: Noordelijk Centrum voor Gezondheidsvraagstukken. Brewin, C.R. (1998). Involuntary autobiographical memories in depression and posttraumatic stress disorder. Applied Cognitive Psychology, 12, 359-370. Brewin, C. R., Andrews, B., & Valentine, J. D. (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology, 68, 748-766. Brewin, C. R., & Beaton, A. (2002). Thought suppression, intelligence, and working memory capacity. Behavior, Research & Therapy, 40, 923-930. Brewin, C.R., Christodoulides, J., & Hutchinson, G. (1996). Involuntary thoughts and involuntary

memories in a nonclinical sample. Cognition and Emotion, 10, 107-

112. Brewin, C. R., Dalgleish, T., & Joseph, S. (1996). A dual representation theory of posttraumatic stress disorder. Psychological Review, 107, 670-686. Brewin, C. R., & Saunders, J. (2001). The effect of dissociation at encoding on intrusive memories for a stressful film. British Journal of Medical Psychology, 74, 467-472. Brewin, C. R., & Smart, L. (2005). Working memory capacity and suppression of involuntary thoughts. Journal of Behavior Therapy and Experimental Psychiatry, 36, 61-68.

124

Broadbent, D.E., Cooper, P.F., FitzGerald, P., & Parkes, K.R. (1982). The Cognitive Failures Questionnaire (CFQ) and its correlates. British Journal of Clinical Psychology, 21, 116. Buckley, T. C., Blanchard, E. B., & Neill, W. T. (2000). Information processing and PTSD: A review of the empirical literature. Clinical Psychology Review, 28, 1041-1065. Bywaters, M., Andrade, J., & Turpin, G. (2004). Intrusive and non-intrusive memories in a non-clinical sample: The effects of mood and affect on imagery vividness. Memory, 12, 467-478. Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., Wilhelm, O., & Engle, R. W. (2005). Working memory span tasks: A methodological review and user's guide. Psychonomic Bulletin & Review, 12, 769-786. Conway, M. A., & Pleydell-Pearce, C. W. (2000). The construction of autobiographical memories in the self-memory system. Psychological Review, 107, 261-288. Conway, M. A., Singer, J. A., & Tagini, A. (2004). The self and autobiographical memory: correspondence and coherence. Social Cognition, 22, 491-529. Crowne, D.P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349-354. Dalgleish, T. (2004). Cognitive approaches to posttraumatic stress disorder: The evolution of multi-representational theorizing. Psychological Bulletin, 130, 228-260. Davies, M., & Clark, D. M. (1998). Predictors of analogue post-traumatic intrusive cognitions. Behavioral and Cognitive Psychotherapy, 26, 303-314. De Groot, A. M. B. (1980). Mondelinge Woordassociatienormen: 100 Woordassociaties op 460 Nederlandse Zelfstandige Naamwoorden.[Oral Word Association Norms: 100 Word Associations on 460 Dutch Nouns.] Lisse: Swets & Zeitlinger. Delis, D.C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1987). The California Verbal Learning Test. San Antonio, TX: Psychological Corporation. Dempster, F. N. (1993). Resistance to interference: Developmental changes in a basic processing dimension. In M. L. Howe & R. Pasnak. Emerging Themes in Cognitive Development. Vol. 1: Foundations (pp. 3-27). New York: Springer. Dempster, F. N., & Corkill, A. J. (1999). Individual differences in susceptibility to interference and general cognitive ability. Acta Psychologica, 101, 395-416.

125

References

Derryberry, D., & Reed, M. A. (2002). Anxiety-related attentional biases and their regulation by attentional control. Journal of Abnormal Psychology, 111, 225–236. Duke, L. M., & Vasterling, J. (2005). Epidemiological and methodological issues in neuropsychological research on PTSD. In J.J. Vasterling & C. R. Brewin (Eds.), Neuropsychology of PTSD: Biological, cognitive and clinical perspectives (pp. 178207). New York, NY: The Guilford Press. Ehlers, A., & Clark, D. M. (2000). A cognitive model of post-traumatic stress disorder. Behavior Research and Therapy, 38, 319-345. Ehlers, A., Hackmann, A., & Michael, T. (2004). Intrusive re-experiencing in post-traumatic stress disorder: Phenomenology, theory, and therapy. Memory, 12, 403-415. Ehlers, A., Hackmann, A., & Steil, R.(2002). The nature of involuntary memories after trauma: the warning signal hypothesis. Behavior Research and Therapy, 40, 9951002. Ehlers, A., Michael, T., Chen, Y. P., Payne, E., & Shan, S. (2006). Enhanced perceptual priming for neutral stimuli in a traumatic context: A pathway to intrusive memories? Memory, 12, 316-328. Ehlers, A., & Steil, R. (1995). Maintenance of intrusive memories in posttraumatic stress disorder: A cognitive approach. Behavioural and Cognitive Psychotherapy, 23, 217249. Ellis, H. C. (1990). Depressive deficits in memory: processing initiative and resource allocation. Journal of Experimental Psychology: General, 119, 60-62. Ellis, H. C. (1991). Focused attention and depressive deficits in memory. Journal of Experimental Psychology: General, 120, 310-312. Engelhard, I. M., Van den Hout, M. A., & Kindt, M. (2003). The relationship between neuroticism, pre-traumatic stress, and post-traumatic stress: A prospective study. Personality and Individual Differences, 35, 381-388. Engle, R. W. (2002). Working memory capacity as executive attention. Current Directions in Psychological Science, 11, 19-23. Engle, R. W., Conway, A. R. A., Tuholski, S. W., & Shisler, R. J. (1995). A resource account of inhibition. Psychological Science, 6, 122-125. Friedman, N.P., & Miyake, A. (2004). The relations among inhibition and interference control

126

functions: A latent-variable analysis. Journal of Experimental Psychology: General, 133,101-135. Friedman, N. P., Miyake, A., Young, S. E., DeFries, J. C., Corley, R. P., & Hewitt, J. K. (2008). Individual differences in executive functions are almost entirely genetic in origin. Journal of Experimental Psychology: General, 2,201-225. Ginzburg, K., Solomon, Z. & Bleich, A. (2002). Repressive coping style, acute stress disorder, and posttraumatic stress disorder after myocardial infarction. Psychosomatic Medicine ,64, 748-757. Hackmann, A., Ehlers, A., Speckens, A., & Clark, D. M. (2004). Characteristics and content of intrusive memories in PTSD and their changes with treatment. Journal of Traumatic Stress, 17, 231-240. Harnishfeger, K. K. (1995). The development of cognitive inhibition: Theories, definitions, and research evidence. In F. N. Dempster & C. J. Brainerd (Eds.), Interference and Inhibition in Cognition (pp. 175-204). San Diego, CA: Academic Press. Hasher, L., Zacks, R.T., & May, C.P. (1999). Inhibitory control, circadian arousal, and age. In D. Gopher & A. Koriat (Eds.), Attention and Performance XVII: Cognitive regulation of performance: Interaction of theory and application (pp. 653-675). Cambridge, MA, US: The MIT Press. Hedden, T. & Yoon, C. (2006). Individual differences in executive processing predict susceptibility to interference in verbal working memory. Neuropsychology, 20, 511528. Hendriks, A. A. J., Hofstee, W. K. B., De Raad, B., & Angleiter, A. (1999). The Five-Factor Personality Inventory (FFPI). Personality and Individual Differences, 27, 307-325. Holmes, E. A. & Bourne, C. (2008). Inducing and modulating intrusive emotional memories: A review of the trauma film paradigm. Acta Psychologica, 127, 553-566. Holmes, E. A., Brewin, C. R., & Hennessey, R. G. (2004). Trauma films, information processing, and intrusive memory development. Journal of Experimental Psychology: General, 133, 3-22. Horne, J.A., & Östberg, O. (1976). A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Journal of Chronobiology, 4, 97-110.

127

References

Horowitz, M., Wilner, N., & Alvarez, W. (1979). Impact of event scale: A measure of subjective stress. Psychosomatic Medicine, 41, 209-218. Johnson, J.H., Sarason, I.G., & Siegel, J.M. (1979). Arousal seeking as a moderator of life stress. Perceptual and Motor Skills, 49, 665-666. Kane, M. J., & Engle, R. W. (2000). Working-memory capacity, proactive interference, and divided attention: Limits on long-term memory retrieval. Journal of Experimental Psychology: Learning,, Memory, and Cognition, 26, 336-358. Kane, M. J., & Engle, R. W. (2003). Working-memory capacity and the control of attention: The contributions of goal neglect, response competition, and task set to Stroop interference. Journal of Experimental Psychology: General, 132, 47-70. Kane, M. J., Hambrick, D. Z. Tuholski, S. W., Oliver, W., Payne, T. W., & Engle, R. W. (2004). The generality of working memory capacity: A latent-variable approach to verbal and visiospatial memory span and reasoning. Journal of Experimental Psychology: General, 133, 189-217. Kane, M. J., Conway, A. R. A., Hambrick, D. Z., & Engle, R. W. (2007). Variation in working memory capacity as variation in executive attention and control. In: A. R. A. Conway, C. Jarrold, M. J. Kane, A. Miyake, & J. N. Towse. Variation in Working Memory (pp. 21-46). New York, NY, US: Oxford University Press. Kaplan, Z., Weiser, M., Reichenberg, A., Rabinowitz, J., Caspi, A., & Bodner, E. (2002). Motivation to serve in the military influences vulnerability to future posttraumatic stress disorder. Psychiatry Research, 109, 45-49. Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M. (1995). Posttraumatic stress disorder in the National Comorbidity Survey. Archives of General Psychiatry, 52, 1048-1060. Kindt, M., & Engelhard, I. M. (2005). Trauma processing and the development of posttraumatic stress disorder. Journal of Behavior Therapy and Experimental Psychiatry, 36, 69-76. Klein, K., & Boals, A. (2001). The relationship of life event stress and working memory capacity. Applied Cognitive Psychology, 15, 565-579. Kramer, J. H., & Delis, D. C. (1991). Interference effect on the California Verbal Learning Test: A construct validation study. Psychological Assessment, 3, 299-302. Kuyken, W., & Brewin, C.R. (1994). Involuntary memories of childhood abuse during depressive episodes. Behavior Research and Therapy, 32, 25-28.

128

Lauterbach, D., & Vrana, S. (2001). The relationship among personality variables, exposure to traumatic events, and severity of posttraumatic stress symptoms. Journal of Traumatic Stress, 14, 29-45. Levy, B. J., & Anderson, M. C. (2002). Inhibitory processes and the control of memory retrieval. Trends in Cognitive Science, 6, 299-305. Levy, B. J., & Anderson, M. C. (2008). Individual differences in the suppression of unwanted memories: the executive deficit hypothesis. Acta Psychologica, 127, 623-635. Lustig, C., Hasher, L., & May, C. P. (2001). Working memory span and the role of proactive interference. Journal of Experimental Psychology: General, 130, 199-207. MacLeod, C. M. (2005). The Stroop task in cognitive research. In A. Wenzel & D. C. Rubin (Eds.), Cognitive methods and their application to clinical research (pp. 17-40). Washington, DC: American Psychological Association. Mayou, R. A., Ehlers, A., Hobbs, M. (2000). Psychological debriefing for road-traffic accident victims: Three-year follow-up of a randomized control trial. British Journal of Psychiatry, 176, 589-593. Macklin, M. L., Metzger, L. J., Litz, B. T., McNally, R. J., Lasko, N. B., Orr, S. P., & Pittman, R. K. (1998). Lower precombat intelligence is a riskfactor for posttraumatic stress disorder. Journal of Consulting and Clinical Psychology, 66, 323-326. McNally, R. J. (2003). Psychological mechanisms in acute response to trauma. Biological Psychiatry, 53, 779-786. McNally, R. J., Bryant, R. A. & Ehlers, A. (2003). Does early psychological intervention promote recovery from posttraumatic stress? Psychological Science in the Public Interest, 4, 45-79. McNally, R. J., & Shin, L. M. (1995). Association of intelligence with severity of posttraumatic stress disorder symptoms in Vietnam combat veterans. American Journal of Psychiatry, 152, 936-938. Mecacci, L., Righi, S., & Rocchetti, G. (2004). Cognitive failures and circadian typology. Personality and Individual Differences, 37, 107-113. Mendiola, M. (2002). An index of self-regulation of emotion and the study of repression in social contexts that threaten or do not threaten self-concept. Emotion, 2, 215-232.

129

References

Michael, T., & Ehlers, A. (2007). Enhanced perceptual priming for neutral stimuli occurring in a traumatic context: Two experimental investigations. Behaviour Research and Therapy, 45, 341-358. Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. (2000). The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: A latent variable analysis. Cognitive Psychology, 41, 49-100. Morgan, H. J., & Janoff-Bulman, R. (1994). Positive and Negative self-complexity: Patterns of adjustment following traumatic versus non-traumatic life experiences. Journal of Social and Clinical Psychology 13, 63-85. Morgan, I. A., Matthews, G., & Winton, M. (1995). Coping and personality as predictors of posttraumatic intrusions, numbing, avoidance and general distress: A study of victims of the Perth flood. Behavioral and Cognitive Psychotherapy, 23, 251-264. Most, S. B., Chun, M. M., & Widders, D. M. (2005). Attentional rubbernecking: Cognitive control and personality in emotion-induced blindness. Psychonomic Bulletin & Review, 12, 654-661. Most, S. B., Smith, S. D., Cooter, A. B., Levy, B. N., & Zald, D. H. (2007). The naked truth: Positive, arousing distracters impair rapid target perception. Cognition and Emotion, 21, 964-981. Muris, P. (2006). Unique and interactive effects of neuroticism and effortful control on psychopathological symptoms in non-clinical adolescents. Personality and Individual Differences, 40, 1409-1419. Muris, P., de Jong, P. J., & Engelen, S. (2004). Relationships between neuroticism, attentional control, and anxiety disorders in non-clinical children. Personality and Individual Differences, 37, 789-797. Murray, J., Ehlers, A., & Mayou, R. A. (2002). Dissociation and posttraumatic stress disorder: Two prospective studies of road traffic accident victims. British Journal of Psychiatry, 180, 363-368. Myers, L.B., & Derakshan, N. (2004). To forget or not to forget: What do repressors forget and when do they forget? Cognition and Emotion, 18, 495-511.

130

Nigg, J. T. (2000). On inhibition/disinhibition in developmental psychopathology: Views from cognitive and personality psychopathology and a working inhibition taxonomy. Psychological Bulletin, 126, 220-246. Nixon, R. D. V., Nehmy, T., & Seymour, M. (2007b). The effect of cognitive load and hyper arousal on negative intrusive memories. Behaviour Research and Therapy, 45, 26522663. Ozer, E. J., Best, S. R., Lipsey, T. L., & Weiss, D. S. (2003). Predictors of posttraumatic stress disorder and symptoms in adults: A meta-analysis. Psychological Bulletin, 129, 52-73. Pennington, B. F., & Ozonoff, S. (1996). Executive functions and developmental psychopathology. Journal of Child Psychology and Psychiatry, 37, 51-87. Pineles, S . L., Shipherd, J. C., Welch, L. P., & Yovel, I. (2007). The role of attentional biases in PTSD: Is it interference or facilitation? Behaviour Research and Therapy, 45, 1903-1913. Reynolds, M., & Brewin, C.R. (1998). Involuntary cognitions, coping strategies and emotional responses in depression, post-traumatic stress disorder and a non-clinical population. Behaviour Research and Therapy, 36, 135-147. Rosen, V. M., & Engle, R. W. (1997). The role of working memory capacity in retrieval. Journal of Experimental Psychology: General, 126, 211-227. Rosen, V. M., & Engle, R. W. (1998). Working memory capacity and suppression. Journal of Memory and Language, 39, 418-436. Rothbaum, O. R., Foa, E. B., Riggs, D. S., Murdock, T., & Walsh, W. (1992). A prospective examination of posttraumatic stress disorder in rape victims. Journal of Traumatic Stress, 3, 455-475. Ruts, W., De Deyne, S., Ameel, E., Vanpaemel, W., Verbeemen, T., & Storms, G. (2004). Dutch norm data for 13 semantic categories and 338 exemplars. Behavior Research Methods, Instruments & Computers, 36, 506-515. Schacter, D. L. (2001). The seven sins of memory: how the mind forgets and remembers. Boston: Hougthon Mifflin Company. Smith, S. D., Most, S. B., Newsome, L. A., & Zald, D. H. (2006). An emotion-induced attentional blink elicited by aversively conditioned stimuli. Emotion, 6, 523-527.

131

References

Spielberger, C.D. (1995). State-Trait Depression Scales (Form X-1). Mind Garden, Palo Alto, CA. Spielberger, C.D., & Vagg, P.R. (1984). Psychometric properties of the STAI: A reply to Ramanaiah, Franzen, and Schill. Journal of Personality Assessment, 48, 95-97. Steel, C., & Holmes, E. A. (2007). The role of involuntary memories in posttraumatic disorder and psychosis. In J. H. Mace (Ed.), Involuntary Memory (pp. 20-49). Oxford, UK: Blackwell. Stuart, A. D. P., Holmes, E. A., & Brewin, C. R. (2006). The influence of a Visio-spatial grounding task on intrusive images of a traumatic film. Behavior Research and Therapy, 44, 611-619. Towse, J. N., & Neil D. (1998). Analyzing human random generation behavior: A review of methods used and a computer program for describing performance. Behavior Research Methods, Instruments, & Computers, 30, 583-591. Tulving, E. (2002). Episodic memory: From mind to brain. Annual review of psychology, 53, 1-25. Tulving, E., & Schacter, D. L. (1990). Priming and human memory systems. Science, 247, 301-306. Turner, M. L., & Engle, R. W. (1989). Is working memory capacity task dependent? Journal of Memory and Language, 28, 127-154. Uddo, M., Vasterling, J. J., Brailey, K., & Sutker, P.B. (1993). Memory and attention in combat-related post-traumatic stress disorder (PTSD). Journal of Psychopathology and Behavioral Assessment, 15, 43-52. Unsworth, N., Schrock, J. C. & Engle, R. W. (2004). Working memory capacity and the antisaccade task: Individual differences in voluntary saccade control. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30, 1302-1321. Unsworth, N. & Engle, R. W. (2007). The nature of individual differences in working memory capacity: Active maintenance in primary memory and controlled search from secondary memory. Psychological Review, 114, 104-132. Van der Ploeg, E., Mooren, T. M., Kleber, R. J., Van der Velden, P. G., & Brom, D. (2004). Construct validation of the Dutch version of the Impact of Event Scale. Psychological Assessment, 16, 16-26.

132

Vassiliki, H., & Tarrier, N. (2001). Personality and peritraumatic dissociation in the prediction of PTSD in victims of road traffic accidents. Journal of Psychosomatic Research, 51, 687-692. Vasterling, J. J., & Brailey, K. (2005). Neuropsychological findings in adults with PTSD. In J.J. Vasterling & C. R. Brewin (Eds.), Neuropsychology of PTSD: Biological, cognitive and clinical perspectives (pp. 178-207). New York, NY: The Guilford Press. Vasterling, J. J., Brailey, K., Constants, J. I., & Sutker, P.B. (1998). Attention and memory dysfunction in posttraumatic stress disorder. Neuropsychology, 12, 125-133. Vasterling, J. J., Duke, L. M., Brailey, K., Constants, J. I., Allain, A. N., Jr., & Sutker, P. B. (2002). Attention, learning, and memory performances and intellectual resources in Vietnam veterans: PTSD and no disorder comparisons. Neuropsychology, 16, 5-14. Verwoerd, J., de Jong, P. J., & Wessel, I. (2006). ACS: Dutch translation of the Attentional Control Scale, originally developed by Derryberry and Reed (2002). Verwoerd, J. R. L., de Jong, P. J., & Wessel, I. (2008). Low attentional control and the development of intrusive memories following a laboratory stressor. Journal of Psychopathology and Behavioral Assessment, 30, 291-297. Verwoerd, J. R. L., & Wessel, I. (2007). Distractibility and individual differences in the experience of intrusive memories. Personality and Individual Differences, 42, 325334. Verwoerd, J. R. L., Wessel, I., & de Jong, P. J. (in press). Individual differences in experiencing intrusive memories: The role of the ability to resist proactive interference. Journal of Behavior Therapy and Experimental Psychiatry. Wegner, D., Schneider, D., Carter, S., & White, T. (1987). Paradoxical effects of thought suppression.. Journal of Personality and Social Psychology, 53, 5-13. Weinberger, D.A. (1990). The construct validity of the repressive coping style. In J.L. Singer (Ed.),

Repression

and

dissociation:

Implications

for

personality

theory,

psychopathology, and health (pp. 337-386). Chicago: University of Chicago Press. Wenzlaff, R. M., & Wegner, D. M. (2004). Thought suppression. Annual Review of Psychology, 51, 59-91.

133

References

Wessel, I., Overwijk, S., Verwoerd, J., & de Vrieze (2008). Pre-stressor cognitive control is related to intrusive cognition of a stressful film. Behaviour Research and Therapy, 46, 496-513. Williams, J.M.G., & Broadbent, K. (1986). Autobiographical memory in suicide attempters. Journal of Abnormal Psychology, 95, 144-149. Williams, J. M. G., Mathews, A., & Macleod, C. (1996). The emotional Stroop task and psychopathology. Psychological Bulletin, 120, Yehuda, R., Keefe, R. S. E., Harvey, P. D., Levengood, R. A., Gerber, D. K., Geni, J., & Siever, L. J. (1995). Learning and memory in combat veterans with posttraumatic stress disorder. American Journal of Psychiatry, 152, 137-139. Yoon, C., May, C.P., & Hasher, L. (2000). Aging, circadian arousal patterns, and cognition. In D.C. Park ., & N. Schwarz (Eds.). Cognitive aging: A primer (pp. 151-171). New York: Psychology Press.

134

Samenvatting

135

Samenvatting

Een natuurlijke reactie in de nasleep van een traumatische gebeurtenis is het ongewild ervaren van sensorische beelden, ook wel intrusieve herinneringen genoemd. Slachtoffers van traumatische gebeurtenissen zoals oorlogsgeweld, seksueel misbruik of een ernstig verkeersongeval worden als het ware teruggeworpen in de tijd om delen van het trauma, inclusief de oorspronkelijke intense emoties opnieuw te beleven. Als voorbeeld van zo´n ervaring geeft de Londense psycholoog en trauma specialist Anke Ehlers een verkeersslachtoffer die bij het zien van stralen zonlicht plotseling een angstig beeld voor zich zag van naderende koplampen. Die lampen had hij ook gezien tijdens het verloop van het verkeersongeval. Dit soort zeer stressvolle, over het algemeen visuele intrusies, kunnen het dagelijks leven danig beperken en lokken in veel gevallen ook sterke vermijdingsreacties uit van situaties en zaken die sterk aan het trauma doen denken. Ondanks de heftige reacties die een traumatische gebeurtenis teweeg kan brengen, vertoont het merendeel van de overlevenden een opmerkelijk veerkracht. Voor de meeste mensen geldt dat ze binnen enkele weken van de steeds terugkerende intrusieve herinneringen/herbelevingen verlost zullen zijn. Slechts bij een relatief klein aantal personen zullen de symptomen voortduren en uitmonden in een klinische stoornis, waarvan de Posttraumatische Stress Stoornis (PTSS) de bekendste is. Of mensen nu wel of niet een PTSS ontwikkelen kan niet goed worden voorspeld op basis van de ernst van het trauma. Deze klinische observatie suggereert dat er individuele verschillen in het spel zijn die mede uitmaken of iemand wel of niet last blijft houden van ongewenste intrusieve herinneringen na een trauma. Maar wat voor individuele verschillen zijn dat dan? Wat voor factoren zorgen ervoor dat sommige mensen wel last blijven houden van intrusieve herinneringen en anderen niet? Deze vraag is de leidraad voor dit dissertatieonderzoek. Om de vragen die in dit proefschrift naar voren te komen beter te kunnen begrijpen, volgt nu eerst een introductie van enkele essentiële concepten uit de aandacht- en geheugen literatuur. In hedendaagse theorieën over het werkgeheugen speelt het begrip controle een belangrijke rol. Het werkgeheugen zou omschreven kunnen worden als de centrale processor van de menselijke geest, verantwoordelijk voor het selecteren, actief houden en manipuleren van informatie ten behoeve van specifieke doelen. Het idee is dat er aangeboren verschillen bestaan in de mate waarin iemand in staat is zijn werkgeheugen te controleren. In de

136

literatuur wordt dit aangeboren verschil in resources/capaciteit uitgedrukt met de homunculus-achtige term “executieve controle”. Het filosofische begrip homunculus voert terug op bekende problemen m.b.t. het wetenschappelijk verklaren van bewustzijn, de vrije wil en de verhouding tussen lichaam en geest. Dat wil zeggen: de notie van executieve controle veronderstelt iemand die deze controle uitvoert – een denkbeeldig mannetje in ons hoofd, ook wel homunculus genoemd. Bij gebrek aan voldoende conceptueel gereedschap zal op deze plaats niet verder worden ingegaan op deze filosofische kwestie/queeste en zal het homunculus-probleem zich in een nieuw jasje als executieve controle blijven voordoen. Uit onderzoek blijkt dat executieve controle een grote rol speelt bij het reguleren van verschillende aspecten van doelgericht gedrag. Kerndoel van de huidige thesis is het stapsgewijs onderzoeken van de voorspellende rol van executieve controle bij het in stand houden van intrusieve herinneringen. De hypothese is dat de mate van executieve controle over het werkgeheugen voorafgaand aan een traumatische ervaring de uitkomst kan voorspellen tussen natuurlijk herstel of het in stand houden van intrusieve herinneringen. Dit idee is meer specifiek en gedetailleerd weergegeven in het model dat aan het aan het einde van het eerste hoofdstuk wordt gepresenteerd. Dit intrusiemodel was het uitgangspunt voor de empirische studies die in deze thesis zijn beschreven. Behalve executieve controle en intrusieve herinneringen spelen in dit model ook herinneringscues in de externe omgeving een belangrijke rol. Uit klinische observaties bij patiënten met PTSS blijkt dat intrusieve herinneringen voornamelijk worden opgeroepen door perceptuele omgevingsstimuli die een directe relatie hebben met stimuli die tijdens het trauma aanwezig waren. Een voorbeeld van zo’n perceptuele herinneringscue is het zonlicht dat voor een angstige herbeleving aan een verkeersongeluk zorgt. In het eerder beschreven voorbeeld van een visuele intrusie heeft het zonlicht een sterke perceptuele relatie met de koplampen die het verkeersslachtoffer tijdens het ongeluk op hem af zag komen. Doordat de herinnering aan het trauma sterk in het geheugen geactiveerd is (dit wordt ook wel priming genoemd) zal het zonlicht ogenschijnlijk automatisch voor de activatie van het angstige beeld van naderende koplampen zorgen. De vraag die in deze dissertatie wordt onderzocht is of natuurlijk herstel na een trauma (bij de meerheid van de slachtoffers) afhangt van de vaardigheid om de aandacht tijdig van perceptuele herinneringscues af te wenden. Controle over herinneringscues zoals het zonlicht kan personen in staat stellen om nieuwe, meer gezonde geheugenassociaties te vormen, zoals zonlicht ---> warmte. Dit betekent dat cues als

137

Samenvatting

zonlicht of koplampen langzaam aan geen angst of gevoelens van dreiging meer oproepen (intrusies), maar hun normale neutrale (of positieve) betekenis weer terug krijgen. Zoals vermeldt in het model vergt dit proces van natuurlijk herstel na een traumatische gebeurtenis een zekere mate van executieve controle. Personen die van nature een gebrekkige executieve controle bezitten zullen in de nasleep van een traumatische gebeurtenis veel moeite hebben met het afwenden van de aandacht van herinneringscues die stressvolle intrusies oproepen. Dit betekent dat deze cues hun sterke associatie met de stressvolle herinneringen behouden, waardoor het erg moeilijk is om nieuwe, meer gezonde associaties te vormen. In het huidige proefschrift worden verschillende componenten van het model onderling en in samenhang met elkaar onderzocht. De nadruk ligt hierbij op het stapsgewijs onderzoeken van de relatie tussen executieve controle en intrusieve herinneringen. Het laatste empirische hoofdstuk richt zich op het additionele belang van perceptuele herinneringscues bij het in stand houden van het frequent ervaren van intrusieve herinneringen. In de verschillende studies wordt gebruik gemaakt van ongeselecteerde groepen psychologiestudenten. Het werken met “gezonde” proefpersonen heeft als voordeel dat het cognitief functioneren bij deze groep voorafgaand aan de ervaring van een stressvolle of traumatische gebeurtenis bepaald kan worden. Hierdoor is het relatief eenvoudig is om de unieke bijdrage van executieve controle bij het in stand houden van symptomen zoals intrusieve herinneringen te bepalen. Bij een groep met PTSS patiënten is dit veel complexer omdat de traumatische ervaring, samen met veel voorkomende comorbide stoornissen zoals depressie en angst (e.g., paniek, fobieën) invloed kan hebben op het verminderd cognitief functioneren na een trauma. Hierdoor is het erg lastig om oorzaak en gevolg m.b.t. gebrekkige executieve controle en traumagerelateerde symptomen te achterhalen.

Studie 1 (retrospectief): Intrusieve herinneringen en afleidbaarheid in het dagelijks leven In de studie die beschreven wordt in hoofdstuk 2 vulde een grote groep van ongeveer 400 studenten een aantal vragenlijsten in om de relatie tussen executieve controle en intrusieve herinneringen te onderzoeken. Een eerste vragenlijst vroeg naar de mate van afleidbaarheid in het dagelijks leven in de vorm van het maken van cognitieve fouten. Deze vragenlijst werd gebruikt als maat voor executieve controle met het idee dat een groot aantal cognitieve fouten

138

vergeleken kan worden met een verminderde executieve controle. Voorbeelden van items op deze cognitieve fouten vragenlijst zijn: “afspraken vergeten” (nooit – zeer vaak), “Beginnen met iets en er halverwege mee ophouden omdat u afgeleid wordt door iets anders” (nooit – zeer vaak), en “In een gesprek iets eruit flappen dat wel eens beledigend voor anderen zou kunnen zijn” (nooit – zeer vaak). Daarnaast werd proefpersonen gevraagd om terug te denken aan een zeer stressvolle/traumatische gebeurtenis uit het verleden en hierover, indien aanwezig, het aantal recente (afgelopen 7 dagen) intrusies en vermijdingsreacties te rapporteren. Voorbeelden van items behorend bij de intrusievragenlijst zijn: “beelden ervan kwamen plotseling boven” (helemaal niet- vaak), “sterke gevoelens overspoelden me” (helemaal niet- vaak) en “ik probeerde er niet over te praten” (helemaal niet- vaak). De verwachting was dat personen met een relatief groot aantal cognitieve fouten het meeste last zouden hebben van recente intrusies en vermijdingsreacties n.a.v. de stressvolle levensgebeurtenis. In overeenstemming met de hypothese was er een significant verband te zien tussen het aantal cognitieve fouten in het dagelijks leven en het aantal intrusieve herinneringen. Eerder onderzoek heeft laten zien dat een negatieve stemming zowel samenhangt met een groot aantal cognitieve fouten als met een hoge frequentie van het aantal intrusieve herinneringen. Het was dus belangrijk om te kijken of de relatie tussen cognitieve fouten en intrusieve herinneringen niet gewoon een bijeffect was van een algemeen negatieve stemming. Dit bleek niet het geval; de relatie bleek onafhankleijk van verschillen in depressie. Behalve voor depressie werd er ook succesvol gecontroleerd voor de leeftijd van de stressvolle gebeurtenis (i.e., hoe lang geleden vond de gebeurtenis plaats?) en voor de impact die de gebeurtenis in het verleden had. Dit patroon van resultaten is in overeenstemming met het idee dat individuele verschillen in executieve controle een unieke bijdrage kunnen leveren aan de verklaring van individuele verschillen in intrusieve herinneringen n.a.v. stressvolle en traumatische gebeurtenissen.

Studie 2 (retrospectief): De rol van specifieke executieve functies bij het verklaren van intrusieve herinneringen De eerste studie maakte gebruik van een zelfrapportage maat voor executieve controle. Het valt echter niet uit te sluiten dat een dergelijke zelfrapportage geen accuraat beeld verschaft van de “echte” executieve controle functies. Om die reden maakt studie 2

139

Samenvatting

gebruik van een computer taak om individuele verschillen in executieve controle in kaart te brengen. Daarnaast is het belangrijk vast te stellen dat executieve controle geen ondeelbaar construct is. In de experimentele geheugen literatuur wordt uitgegaan van het idee dat executieve controle verschillende onafhankelijke functies dient. Zo kan er controle plaatsvinden op het niveau van motorische acties (het remmen van een automatische respons) of op het niveau van het werkgeheugen (het tegengaan/remmen van ongewenste informatie). In de studie in hoofdstuk drie werd de executieve functie die betrokken is bij het weren van ongewenste informatie uit het werkgeheugen vergeleken met de executieve functie om een automatische respons tegen te gaan. Ter illustratie: een goede vriend geeft u zijn nieuwe telefoonnummer door. Bij de meeste mensen zal het de eerste weken zeer moeilijk zijn om dit nieuwe nummer direct op te roepen wanneer ze de vriend willen bellen, omdat het veel sterker geactiveerde oude nummer zich steeds opnieuw zal opdringen. Een flexibele toegang tot het nieuwe nummer vergt een goed ontwikkelde (executieve) vaardigheid om interferentie van het oude nummer uit het werkgeheugen te weren. Deze vaardigheid zorgt ervoor dat het oude nummer langzaam minder actief wordt in het geheugen om hiermee het nieuwe nummer voorrang te geven. Een mooi voorbeeld van de vaardigheid om een automatische respons tegen te gaan wordt gegeven door de Amerikaanse geheugen psycholoog Michael Anderson. Hij beschrijft een situatie waarin hij op een warme dag het raam open wil zetten om wat frisse lucht op te vangen. Op het moment dat hij dit doet valt er een cactus naar beneden die in het raamkozijn stond. In een eerste reactie wil hij de plant opvangen om te voorkomen dat de pot op de grond valt en breekt, maar realiseert zich in een fractie van een seconde dat dit misschien niet zo’n goed idee is. Net op tijd kan hij de al in gang gezette actie tegengaan en een zeer pijnlijke ervaring voorkomen. Uit het bovenstaande moge duidelijk zijn dat zowel de vaardigheid om ongewenste informatie uit het werkgeheugen te weren, als de vaardigheid om een automatische response te remmen van essentieel belang zijn voor de controle over specifieke componenten van ons gedrag. Voor het weren van ongewenste herinneringen lijkt het tegengaan van responsen minder relevant dan het tegengaan van irrelevante informatie in het werkgeheugen. Om te onderzoeken of inderdaad voornamelijk een gebrekkige vaardigheid om irrelevante informatie uit het werkgeheugen te weren samenhangt met het frequent ervaren van intrusieve herinneringen, zijn proefpersonen in studie 3 onderworpen aan een aantal computertaken. Dit om beide executieve functies in kaart te brengen. Als maat voor de

140

vaardigheid om irrelevante informatie uit het werkgeheugen te weren werd een taak gebruikt waarbij proefpersonen verschillende lijsten met woordparen moesten leren. Deze lijsten met woordparen waren zo samengesteld dat het leren van de tweede lijst sterk interfereerde met de woorden die eerder geleerd waren (e.g., bakker - brood [lijst 1]; bakker - meel [lijst 2]; slager - vlees [lijst 1]; slager – touw [lijst 2]). De snelheid waarmee proefpersonen de tweede lijst tot een bepaald criterium konden leren werd gebruikt als maat voor de vaardigheid om interferentie tegen te gaan en gerelateerd aan het aantal intrusieve herinneringen. Het aantal intrusies werd vervolgens gerelateerd aan de taakprestatie op een tweetal computertaken welke in eerder onderzoek model stonden voor de vaardigheid om een automatische respons te onderdrukken: de kleurwoorden Strooptaak en de random number generation taak. Bij deze laatste taak wordt van proefpersonen gevraagd om cijfers tussen de 1 en de 10 in een willekeurige volgorde op te noemen, 1 getal per seconde. Het is bij deze taak de kunst om een automatisch opkomende neiging om te gaan tellen (1, 2, 3…) tegen te gaan. Voortbouwend op de resultaten van studie 1 bleek een gebrekkige vaardigheid om interferentie uit het werkgeheugen te weren gerelateerd aan een groot aantal recente intrusies aan een stressvolle gebeurtenis uit het verleden. In overeenstemming met de voorspelling dat vooral de controle over het werkgeheugen van belang is voor de verklaring van intrusieve herinneringen was er geen verband tussen de frequentie van het aantal intrusies en de prestatie op de taken die model stonden voor de vaardigheid om een automatische response te onderdrukken. Wederom bleek de relatie tussen executieve controle en intrusieve herinneringen onafhankelijk van depressie, geslacht en het aantal maanden geleden dat de stressvolle gebeurtenis plaatsvond. Deze resultaten leveren aldus additionele ondersteuning voor de hypothese dat natuurlijk herstel of persistentie van intrusieve herinneringen na een stressvolle

levensgebeurtenis

(deels)

afhangt

van

een

specifieke

executieve

functie/vaardigheid. Dit is de vaardigheid om ongewenste informatie uit het werkgeheugen te weren.

Studie 3 (prospectief): Zelfgerapporteerde executieve controle als voorspeller van het aantal intrusieve herinneringen aan een stressvolle film De studies in hoofdstuk 2 en 3 onderzochten de relatie tussen executieve controle en intrusieve herinneringen op een moment in de tijd. Het valt op basis van die gegevens niet uit te sluiten dat juist vanwege de intrusies de executieve functies achteruit zijn gegaan i.p.v. dat

141

Samenvatting

gebrekkige executieve functies hebben geleid tot meer intrusies. Een belangrijke volgende stap is daarom om te toetsen of verschillen in executieve controle voorafgaand aan het trauma ook daadwerkelijk voorspellend zijn voor het al dan niet optreden van intrusieve herinneringen. Een belangrijke voorwaarde voor een dergelijke studie is dat individuele verschillen niet kunnen worden herleid tot verschillen in de traumatische ervaring. Een relevante ingang voor prospectief onderzoek naar de rol van executieve controle is het nabootsen van een traumatische gebeurtenis in een gecontroleerde analoge situatie. Uit eerder onderzoek blijkt dat het aanbieden van een stressvolle film als een analoog trauma een valide methode is om individuele verschillen in de frequentie van het aantal intrusieve herinneringen te onderzoeken (zie Holmes & Bourne, 2008 voor een overzicht). Het idee hierachter is dat de ervaring van een stressvolle film en de ervaring van een trauma qua impact en intensiteit sterk van elkaar verschillen, maar kwalitatief geen ander soort ervaringen zijn. Studie 3 was bedoeld als eerste stap om de invloed van pre “trauma” executieve controle op het in stand houden van intrusieve herinneringen te onderzoeken. In deze studie werd een stressvol filmfragment gebruikt als analoog trauma. Voorafgaand aan de presentatie van de stressvolle film vulden proefpersonen een vragenlijst in waarin gevraagd werd naar de vaardigheid om de aandacht te controleren. Voorbeelden: “Als ik me concentreer, kan ik mijn aandacht zo richten dat ik niet merk wat er om me heen gebeurt” (bijna nooit – altijd), “Ik kan snel van de ene naar de andere taak overschakelen” (bijna nooit – altijd), en “Als er een afleidende gedachte in me opkomt, kan ik mijn aandacht hier gemakkelijk van wegleiden” (bijna nooit – altijd). Na afloop van de film vulde de helft van de proefpersonen een dagboekje in om gedurende drie dagen plotseling opkomende herinneringen aan de film te noteren. Na afloop van de drie dagen kwamen ze nog een keer terug voor een tweede bijeenkomst, waar ze het dagboekje inleverden en een vragenlijst invulden die nogmaals vroeg naar het aantal intrusieve herinneringen aan de film. De andere helft van de proefpersonen vulden na het zien van de film geen dagboekje in en kwamen zonder verdere informatie na drie dagen terug om de intrusievragenlijst in te vullen. In lijn met de hypothese was er voor de dagboekgroep een verband te zien tussen de mate van aandachtscontrole voorafgaand aan de film en het aantal intrusieve herinneringen tijdens de tweede bijeenkomst. Dit verband trad echter niet op bij de controle groep die alleen een vragenlijst achteraf invulde. Dit resultaat suggereert het bestaan van een prospectieve relatie tussen executieve controle en intrusieve herinneringen na een stressvolle gebeurtenis.

142

Verder lijkt het gebruik van een dagboekje een sensitievere methode om natuurlijk en spontaan opkomende intrusies te registreren dan enkel een vragenlijst achteraf.

Studie 4 (prospectief): De executieve vaardigheid om ongewenste informatie uit het werkgeheugen te weren voorspelt het aantal intrusies n.a.v. een stressvolle film De resultaten van studie 3 waren in overeenstemming met het idee dat cognitieve controle prognostische waarde heeft voor het ontwikkelen van intrusieve herinneringen. Echter, de kleine groep proefpersonen en het feit dat executieve controle werd gemeten met een vragenlijst vraagt om replicatie van de resultaten in een grotere groep waarbij gebruik gemaakt wordt van een computertaak i.p.v. zelfrapportage. Deze huidige prospectieve studie richtte zich specifiek op de rol van individuele verschillen in de executieve functie/vaardigheid om ongewenste informatie uit het werkgeheugen te weren. De vraag was of een gebrekkige controle over het werkgeheugen voorafgaand aan de presentatie van een stressvolle film samenhangt met een relatief groot aantal dagboekintrusies in de 7 dagen na het bekijken van de film. Om individuele verschillen op deze executieve functie in kaart te brengen werd een variant gebruikt op de computertaak waarbij proefpersonen interfererende lijsten met woordparen moesten leren (zie studie 2/hoofdstuk 3). De prestatie op deze taak werd gerelateerd aan het aantal dagboekintrusies 7 dagen na de vertoning van de stressvolle film. De resultaten lieten een duidelijk verband zien dat in lijn was met de assumpties uit het intrusiemodel: executieve controle over het werkgeheugen voorafgaand aan een stressvolle gebeurtenis was een goede voorspeller van het aantal intrusieve herinneringen aan deze gebeurtenis. Dit gevonden effect bleek niet afhankelijk te zijn van additionele factoren die een bewezen invloed hebben op het in stand houden van intrusieve herinneringen na een traumatische situatie. Na statistische controle voor de invloed van depressie, geslacht, de impact van de stressvolle gebeurtenis (de film) en het wel of niet hebben ervaren van traumatische gebeurtenissen in het verleden, bleef executieve controle een goede voorspeller van het aantal intrusieve herinneringen.

143

Samenvatting

Studie 5 (prospectief): De rol van perceptuele herinneringscues bij het voorspellen van intrusieve herinneringen aan een stressvolle film In de studie die beschreven wordt in hoofdstuk 6 werd de focus verlegd naar een nog niet onderzochte component van het intrusiemodel. Er werd gekeken naar de invloed die herinneringscues in de externe omgeving kunnen hebben op het in stand houden van intrusieve herinneringen. De vraag was of het niet kunnen afwenden van de aandacht van perceptuele herinneringscues aan een stressvolle film voorspellend is voor het aantal filmgerelateerde intrusieve herinneringen in de week na het bekijken van deze film. De keus voor het selecteren van perceptuele cues was gebaseerd op de klinische observatie dat de meerderheid van de herinneringscues na een traumatische gebeurtenis een perceptueel karakter bezitten (zie het voorbeeld over het zonlicht). In de huidige studie werd gebruik gemaakt van een aandachtstaak waar een groep proefpersonen een snelle sequentiële reeks landschapsfoto’s op het computerscherm te zien kreeg (100 ms per foto). Tussen deze landschapsfoto’s zat 1 foto van een gebouw; deze was 90 graden geroteerd en fungeerde als doelfoto. Proefpersonen werd gevraagd om na iedere trial van 17 opeenvolgende foto’s de richting van de rotatie van het gebouw aan te geven (naar links of naar rechts). Dit konden ze doen met de pijltjes toetsten. De afhankelijke variabele bij deze taak was het aantal fouten in rotatiedetectie van de doelfoto’s. Dit aantal fouten werd vergeleken tussen twee condities. In de eerste conditie kwam er vlak voor het verschijnen van de doelfoto een afleidende foto in beeld. Deze afleidende foto bestond uit een neutraal gezicht of dier en week duidelijk af van de landschapsfoto’s die als achtergrond dienden. In de tweede conditie was er ook een afleidende foto te zien, maar nu bestond deze uit personen en objecten uit de eerder vertoonde stressvolle film. Het verschil in het aantal fouten tussen de trials waar film afleiders voorkwamen en de trials waar neutrale afleiders voorkwamen werd gebuikt als maat om problemen met het afwenden van de aandacht te bepalen. Het idee hierachter was dat de neiging om relatief veel fouten te maken op trials waar film afleiders voorkwamen (t.o.v. neutrale afleiders) model staat voor problemen met het loskoppelen van de aandacht van herinneringscues aan een eerdere stressvolle gebeurtenis. Deze verschilscore werd vervolgens gebruikt om het aantal intrusieve herinneringen aan de film te voorspellen. De resultaten waren in lijn met de eerdere verwachtingen: De neiging om relatief veel fouten in rotatiedetectie te maken na het verschijnen van een herinneringscue aan de stressvolle film hing sterk samen met het aantal dagboek herinneringen in de week na het bekijken van deze

144

film. Dit betekent dat het ongewild vasthouden van de aandacht op de stressvolle aspecten van herinneringscues voor een relatief hoge frequentie van het aantal intrusieve herinneringen kan zorgen.

Conclusie Dit proefschrift heeft zich beziggehouden met de vraag of een aangeboren verschil in controle over het werkgeheugen (i.e., mentale capaciteit) kan voorspellen of iemand na een traumatische gebeurtenis nog jarenlang geplaagd zal worden door steeds terugkerende intrusieve herinneringen. In lijn met het intrusiemodel uit het eerste hoofdstuk lieten de resultaten van de verschillende empirische studies een consistent patroon zien: Een gebrekkige executieve controle voorgaand aan een stressvolle gebeurtenis blijkt voorspellend voor het aantal intrusieve herinneringen aan deze gebeurtenis. De resultaten van deze analoge studies bieden interessante aanknopingspunten om de rol van premorbide executieve controle als kwetsbaarheidfactor verder te onderzoeken. Toekomstig onderzoek zou zich kunnen richten op de vraag of pretrauma executieve controle een goede voorspeller is van het cruciale verschil tussen natuurlijk herstel dan wel het ontwikkelen van PTSS na een traumatische gebeurtenis. Hiervoor zou een groep personen met een verhoogd risico om een trauma mee te maken (e.g., soldaten voorafgaand en na afloop van een uitzending) gedurende een bepaalde periode gevolgd kunnen worden.

145

Samenvatting

146

Dankwoord

147

Dankwoord

Dan is het nu tijd voor het laatste en ongetwijfeld meest gelezen gedeelte van dit proefschrift: Het dankwoord. Een aantal personen hebben de afgelopen vier jaar - direct dan wel indirect bijgedragen aan het schrijven van mijn proefschrift. Hen wil ik graag persoonlijk bedanken. Allereerst mijn copromotor en begeleider Ineke Wessel. Beste Ineke, na een wat stroeve start hebben we de juiste toon gevonden. Ik zie onze samenwerking als zeer vruchtbaar en wil je bedanken voor de dingen die ik de afgelopen jaren van je heb geleerd. Hoe moeilijk was het voor mij om kort en bondig te zijn tijdens het schrijven….. Dankzij jouw kritische blik en grondig commentaar is mijn schrijfvaardigheid met sprongen vooruit gegaan, wat heeft geleid tot een aantal mooie publicaties. Ik zie onze toekomstige samenwerking en discussies tijdens mijn post doc in Groningen met plezier tegemoet. Vervolgens mijn promotor Peter de Jong. Beste Peter, dank voor je enthousiasmerende en inspirerende houding, het grote aantal creatieve ideeën dat me steeds weer vooruitgeholpen heeft en het waanzinnig snelle commentaar op mijn stukken. Het is toch de droom van elke AIO om een ’s middags opgestuurd stuk de volgende ochtend al weer met uitgebreide suggesties in je mailbox terug te hebben. Heel veel dank hiervoor, ook voor je constante betrokkenheid gedurende de ups en downs van mijn promotie traject. Uiteraard wil ik ook een aantal woorden richten tot mijn directe collega’s, te beginnen met mijn kamergenoten. Esther en Sippie, bedankt voor de gezellige tijd en de goede gesprekken tijdens de eerste helft van mijn AIO-tijd. Complimenten voor het feit dat jullie het zo lang uithielden met een warhoofd en chaoot als ondergetekende…; ik kijk er met plezier op terug. Vervolgens een woord van dank voor de (voormalige) bezetting van kamer 315 – bij mijn weten de enige kamer met een officieel kameruitje. Charmaine (please, learn the Dutch language quickly, so that you can read my acknowledgements), Els, Klaske en Koen (Heineken rulezz!!), het was een bijzonder fijne tijd om met jullie een kamer te mogen delen. Heel veel dank hiervoor. Ik kijk nu al uit naar het volgende kameruitje. Klaske, bedankt voor de mentale steun gedurende de afrondende fase van het proefschrift. Uiteraard wil ik ook de rest van de (voormalig) AIO’s en collegae K & O bedanken voor de onmisbare support tijdens lunch, borrels, congressen etc etc. In alfabetische volgorde: Annemiek, Boukje, Corine, Daan, Esther S., Eva, Femke, Jessica, Jorg, Maartje, Madelon, Marieke, Martine, Rineke, Victorine en natuurlijk de vaste staf die hier niet bij name is genoemd. Een speciaal woord van dank wil ik reserveren voor Maurice Nieuwenhuis. Tijdens je studie en vooral gedurende het jaar dat je als vrijwilliger op onze afdeling werkte hebben we samen

148

vele, vele uren in het lab doorgebracht (250 - 300 proefpersonen getest!). Dank hiervoor. Ik wens je heel veel succes toe tijdens je nieuwe functie als promovendus; ik kom zeker een keertje langs voor een bakje koffie. Laat ik het secretariaat niet vergeten. Fenna, Ingrid, Josje, Lucia en Mirjam, heel veel dank voor het feit dat ik altijd bij jullie terecht kon; voor allerlei logistieke problemen, of gewoon voor een praatje. Bedankt ook alle EPP-promovendi voor alle gezellige momenten tijdens en vooral na afloop van de leerzame symposia. Tenslotte wil ik Bert Hoekzema en Pieter Zandbergen bedanken voor de onmisbare technische support tijdens het opzetten van de experimentele studies. Vriendenclub/koffieclub Waarder en omstreken, jullie namen mogen op deze plaats zeker niet ontbreken. Harry, Jantiene en “lieve” Lydia, bedankt voor de open deur, de vele kopjes koffie en de gezellige avonden aan de speltafel. Laten we ze vooral in ere houden. Evert-Jan en Dijanne, Dick en Franka, Jan en Cora, Anton en Dikkie, Rien en Marijke, Wolter en Marja en Johan Bos, allen hartelijk dank voor de morele support bij het schrijven van mijn proefschrift. Rien, wat gaaf dat je als paranimf naast me wilt staan tijdens mijn verdediging! Bouqie. Een simpel woord met een rijke inhoud: Nikolai, Piotr en Vladimir (Matthijs, Martijn en Frederik), wij 4-en vieren dit jaar ons eerste lustrum. Bedankt voor het samen beleven van literatuur, goede muziek en de zuivere smaak van sherrie. Boven alles staat echter onze vriendschap. Ik hecht hier erg veel waarde aan en hoop dat deze nog jarenlang voort mag duren. Frederik, bedankt voor je deskundige advies bij het ontwerpen van de omslag. Matthijs, bedankt dat je ook als paranimf bij mijn verdediging aanwezig wilt zijn. Ik wil natuurlijk ook mijn familie bedanken, in het bijzonder mijn moeder, broer, schoonzusje en de kleine Thomas. Mark, je brede algemene ontwikkeling maakt dat ik veel van je geleerd heb over wetenschap, literatuur, filosofie, religie en de betere films. Ik hoop dat we onze lange gesprekken ook in de toekomst een plaats kunnen geven. Moeder en schoonzusje, bedankt dat jullie altijd voor mij klaar willen staan, zowel dichtbij als op afstand. Het belang hiervan kan niet genoeg onderstreept worden. Laat ik afsluiten met die personen wier namen niet in dit dankwoord voorkomen, maar die toch hebben bijgedragen aan mijn morele en intellectuele ontwikkeling als mens en wetenschapper. Misschien zijn we elkaar uit het oog verloren, misschien spraken we elkaar slechts kortstondig op een congres, op vakantie, tijdens mijn vorig leven in de sport, of waar dan ook ter wereld. Slechts een klein aantal van jullie zal deze woorden daadwerkelijk lezen,

149

Dankwoord

wat niet wegneemt dat ik een ieder hartelijk wil bedanken voor het feit dat ik voor kortere of langere tijd lief en leed met jullie heb mogen delen. Zonder jullie had ik nooit die keuzes kunnen maken die uiteindelijk hebben geleid tot de totstandkoming van dit proefschrift.

Johan Groningen, januari 2009

150

Curriculum Vitae

Johan Verwoerd was born in Woerden on April 5th, 1975. After being trained as an instructor in the field of sports and recreation, he started his studies of psychology in September 1999. In February 2004 he received his Master’s degree in cognitive psychology at Leiden University under supervision of Dr. Gezinus Wolters. His Master’s thesis described the role of perceptual and conceptual priming in prospective memory tasks. In September 2004 he moved to Groningen for a PhD at the department of Clinical and Developmental psychology under supervision of Dr. Ineke Wessel and Prof. dr. Peter de Jong. This resulted in the present dissertation. Since Oktober 2008, Johan has extended his work in Groningen as a post doctoral researcher.

Publications Verwoerd, J., Wessel., I., de Jong, P. J., & Nieuwenhuis, M. M. W. (in press). Preferential processing of visual trauma-film reminders predicts subsequent intrusive memories. Cognition and Emotion. Verwoerd, J., Wessel, I., & de Jong, P. J. (in press). Individual differences in experiencing intrusive memories: The role of the ability to resist proactive interference. Journal of Behavior Therapy and Experimental Psychiatry. Wessel, I., Overwijk, S., Verwoerd, J., & de Vrieze, N. (2008). Pre-stressor cognitive control is related to intrusive cognition of a stressful film. Behaviour Research and Therapy, 46, 496513. Verwoerd, J., de Jong, P. J., & Wessel, I. (2008). Low attentional control and the development of intrusive memories following a laboratory stressor. Journal of Psychopathology and Behavioural Assessment, 30, 291-297. Verwoerd, J. & Wessel, I. (2007). Distractibility and individual differences in the experience of intrusive memories. Personality and Individual Differences, 42, 325-334.

Verwoerd, J., Wessel., I., de Jong, P. J., & Nieuwenhuis, M. M. W. (2008). Pre-stressor interference control and intrusive memories. Manuscript submitted for publication.

151

Curriculum Vitae

152

153

154