writing project

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=pmem20

Memory

ISSN: 0965-8211 (Print) 1464-0686 (Online) Journal homepage: https://www.tandfonline.com/loi/pmem20

Inhibitory control of threat remembering in PTSD

Andrei-Cristian Tudorache, Wissam El-Hage, Géraldine Tapia, Nelly Goutaudier, Sandrine Kalenzaga, Badiâa Bouazzaoui, Nemat Jaafari & David Clarys

To cite this article: Andrei-Cristian Tudorache, Wissam El-Hage, Géraldine Tapia, Nelly Goutaudier, Sandrine Kalenzaga, Badiâa Bouazzaoui, Nemat Jaafari & David Clarys (2019) Inhibitory control of threat remembering in PTSD, Memory, 27:10, 1404-1414, DOI: 10.1080/09658211.2019.1662053

To link to this article: https://doi.org/10.1080/09658211.2019.1662053

Published online: 05 Sep 2019.

Submit your article to this journal

Article views: 219

View related articles

View Crossmark data

Citing articles: 1 View citing articles

Inhibitory control of threat remembering in PTSD Andrei-Cristian Tudorachea, Wissam El-Hageb, Géraldine Tapiac, Nelly Goutaudiera, Sandrine Kalenzagaa, Badiâa Bouazzaouia, Nemat Jaafarid and David Clarysa

aCentre de Recherches sur la Cognition et l’Apprentissage, UMR CNRS 7295, Universite ́ de Poitiers, Université de Tours, Poitiers, Tours, France; bUMR 1253, iBrain, Université de Tours, CHRU de Tours, Inserm, Tours, France; cLaboratoire de Psychologie EA4139, Université de Bordeaux, Bordeaux, France; dUnité de recherche clinique intersectorielle en psychiatrie à vocation régionale Pierre Deniker du Centre Hospitalier Henri Laborit, Inserm CIC-P 1402, Inserm U 1084 Experimental and Clinical Neurosciences Laboratory, Université de Poitiers, CHU Poitiers, Groupement De Recherche CNRS 3557, Poitiers, France

ABSTRACT Intrusive traumatic recollections suggest an inability in Posttraumatic Stress Disorder (PTSD) to control and notably to inhibit memories for trauma-related information. Supported by inhibitory deficits found on experimental settings in PTSD, memory functioning and memory biases in the disorder were usually explained through inhibitory and control deficits in the processing of trauma-related information. The present study aimed to directly assess this hypothesis by investigating memory control abilities for emotional information in PTSD. For this purpose, 34 patients diagnosed with PTSD were compared to 37 non-PTSD controls on an item-cued directed forgetting paradigm for emotional words combined with a Remember/Know recognition procedure. Results revealed enhanced amounts of Remember recognitions for trauma-related words in PTSD. Moreover, we replicated findings of memory control impairments in the disorder. However, such impairments only occurred for non-trauma- related words. Accordingly, it appeared that PTSD patients presented preserved memory control abilities for trauma-related words, at the expenses of other emotional valences. Surprisingly, PTSD patients presented a preserved ability to control and notably to inhibit their memory functioning for trauma-related material. In addition to potential theoretical and clinical relevance, these results are discussed in the light of resource reallocation hypotheses and vigilant-avoidant theories of information processing in PTSD.

ARTICLE HISTORY Received 1 March 2019 Accepted 25 August 2019

KEYWORDS Posttraumatic Stress Disorder (PTSD); memory; inhibition; emotion; remember/know; directed forgetting

The constant re-experiencing of traumatic events through involuntary and uncontrollable intrusive reminiscences is one of the most salient and distressing symptoms of Post- traumatic Stress Disorder (PTSD; Diagnostic and statistical manual of mental disorders 5th ed., DSM-5; American Psy- chiatric Association, APA, 2013). Among other specificities, the strong emotional and sensorial reliving, the lack of control or the enhanced vividness (i.e., flashbacks) dis- tinguish intrusive symptoms from other types of memories (APA, 2013; Brewin, Gregory, Lipton, & Burgess, 2010; Ehlers & Clark, 2000; Foa, Molnar, & Cashman, 1995; Harvey & Bryant, 1999). Conceived as a central factor not only in the distress caused by the disorder, but also in the develop- ment and maintenance of symptoms (e.g., Brewin et al., 2010; Ehlers & Clark, 2000; Rubin, Berntsen, & Bohni, 2008), memory functioning in PTSD inspired numerous amounts of studies in the past decades. However, it is important to note that, despite large quantities of data in this area, investigations of memory functioning based on standardised memory tests have yielded inconsistent findings, reporting alternately impaired or preserved memory skills in PTSD (see review by Tapia, Clarys, El-

Hage, & Isingrini, 2007). Nevertheless, some interesting results have emerged from experimental approaches in which memory is assessed in conjunction with other fea- tures of PTSD symptomatology, such as emotional proces- sing (Golier, Yehuda, Lupien, & Harvey, 2003), attention allocation (Paunovic, Lundh, & Öst, 2002), states of con- sciousness (Tapia, Clarys, Bugaiska, & El-Hage, 2012), and executive functioning (Cottencin et al., 2006).

Among these features, emotional processing was largely questioned in PTSD. Regarding memory, the highly distressing emotional reliving associated with trauma memories drew attention to the interactions between emotions and memory functioning in PTSD, as highlighted in the review by Tapia and colleagues (2007). In studies conducted in community samples, manipulation of the emotional valence of processed material usually highlighted a beneficial effect for memory performance of emotionally salient stimuli compared with neutral ones (see review by Hamann, 2001). However, transposed to PTSD patients, the study of emotional processing has revealed biases in favour of threatening information specifically. While performing poorly for neutral and

© 2019 Informa UK Limited, trading as Taylor & Francis Group

CONTACT Andrei-Cristian Tudorache [email protected] Centre de Recherches sur la Cognition et l’Apprentissage, Maison des Sciences de l’Homme et de la Société, CNRS UMR 7295 - Université de Poitiers, 5 rue Théodore Lefebvre, TSA 21103, F-86073, Poitiers CEDEX 9, France

MEMORY 2019, VOL. 27, NO. 10, 1404–1414 https://doi.org/10.1080/09658211.2019.1662053

positive stimuli, PTSD patients demonstrated improved memory performance for negative or trauma-related stimuli compared with non-PTSD controls (e.g., Golier et al., 2003; Lin, Hofmann, Qian, & Li, 2015; Moradi, Taghavi, Neshat-Doost, Yule, & Dalgleish, 2000; Tapia et al., 2012; Thomas, Goegan, Newman, Arndt, & Sears, 2013; Zeitlin & McNally, 1991).

In addition to their emotional intensity, traumatic remi- niscences can become markedly vivid and conscious to the point that patients can have the impression that the event is occurring at the moment (APA, 2013; Ehlers & Clark, 2000). While the involuntary, unorganised and fragmented aspects of intrusive symptoms led cognitive models of PTSD to posit deficits in memory contextualisation (e.g., Brewin et al., 2010; Ehlers & Clark, 2000), the “here and now” quality suggest contextualisation could otherwise reach unusually high levels for specific aspects of trauma memories. Although rarely investigated in PTSD studies, this consideration echoes Tulving’s (1985) proposal to sep- arate the declarative memory system into episodic and semantic components based on the states of conscious- ness associated with memory. Whereas the noetic con- sciousness refers to the awareness of knowing and reflects the semantic memory system, the autonoetic con- sciousness, associated with episodic memory, implies the ability to “travel back in time”, getting the impression of reliving the encoding of an event. Accordingly, semantic memory is considered a stock of general and uncontextua- lized knowledge about the world and environment (e.g., the knowledge of capital cities), whereas episodic memory is a stock of highly contextualised personally experienced events (e.g., the remembering of your last holiday), including notably emotional and perceptual details. Based on auto-evaluations of their associated states of consciousness, the Remember/Know (R/K) para- digm was designed for experimental assessments of these two memory forms (Tulving, 1985). This paradigm respectively reflects autonoetic and noetic states of con- sciousness by distinguishing a Remember (R) response type when a stimulus is retrieved with contextual details from its encoding (e.g., associated items, memories, images, feelings), and a Know (K) response type when no contextual information could be provided.

As far as the characteristics of PTSD intrusive memories are concerned, the R/K paradigm has initially been used to assess the hypothesis of general impairments in contextua- lising memory acquisitions. A study conducted in this frame- work confirmed this hypothesis by revealing a specific deficit in R recognitions for neutral words in PTSD partici- pants as compared to un-traumatized control ones. In con- trast, K responses were preserved (Tapia, Clarys, El Hage, Belzung, & Isingrini, 2007). In another study, Tapia et al. (2012) compared PTSD to un-traumatized anxious/depress- ive participants and healthy control ones on a R/K recog- nition procedure of positive, neutral or negative words. Interestingly, this study highlighted that recognition of negative stimuli was accompanied by R responses to a

greater extent in the PTSD group compared with the control sample (Tapia et al., 2012). Suggesting that recollec- tion of negative stimuli in PTSD could be associated with enhanced reliving of the memory acquisition, this result emphasises the relevance of the R/K paradigm in exper- imental approaches of intrusive symptoms. However, it is interesting to note that there were not statistically signifi- cant differences when PTSD patients were compared with anxious and depressive patients. Moreover, the negative stimuli used in this study encompassed both trauma- related stimuli (e.g., accident, weapon, death) and non- trauma-related ones (e.g., sadness, tiredness, delirium). It is thus unclear whether or not qualitatively improved access to negative information reflects a general negativity bias in PTSD due to anxious or depressive comorbidities, or a specific preferential processing of reminders of traumatic experiences. Focusing on this issue appears to be critical for interpretation of R/K findings in PTSD.

Accordingly, the present study sought to replicate Tapia et al. (2012) methodology, using a R/K paradigm to dis- tinguish between the encoding and recognition of positive, neutral, non-trauma-negative and trauma-related words. In addition to examination of emotional biases in PTSD memory functioning, a second aim was to evaluate the cog- nitive mechanisms underlying such biases. In this regard, both the inability of PTSD patients to inhibit their trauma- related thoughts and experimental investigations of infor- mation processing in PTSD suggested the presence of inhibitory deficits in the processing of negative or trauma- related information (e.g., El Khoury-Malhame et al., 2011; Fleurkens, Rinck, & van Minnen, 2011; Paunovic et al., 2002). Mainly investigated through Emotional Stroop tasks, inhibitory deficits on experimental settings were translated by enhanced interferences in colour naming in the presence of negative or trauma-related stimuli (see for instance the meta-analysis by Cisler et al., 2011). These inhibitory deficits may appear disjointed from clinical inhibi- tory deficits in preventing trauma-related intrusions and re- experiencing. However, some studies revealed that inhibi- tory or control deficits on cognitive tasks specifically corre- lated with re-experiencing symptoms (e.g., Bomyea, Amir, & Lang, 2012; Vasterling, Brailey, Constans, & Sutker, 1998). As a consequence, such inhibitory deficits became common explanations of memory biases in PTSD. However, studies focusing on this hypothesis highly diverge regarding both findings and methodological choices. Moreover, to our knowledge, few studies have experimentally investigated memory control abilities for emotional material in PTSD and, to date, none have jointly examined the distinction between R or K memories.

Among methodological choices, the directed forgetting task appeared highly promising for memory control assess- ments. Conceived as a classical memory task, the directed forgetting procedure differed in that instructions to mem- orise or inhibit the memorisation are given after each item (item method) or after a list of items (list method). Prior research has shown that healthy participants completing

MEMORY 1405

this paradigm often display a directed forgetting effect, illustrated by worse memory performance for “to-be-for- gotten” items than for “to-be-remembered” ones (MacLeod, 1999). Regarding PTSD, studies have found a reduced directed forgetting effect, suggesting that PTSD could be associated with memory inhibition deficits (Cot- tencin et al., 2006; Zwissler et al., 2012). Interestingly, McNally, Metzger, Lasko, Clancy, and Pitman (1998) com- pared participants with and without PTSD on a directed forgetting paradigm using positive versus neutral versus trauma-related words. Contrary to their expectations, PTSD participants tended to present a directed forgetting effect only for trauma-related words. However, rather than reflecting an improved ability to inhibit the encoding of threatening stimuli, this result was mainly due to an enhanced recall in PTSD participants for trauma-related words “to-be-remembered” compared with other types of words. Furthermore, no differences for recall of trauma- related words were observed between groups. Considering these prior findings, results remain unclear regarding memory control and inhibitory abilities in PTSD.

In sum, the present study aimed to investigate the role played by inhibitory deficits in the ease to access R recog- nition during retrieval of threatening information in PTSD. Our methodology consisted of a novel item-cued directed forgetting task for emotional words combined with a R/K recognition procedure. In light of prior studies, we antici- pated replication of Tapia et al.’s (2012) findings of an increased production of R recognitions for threatening words in the “to remember” encoding condition. However, earlier studies have used negative words regardless of their traumatic relevance. In the present study, we hypoth- esised that that effects are not related to a general nega- tivity bias per se, but mainly to the fact that a significant portion of the words used could be trauma relevant. In line with studies suggesting a specific sensitivity in PTSD towards trauma-related material (e.g., Stanford, Vasterling, Mathias, Constans, & Houston, 2001), we expected that the PTSD group would present increased R recognitions during retrieval of trauma-related words specifically.

Secondly, we wanted to assess the extent to which inhibi- tory deficits were associated with memory biases. Given that PTSD patients both presented inhibitory deficits in the processing of trauma-related information and in pre- venting the onset of their traumatic memories, we hypoth- esised decreased ability to succeed in the queries of the “to forget” encoding condition. Consequently, directed forget- ting effects should be limited to non-trauma related words in the PTSD group as a result of an increased production of R recognitions for trauma-related words “to be forgotten”. Finally, consistent with Tapia et al.’s (2012) findings, we pre- dicted K recognitions to be unaffected in PTSD.

Materials and methods

Participants

Our sample was composed of 34 individuals diagnosed with PTSD (25 women and 9 men) and 37 non-PTSD control sub- jects (25 women and 12 men). Participants in the PTSD group were treatment-seeking patients of French university hospitals. PTSD symptoms, as defined by the DSM-IV (APA, 1994), were measured using the Clinician Administered PTSD Scale (CAPS; Blake et al., 1995). Their control counter- parts were recruited through university mailing lists. Only respondents matching patients’ ages, years of education, and with no traumatic experiences on the Life Events Check- list (Gray, Litz, Hsu, & Lombardo, 2004) were included. For both groups, exclusion criteria were: the presence of other comorbid psychiatric diagnoses (e.g., bipolar disorder, sub- stance abuse, psychotic-related disorders), a main diagnosis of Major Depressive Disorder, newly introduced medi- cations (i.e., <1 month), and/or history of head injury. PTSD patients were included regardless of the type of trauma(s) experienced (most occurrences were related to sexual or physical assaults, relative’s death by suicide, ill- nesses, or severe injuries). Most of the individuals in the PTSD group experienced multiple traumatic experiences. Groups differed significantly on depressive symptoms, assessed with the Montgomery and Asberg Depression Rating Scale (MADRS; Montgomery & Åsberg, 1979), disso- ciative symptoms, assessed with the Dissociative Experi- ence Scale Taxon (DES-T, Waller, Putnam, & Carlson, 1996), and anxiety symptoms, assessed with the State Trait Anxiety Inventory (STAI; Spielberger, 1983). Participants’ characteristics are presented in Table 1. Participation in the study was voluntary, and all participants gave their informed written consent. Recruitment and testing pro- cedures received the approval of the psychiatrists in charge of our PTSD participants and were in accordance with ethical guidelines of the declaration of Helsinki.

Measures

The Life Events Checklist is a 17-item scale assessing for exposure to most commonly encountered traumatic events. While the first 16 items present most commonly

Table 1. Means and standard deviations of age, yeas of education and clinical scales’ scores for the two groups.

PTSD Controls

F(1,69)

(n = 34) (n = 37)

M SD M SD

Age (years) 35.24 17.13 34.12 16.36 <1 Educational level (years) 12.80 1.69 12.00 1.89 <1 CAPS total 51.64 15.95 – – – STAI-S 56.20 23.33 26.66 6.85 22.11*** STAI-T 63.20 21.10 36.13 9.86 20.24*** MADRS 18.73 10.66 3.60 6.29 22.40*** DES-T 37.13 41.39 3.44 4.79 9.80**

Note: **p < .01; ***p < .001; M: mean; SD: standard deviation. The Clinician Administered Posttraumatic Stress disorder Scale (CAPS) is from Blake et al. (1995); The State Trait Anxiety Inventory (STAI-S; STAI-T) is from Spiel- berger (1983); The Montgomery and Asberg Depression Rating Scale (MADRS) is from Montgomery and Åsberg (1979); The Dissociative Experi- ence Scale Taxon (DES-T) is from Waller et al. (1996).

1406 A.-C. TUDORACHE ET AL.

experienced events, the last item allows respondents to add other potential traumatic events not listed before. Each item is rated on a 4-point scale (1 = happened to me, 2 = witnessed it, 3 = learned about it, and 4 = does not apply). Gray et al. (2004) reported overall adequate psycho- metric properties.

The CAPS (Blake et al., 1995) is a largely used structured interview for PTSD diagnosis and symptoms assessment. The version used in this study, encompasses, through 17 items, the DSM-IV diagnostic criteria for PTSD. Weathers, Keane, and Davidson (2001) reported overall adequate psy- chometric properties.

The MADRS (Montgomery & Åsberg, 1979) is a 10-item structured interview forassessmentof depressive symptoms (i.e., apparent sadness, reported sadness, inner tension, reduced sleep, reduced appetite, concentration difficulties, lassitude, inability to feel, pessimistic thoughts, suicidal idea- tions). Symptom severity is rated on a 7-point scale ranging from 0 to 6. Total score range from 0 to 60 with high scores indicating elevated anxiety and depressive symptoms and a cut-off score of 20 as an indicator of probable depression. Psychometric properties have been reported to be generally good to excellent (Bondolfi et al., 2010).

The DES-T (Waller et al., 1996) is an 8-item self-admini- strated measure of pathological dissociation. For each item, respondents have to rate the frequency of experien- cing the dissociative state described (from 0% = Never to 100% = Always). The total score is usually obtained by aver- aging percentages obtained on individual items with high percentages indicating elevated dissociation. Waller et al. (1996) estimated adequate reliability.

The STAI (Spielberger, 1983) is composed of two 20-item self-administrated questionnaires assessing anxiety symp- toms expressed at the moment (STAI State) and general anxiety (STAI Trait). Respondents rate their experiences on 4-point Likert scales (STAI State: 1 = no at all, 4 = very much; STAI Trait: 1 = almost never, 4 = almost always). High scores indicate elevated anxiety. Barnes, Harp, and Jung (2002) reported overall good psychometric properties.

Materials

Memory was assessed using 80 French words divided into positive, neutral, negative and trauma-related according to French databases (Bonin et al., 2003; Messina, Morais, & Cantraine, 1989; Vikis-Freibergs, 1976). Since multiple trauma experiences were expected, trauma-related words were selected in order to reflect mostly encountered trau- matic events according the authors’ clinical experiences with PTSD patients. Additionally, imagery and number of syllables were controlled using BRULEX lexical data (Content, Mousty, & Radeau, 1990). The 80 selected words were divided into two 40-words lists, both composed of 10 positive, 10 neutral, 10 negative and 10 trauma-related words. The lists were counterbalanced between partici- pants to serve alternatively as target words for the encoding task or as distracter words for the recognition task. Each list

was divided into two 20-words sub-lists composed of five words per emotional valence. These counterbalanced sub- lists constituted the target words for the “to remember” and “to forget” encoding conditions associated with the directed forgetting paradigm. Finally, the recognition task was operationalised using a random presentation of the 40 studied words and the 40 distracter words.

Procedure

Prior to the study, potential participants were provided with general information regarding the study aims and methodology. After giving their written consent, they were administered the DES, CAPS, MADRS and STAI scales.

The directed forgetting protocol was computerised using E-Prime 2.0 software (Psychology Software Tools Inc., Schnei- der, Eschman, & Zuccolotto, 2002). Each memory phase was preceded by a training trial composed of six unstudied words for the encoding phase, completed by six unstudied distrac- tors for the recognition phase. After familiarisation, the test phase started, preceded by the same instructions. For the encoding phase, each word appeared for 3 s, followed by a 3 spresentationofthe“toremember”or“toforget” encoding instructions. The general instructions specified to memorise the words when they were followed by “to remember” and to inhibit their memorisation when they were followed by “to forget”. The four minutes encoding phase was separated from the recognition phase by a five minutes retention interval.

After this interval, participants completed the R/K recog- nition procedure on the 40 studied words and the 40 unstu- died distracters. Instructions were once again provided both on the screen and orally. The words were randomly pre- sented. Participants were asked to press the “yes” key if they had already seen the word, regardless of the encoding instructions, and the “no” key if they did not remember seeing the word. If the “yes” key was pressed, participants were invited to report the state of consciousness associated with the recognition by pressing the “A”, “B” or “C” key. Neutral terminology of “A”, “B” and “C” rather than “R”, “K” or “Guess (G)” was chosen according McCabe and Geraci’s (2009) recommendations. Instructions specified that an “A” answer implies a specific remembering of the word in which contextual details of its presentation during encod- ing phase could be recalled (including notably associated personal memories, images, or inter-item associations), whereas a “B” answer implies a strict remembering of the word without retrieval of any contextual details. Finally, the “C” answer was designed in order to avoid unreliable “B” responses by assessing uncertainty about the presence of the word during encoding (Gardiner & Conway, 1999). Regarding the R/K paradigm, “A”, “B” and “C” responses respectively represent R, K and G recognitions. In order to ensure the accurate categorisation, participants were invited to justify their first three R answers by reporting the images, thoughts, memories or feelings associated with the recognised word at the encoding. Conversely, for

MEMORY 1407

the first three K recognitions, participants were invited to confirm that no such information was elicited by, or associ- ated with, the word at the time of encoding.

Analyses

Dependent variables were the overall proportion of recog- nised words (number of recognised words/number of words per category) and the proportion of recognised words labelled with a R recognition for each experimental condition (R responses/number of words per category). In line with the Independent R/K (IRK) procedure (Yonelinas & Jacoby, 1995), K recognitions were computed indepen- dently from R recognitions by dividing the total of K responses by the number of trials without R responses [K responses/(number of words per category - R responses)]. Given that G recognitions were only designed for improv- ing the quality of K responses and that their proportion was judged to be too low, they were not analysed.

In order to investigate directed forgetting effects, three 2 (Group: PTSD versus controls) × 2 (Encoding: to remember versus to forget) × 4 (Valence: positive versus neutral versus negative versus trauma-related) analyses of variance (ANOVA) were conducted on the total of recognised words, R responses, and IRK scores. Addition- ally, in order to investigate the memory performance and the memory inhibition abilities of the participants, R responses and IRK scores were analysed separately by a 2 (Group) × 4 (Valence) ANOVA for each Encoding. Overall, the Group was considered as a between-groups factor, whereas the Valence and the Encoding were con- sidered as within-groups factors. Analyses were con- ducted using Dell Statistica 12 (Dell Inc., 2015). Given that high correlations between clinical scales (DES, STAI, MADRS, CAPS) were found (Pearson’s r ranging from .64 to .77) and there were no significant covariant effects of the DES, STAI and MADRS scales, covariant analyses are not reported.

Results

Directed forgetting effects on total recognition scores

The analyses conducted on total recognition scores by Group, Valence, and Encoding revealed no significant Group effect, F(1,69) < 1, indicating that the two groups dis- played the same recognition pattern. Consistent with the classical directed forgetting effect, there was a significant effect of Encoding, F(1,69) = 56.31, p < .001, ηp² = .45, indi- cating a decreased performance for the items “to-be-for- gotten” compared with the items “to-be-remembered”. Moreover, there was a significant interaction between Group and Encoding, F(1,69) = 12.18, p < .001, ηp² = .15. Contrast analyses revealed that the difference between the words “to-be-remembered” and the words “to-be- forgotten” was reduced in the PTSD group, F(1,69) = 7.73,

p = .007, ηp² = .09, compared to the control group, F(1,69) = 63,11, p < .001, ηp² = .48, suggesting a reduced directed forgetting effect in PTSD. No other effect was significant.

Directed forgetting effects on R responses

The analysis conducted on R responses by Group, Encoding, and Valence revealed no significant effect of Group, F(1,69) = 2.19, p = .143, ηp² = .03, suggesting that control participants and PTSD patients produced similar amounts of R recognitions. However, there was a significant effect of the Encoding, F(1,69) = 44.92, p < .001, ηp² = .37, reflect- ing the classical directed forgetting effect (i.e., more R responses for the “to-be-remembered” encoding condition than for the “to-be-forgotten” condition). More- over, the significant interaction between Encoding and Group, F(1,69) = 6.21, p = .015, ηp² = .05, revealed a reduced directed forgetting effect in the PTSD group, F(1,69) = 12.83, p = .001, ηp² = .28, compared with the control group, F(1,69) = 33.71, p < .001, ηp² = .48. No signifi- cant interaction was found between Valence and Group, F(3,207) = 1.24, p = .29, ηp² = .02, nor between Valence and Encoding, F(3,207) < 1. There was a significant Valence effect, F(3,207) = 3.06, p = .029, ηp² = .04, and a significant interaction between the three variables, F(3,207) = 4.21, p = .006, ηp² = .06. Contrast analyses revealed that control participants presented a significant effect of the Encoding for all emotional valences. More precisely, R recognition was higher in the “to remember” condition compared with the “to forget” condition for positive F(1,69) = 14.56, p < .001, ηp² = .28, neutral F(1,69) = 23.87, p < .001, ηp² = .43, negative F(1,69) = 28.15, p < .001, ηp² = .44, and trauma-related words F(1,69) = 6.47, p = .013, ηp² = .28. However, PTSD patients only presented this directed for- getting effect for trauma-related words F(1,69) = 18.76, p < .001, ηp² = .37, while neither positive F(1,69) < 1, neutral F(1,69) = 3.18, p = .08, ηp² = .08 nor negative words F(1,69) = 1.39, p = .243, ηp² = .04 reached significance. Find- ings are presented in Figure 1.

Directed forgetting effects on IRK scores

The analysis conducted on IRK scores by Group, Encoding, and Valence revealed no significant effect of Group, F(1,69) = 1.64, p = .20, ηp² = 02, suggesting that PTSD and control participants produced similar amounts of IRK recog- nitions. The significant effect of the Encoding reflects the classical directed forgetting effect, F(1,69) = 10.98, p = .001, ηp² = .12. Moreover, there was a significant interaction between Encoding and Group, F(1,69) = 8.62, p = .005, ηp² = .10. Contrast analyses revealed that only control par- ticipants presented a directed forgetting effect on IRK scores F(1,69) = 20.39, p < .001, ηp² = .35, whereas PTSD patients did not, F(1,69) < 1. No other effect was significant. Means andstandarddeviationsarepresentedinAppendix A.

1408 A.-C. TUDORACHE ET AL.

Memory performance

The analysis conducted on R recognitions for the words “to-be-remembered” revealed no significant Group effect, F(1,69) < 1, suggesting that overall the two groups produced the same amount of R recognitions. Moreover, while no sig- nificant Valence effect was found, F(3,207) = 1.51, p = .212, ηp² = .02, there was a significant interaction between Group and Valence, F(3,207) = 3.71, p = .012, ηp² = .05. Con- trast analyses revealed that PTSD and control participants only differed on the recognition of trauma-related words, with PTSD patients producing significantly more R recog- nitions than control participants for trauma-related words, F(1,69) = 5.65, p = .020, ηp² = .07, but not for positive, neutral, or negative words, all Fs(1,69) < 1.

The analysis of IRK scores for the words “to-be-remem- bered” revealed a significant Group effect, F(1,69) = 4.69, p = .034, ηp² = .06, indicating that the control group produced more K recognitions than the PTSD group. There was neither Valence effect, F(3,207) < 1, nor interaction between Group and Valence, F(3,207) = 1.98, p = .119, ηp² = .03.

Memory inhibition

Analyses conducted on the words “to-be-forgotten” revealed that PTSD patients produced significantly more R recognitions for these words than their control counterparts, F(1,69) = 5.95, p = .017, ηp² = .08. There was no significant Valence effect, F (3,207) = 2.28, p = .080, ηp² = .03, nor a significant interaction between Valence and Group, F(3,207) = 1.17, p = .32, ηp² = .01. Finally, there was no significant effect on IRK scores.

Discussion

Aiming to assess memory control abilities in PTSD, the current study was led by the hypothesis that PTSD patients

would demonstrate increased R recognitions for threaten- ing material as a result from an inability to successfully inhibit the encoding of trauma-related information.

First, regarding memory functioning, our findings revealed that, overall, PTSD patients recognised the same amount of words as healthy controls. While in line with some prior research (e.g., Tapia et al., 2012; Vrana, Roodman, & Beckham, 1995), this result otherwise diverges from other findings (e.g., Golier et al., 2003; Moradi et al., 2000). Explanations of these discrepancies might emerge from methodological considerations. Indeed, most studies having revealed memory impairments in PTSD used free recall rather than recognition procedures. Accordingly, present findings are in line with prior studies emphasising that, in comparison to free recalls, memory deficits in PTSD may be reduced on cued recalls or recognition procedures (e.g., Jenkins, Langlais, Delis, & Cohen, 1998; Zoellner, Sacks, & Foa, 2003). In light of the inconsistent outcomes regarding this issue (see review by Tapia, Clarys, El-Hage, et al., 2007), it is difficult to conclude that memory functioning is unaffected by PTSD. Moreover, although no overall impairments were reported, it is impor- tant to note that PTSD patients produced more R recog- nitions than controls for trauma-related words specifically. Consistent with our hypothesis, this result suggests that the enhanced R recognition found by Tapia et al. (2012) for negative words might reflect the high trauma-relevance of the negative stimuli used, rather than a general negativity bias.

Typically, cognitive theories conceived emotional biases as factors of maintenance for many disorders (e.g., Moradi et al., 2000). Describing a vicious cycle from negative emotions to emotional biases, such theories could account for memory functioning in PTSD. Thus, the high prevalence of intrusive traumatic memories would favour processing and notably the encoding of trauma-related

Figure 1. Difference in proportions (and standard error bars) of R responses between the “to remember” and “to forget” conditions (i.e., directed forgetting effect) by group and emotional valence.

MEMORY 1409

material, which could in turn elicit the memories of trauma. Considering that only R recognitions were affected, memory biases in PTSD could be translated by enhanced associations between trauma-related materials and contex- tual information. In this view, the constant re-experiencing of personal traumatic memories and associated features (e.g., emotions, sensations, images) could have provided numerous associative cues for trauma-related stimuli, con- tributing to their improved encoding and increasing the probability to obtain R responses during recollection. Additionally, trauma-related stimuli might themselves trigger personal traumatic memories (e.g., Long, Chamber- lain, & Vincent, 1994; Williams, Mathews, & MacLeod, 1996). In line with cognitive models of PTSD (e.g., Ehlers & Clark, 2000), current findings suggest that PTSD patients easily associate environmental stimuli with their own traumatic experience, resulting in memory improvements, and poss- ibly, in return, reinforcing their own traumatic memories.

However, statements of enhanced memory functioning for trauma-related material in PTSD should be taken care- fully regarding the non-significant differences between emotional valences within the PTSD group. Nevertheless, PTSD patients only presented a directed forgetting effect for R recognitions of trauma-related words. Supporting the findings of McNally et al. (1998) and the hypothesis of biased information processing towards trauma-related material in the disorder (e.g., Golier et al., 2003; Lin et al., 2015; Stanford et al., 2001; Zeitlin & McNally, 1991), this effect conversely challenged the hypothesis of inhibitory deficits in memory functioning for threatening information.

Nevertheless, overall our data replicate prior findings of inhibitory impairments in the disorder (e.g., Stein, Kennedy, & Twamley, 2002; Wu et al., 2010), by revealing an increased recognition of “to-be-forgotten” words in PTSD patients when compared to controls. However, while this effect was, as expected, specific to R responses, PTSD patients presented a preserved directed forgetting effect on R recognitions of trauma-related words. As a conse- quence, contrary to predictions, despite generally impaired inhibition ability, it appeared that PTSD patients presented a preserved ability to control their memory functioning for trauma-related information.

Regarding difficulties encountered by PTSD patients in preventing the outcome of their traumatic reminiscences, this effect appears quite counterintuitive. While on the one hand patients are suffering from involuntary intrusive memories, which are moreover easily triggered by environ- mental stimuli (e.g., Ehlers & Clark, 2000), on the other hand, associative encodings for trauma-related stimuli can be willingly controlled in experimental settings. In line with Stanford et al.’s (2001) findings, a possible expla- nation might emerge from a preferential allocation of cog- nitive resources towards threatening elements in PTSD. As a consequence, the processing of non-threatening infor- mation would face resources limitations. According to their findings, the trauma saliency of the stimuli is a deter- mining factor of the cognitive resources re-allocation.

Consequently, in addition to providing a possible expla- nation for the reduced directed forgetting effect found in positive and neutral words, these data also offer interpret- ation clues for the reduced directed forgetting effect for negative words, despite the anxious and depressive comorbidities found in the PTSD group.

Interpretations of this preserved directed forgetting effect in PTSD are limited by non-significant differences on R recognitions between words’ emotional valences within the PTSD group in both “to remember” and “to forget” conditions. However, the significant difference between the two conditions could reflect PTSD symptoma- tology, which is characterised by the coexistence of both intrusive and avoidant symptoms. While intrusive symp- toms might have favoured R recognitions for trauma- related words, the avoidant symptomatology, described as a persistent avoidance of stimuli and thoughts associ- ated with the traumatic event (DSM-5; APA, 2013), could have trained patients to easily disengage from trauma- related information when they were perceived to be irrele- vant to the task.

Interestingly, the combined role of these processes has already been pointed out in the Vigilant-Avoidant model of attention biases (Mogg, Mathews, & Weinman, 1987). Developed in PTSD for integrating divergences of findings in the studies exploring attention allocation, this model suggests that patients could develop either an attention focus towards threat when attention is measured in the first steps of processing, or a subsequent attention distraction away from threat when measures are delayed (Adenauer et al., 2010; Thomas et al., 2013). Occurring gen- erally sequentially, the initial focus is usually conceived as unstoppable as it depends on bottom-up uncontrolled pro- cesses, whereas the secondary disengagement, which is more top-down dependent, could be controlled when con- scious efforts are deployed (Bardeen & Orcutt, 2011). As the directed forgetting paradigm evaluated voluntary and control-dependent inhibitory processes, it is possible to analyse findings in terms of attention allocation. From this perspective, if in both “to-remember” and “to forget” conditions attention could have been initially allocated to trauma-related stimuli, thus contributing to an encoding ease, the secondary control-dependent disengagement could have been either blocked in the “to remember” con- dition or used in order to ease the inhibition of the “to-be- forgotten” words. Consequently, it is possible that the inhibitory deficits hypothesis is not necessarily irrelevant but that such deficits could occur on more uncontrolled processes, probably in the first steps of information processing.

Given the limitations of the current study, some points could be taken into account as far as future studies are con- cerned. First of all, regarding our sample, although all patients met criteria for clinical diagnosis of PTSD, they differed on the traumatic events experienced. Conse- quently, even if the words labelled as trauma-related appeared relevant for the most commonly encountered

1410 A.-C. TUDORACHE ET AL.

traumatic events (i.e., interpersonal violence, physical and sexual abuse), we were not able to ensure that every trauma-related word was actually relevant for each partici- pant. Facing this inherent limitation, a clue for future studies would be to include ratings of the traumatic rel- evance of the words used. Moreover, it is possible that negative or even neutral or positive words could also have constituted reminders of traumatic experiences. While this constraint seems limited regarding the clear differences observed between trauma-related and other type of words, perhaps effect sizes would have benefited from assessments of the individual trauma relevance of the processed material. Along with these considerations, further studies could additionally benefit from contrasting different types of stimuli. Considering that trauma-related intrusions notably include images and visual memories (Brewin et al., 2010; Ehlers & Clark, 2000), it would be inter- esting to replicate this investigation with pictorial stimuli. Finally, it would be interesting to include a group of partici- pants who have been exposed to traumatic events without developing PTSD. The advantage of such a group would be to allow for examination of whether the trauma relevance bias could be attributed to the development of the dis- order or rather reflective of the familiarity of trauma- related words.

In conclusion, the aim of the current study was to assess inhibitory control abilities in memory functioning for trauma-related information in PTSD. While only few studies have used the R/K paradigm for memory investi- gations in PTSD, our findings revealed specifically increased R recognitions for trauma-related material in PTSD patients when compared to non-PTSD controls. Moreover, despite generally impaired memory inhibition skills, PTSD patients presented a preserved ability to control and notably to inhibit their memory functioning for trauma-related material specifically. Consequently, at this point, rather than the expected impaired control abil- ities, memory processing of trauma-related material seemed consistent with theories of cognitive resources reallocation toward threatening material, at the expense of other types of information. While unexpected, the pre- served ability to control memory functioning for threaten- ing information in PTSD could find some support in a recent study revealing a generalised improvement in memory inhibition following trauma exposure in a Think/ No-Think task (Hulbert & Anderson, 2018). While limited and specific to trauma-related material, the preserved memory control abilities found in our study suggest that supporting the reallocation of cognitive resources from trauma-related to general information might reduce threat-related biases in PTSD, and by extension could con- tribute to symptoms reductions.

Acknowledgements

The authors thank Ashley Smith Watts for English editing of this manu- script and the International Neuropsychological Society (INS)

International Liaison Committee (ILS) Research Editing and Consulting Program for their assistance. Additionally, the authors thank the Poitou-Charentes region and the Collaborative Research Actions (ARC, 2014) from the universities of Tours and Poitiers for providing material support.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Adenauer, H., Pinösch, S., Catani, C., Gola, H., Keil, J., Kißler, J., & Neuner, F. (2010).Earlyprocessingofthreatcuesinposttraumaticstress disorder —evidence for a cortical vigilance-avoidance reaction. Biological Psychiatry, 68(5), 451–458. doi:10.1016/j.biopsych.2010.05.015

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders: DSM-IV (4th ed.). Washington, DC: American Psychiatric Association.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders: DSM-5 (5th ed.). Washington, DC: American Psychiatric Publishing.

Bardeen, J. R., & Orcutt, H. K. (2011). Attentional control as a moderator of the relationship between posttraumatic stress symptoms and attentional threat bias. Journal of Anxiety Disorders, 25(8), 1008– 1018. doi:10.1016/j.janxdis.2011.06.009

Barnes, L. L. B., Harp, D., & Jung, W. S. (2002). Reliability generalization of scores on the Spielberger state–trait anxiety inventory. Educational and Psychological Measurement, 62, 603–618. doi:10. 1177/0013164402062004005

Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., Gusman, F. D., Charney, D. S., & Keane, T. M. (1995). The development of a clinician- administered PTSD Scale. Journal of Traumatic Stress, 8(1), 75–90. doi:10.1002/jts.2490080106

Bomyea, J., Amir, N., & Lang, A. J. (2012). The relationship between cog- nitive control and posttraumatic stress symptoms. Journal of Behavior Therapy and Experimental Psychiatry, 43(2), 844–848. doi:10.1016/j.jbtep.2011.12.001

Bondolfi, G., Jermann, F., Rouget, B. W., Gex-Fabry, M., McQuillan, A., Dupont-Willemin, A., … Nguyen, C. (2010). Self- and clinician-rated Montgomery–Åsberg depression rating scale: Evaluation in clinical practice. Journal of Affective Disorders, 121(3), 268–272. doi:10.1016/ j.jad.2009.06.037

Bonin, P., Méot, A., Aubert, L., Malardier, N., Niedenthal, P., & Capelle- Toczek, M. (2003). Normes de concrétude, de valeur d’imagerie, de fréquence subjective et de valence émotionnelle pour 866 mots. = Concreteness, imageability, subjective frequency and emo- tionality ratings for 866 words. L’année Psychologique, 103(4), 655– 694. doi:10.3406/psy.2003.29658

Brewin, C. R., Gregory, J. D., Lipton, M., & Burgess, N. (2010). Intrusive images in psychological disorders: Characteristics, neural mechan- isms, and treatment implications. Psychological Review, 117(1), 210–232. doi:10.1037/a0018113

Cisler, J. M., Wolitzky-Taylor, K. B., Adams, T. G., Jr., Babson, K. A., Badour, C. L., & Willems, J. L. (2011). The emotional Stroop task and posttraumatic stress disorder: A meta-analysis. Clinical Psychology Review, 31(5), 817–828. doi:10.1016/j.cpr.2011.03.007

Content, A., Mousty, P., & Radeau, M. (1990). BRULEX. Une base de données lexicales informatisée pour le français écrit et parlé. = BRULEX: A computerized lexical data base for the French language. L’année Psychologique, 90(4), 551–566. doi:10.3406/ psy.1990.29428

Cottencin, O., Vaiva, G., Huron, C., Devos, P., Ducrocq, F., Jouvent, R., … Thomas, P. (2006). Directed forgetting in PTSD: A comparative study versus normal controls. Journal of Psychiatric Research, 40(1), 70–80. doi:10.1016/j.jpsychires.2005.04.001

MEMORY 1411

Ehlers, A., & Clark, D. M. (2000). A cognitive model of posttraumatic stress disorder. Behaviour Research and Therapy, 38(4), 319–345. doi:10.1016/S00057967(99)00123-0

El Khoury-Malhame, M., Lanteaume, L., Beetz, E. M., Roques, J., Reynaud, E., Samuelian, J., … Khalfa, S. (2011). Attentional bias in post-traumatic stress disorder diminishes after symptom ameliora- tion. Behaviour Research and Therapy, 49(11), 796–801. doi:10.1016/ j.brat.2011.08.006

Fleurkens, P., Rinck, M., & van Minnen, A. (2011). Specificity and gener- alization of attentional bias in sexual trauma victims suffering from posttraumatic stress disorder. Journal of Anxiety Disorders, 25(6), 783–787. doi:10.1016/j.janxdis.2011.03.014

Foa, E. B., Molnar, C., & Cashman, L. (1995). Change in rape narratives during exposure therapy for posttraumatic stress disorder. Journal of Traumatic Stress, 8(4), 675–690. doi:10.1002/jts.2490080409

Gardiner, J. M., & Conway, M. A. (1999). Levels of awareness and var- ieties of experience. In B. H. Challis, & B. M. Velichkovsy (Eds.), Stratification of consciousness and cognition (pp. 237–254). Amsterdam/Philadelphia: John Benjamin Publishing Company.

Golier, J. A., Yehuda, R., Lupien, S. J., & Harvey, P. D. (2003). Memory for trauma-related information in Holocaust survivors with PTSD. Psychiatry Research, 121(2), 133–143. doi:10.1016/S0925-4927 (03)00120-3

Gray, M. J., Litz, B. T., Hsu, J. L., & Lombardo, T. W. (2004). Psychometric properties of the life events checklist. Assessment, 11(4), 330–341. doi:10.1177/1073191104269954

Hamann, S. (2001). Cognitive and neural mechanisms of emotional memory. Trends in Cognitive Sciences, 5(9), 394–400. doi:10.1016/ S1364-6613(00)01707-1

Harvey, A. G., & Bryant, R. A. (1999). The role of anxiety in attempted thought suppression following exposure to distressing or neutral stimuli. Cognitive Therapy and Research, 23(1), 39–52. doi:10.1023/ A:1018758623889

Hulbert, J. C., & Anderson, M. C. (2018). What doesn’t kill you makes you stronger: Psychological trauma and its relationship to enhanced memory control. Journal of Experimental Psychology: General, 147(12), 1931–1949. (Supplemental). doi:10.1037/ xge0000461.supp

Jenkins, M. A., Langlais, P. J., Delis, D., & Cohen, R. (1998). Learning and memory in rape victims with posttraumatic stress disorder. The American Journal Of Psychiatry, 155(2), 278–279. doi:10.1176/ajp. 155.2.278

Lin, M., Hofmann, S. G., Qian, M., & Li, S. (2015). Enhanced association between perceptual stimuli and trauma-related information in indi- viduals with posttraumatic stress disorder symptoms. Journal of Behavior Therapy and Experimental Psychiatry, 46, 202–207. doi:10. 1016/j.jbtep.2014.10.008

Long, N., Chamberlain, K., & Vincent, C. (1994). Effect of the Gulf War on reactivation of adverse combat-related memories in Vietnam veter- ans. Journal Of Clinical Psychology, 50(2), 138–144. doi:10.1002/ 1097-4679(199403)50:2<138::AID- JCLP2270500203>3.0.CO;2-T

MacLeod, C. M. (1999). The item and list methods of directed forget- ting: Test differences and the role of demand characteristics. Psychonomic Bulletin & Review, 6(1), 123–129. doi:10.3758/ BF03210819

McCabe, D. P., & Geraci, L. D. (2009). The influence of instructions and terminology on the accuracy of remember-know judgments. Consciousness and Cognition: An International Journal, 18(2), 401– 413. doi:10.1016/j.concog.2009.02.010

McNally, R. J., Metzger, L. J., Lasko, N. B., Clancy, S. A., & Pitman, R. K. (1998). Directed forgetting of trauma cues in adult survivors of childhood sexual abuse with and without posttraumatic stress dis- order. Journal of Abnormal Psychology, 107(4), 596–601. doi:10. 1037/0021-843X.107.4.596

Messina, D., Morais, J., & Cantraine, F. (1989). Valeur affective de 904 mots de la langue française. = Emotional value of 904 words in the French language. Cahiers De Psychologie Cognitive/Current Psychology of Cognition, 9(2), 165–187.

Mogg, K., Mathews, A., & Weinman, J. (1987). Memory bias in clinical anxiety. Journal of Abnormal Psychology, 96(2), 94–98. doi:10.1037/ 0021-843X.96.2.94

Montgomery, S. A., & Åsberg, M. (1979). A new depression scale designed to be sensitive to change. The British Journal of Psychiatry, 134, 382–389. doi:10.1192/bjp.134.4.382

Moradi, A. R., Taghavi, R., Neshat-Doost, H. T., Yule, W., & Dalgleish, T. (2000). Memory bias for emotional information in children and ado- lescents with posttraumatic stress disorder: A preliminary study. Journal of Anxiety Disorders, 14(5), 521–534. doi:10.1016/S0887- 6185(00)00037-2

Paunovic, N., Lundh, L., & Öst, L. (2002). Attentional and memory bias for emotional information in crime victims with acute posttraumatic stress disorder (PTSD). Journal of Anxiety Disorders, 16(6), 675–692. doi:10.1016/S0887-6185(02)00136-6

Rubin, D. C., Berntsen, D., & Bohni, M. K. (2008). A memory-based model of posttraumatic stress disorder: Evaluating basic assumptions underlying the PTSD diagnosis. Psychological Review, 115(4), 985– 1011. doi:10.1037/a0013397

Schneider, W., Eschman, A., & Zuccolotto, A. (2002). E-prime user’s guide. Pittsburgh, PA: Psychology Software Tools.

Spielberger, C. D. (1983). Manual for the State-Trait anxiety Inventory (Form Y) (“self- evaluation questionnaire”). Palo Alto, CA: Mind Garden, Consulting Psychologists Press.

Stanford, M. S., Vasterling, J. J., Mathias, C. W., Constans, J. I., & Houston, R. J. (2001). Impact of threat relevance on P3 event-related poten- tials in combat-related post- traumatic stress disorder. Psychiatry Research, 102, 125–137. doi:10.1016/S0165-1781(01)00236-0

Stein, M. B., Kennedy, C. M., & Twamley, E. W. (2002). Neuropsychological function in female victims of intimate partner violence with and without posttraumatic stress disorder. Biological Psychiatry, 52(11), 1079–1088. doi:10.1016/S00063223 (02)01414-2

Tapia, G., Clarys, D., Bugaiska, A., & El-Hage, W. (2012). Recollection of negative information in posttraumatic stress disorder. Journal of Traumatic Stress, 25(1), 120–123. doi:10.1002/jts.21659

Tapia, G., Clarys, D., El-Hage, W., & Isingrini, M. (2007). Les troubles cog- nitifs dans le post-traumatic stress disorder (PTSD): Une revue de la littérature. = Cognitive deficits in post-traumatic stress disorder: A review. L’année Psychologique, 107(3), 489–523. doi:10.4074/ S0003503307003065

Tapia, G., Clarys, D., El Hage, W., Belzung, C., & Isingrini, M. (2007). PTSD psychiatric patients exhibit a deficit in remembering. Memory, 15(2), 145–153. doi:10.1080/09658210601145965

Tapia, G., Clarys, D., Isingrini, M., & El-Hage, W. (2007). Mémoire et émotion dans le trouble de stress post-traumatique (TSPT). = Memory and emotion in post-traumatic stress disorder (PTSD). Canadian Psychology/Psychologie Canadienne, 48(2), 106–119. doi:10.1037/cp2007012

Thomas, C. L., Goegan, L. D., Newman, K. R., Arndt, J. E., & Sears, C. R. (2013). Attention to threat images in individuals with clinical and subthreshold symptoms of post- traumatic stress disorder. Journal of Anxiety Disorders, 27(5), 447–455. doi:10.1016/j.janxdis.2013.05. 005

Tulving, E. (1985). Memory and consciousness. Canadian Psychology/ Psychologie Canadienne, 26(1), 1–12. doi:10.1037/h0080017

Vasterling, J. J., Brailey, K., Constans, J. I., & Sutker, P. B. (1998). Attention and memory dysfunction in posttraumatic stress disorder. Neuropsychology, 12(1), 125–133. doi:10.1037/0894-4105.12.1.125

Vikis-Freibergs, V. (1976). Abstractness and emotionality values for 398 French words. Canadian Journal of Psychology/Revue Canadienne De Psychologie, 30(1), 22–30. doi:10.1037/h0082041

Vrana, S. R., Roodman, A., & Beckham, J. C. (1995). Selective processing of trauma relevant words in posttraumatic stress disorder. Journal of Anxiety Disorders, 9(6), 515–530. doi:10.1016/0887-6185 (95)00028-M

Waller, N., Putnam, F. W., & Carlson, E. B. (1996). Types of dissociation and dissociative types: A taxometric analysis of dissociative

1412 A.-C. TUDORACHE ET AL.

experiences. Psychological Methods, 1(3), 300–321. doi:10.1037/ 1082-989X.1.3.300

Weathers, F. W., Keane, T. M., & Davidson, J. T. (2001). Clinician- Administered PTSD Scale: A review of the first ten years of research. Depression and Anxiety, 13(3), 132–156. doi:10.1002/da.1029

Williams, J. M. G., Mathews, A., & MacLeod, C. (1996). The emotional Stroop task and psychopathology. Psychological Bulletin, 120(1), 3–24. doi:10.1037/0033-2909.120.1.3

Wu, J., Ge, Y., Shi, Z., Duan, X., Wang, L., Sun, X., & Zhang, K. (2010). Response inhibition in adolescent earthquake survivors with and without posttraumatic stress disorder: A combined behavioral and ERP study. Neuroscience Letters, 486(3), 117–121. doi:10.1016/ j.neulet.2010.07.040

Yonelinas, A. P., & Jacoby, L. L. (1995). The relation between remember- ing and knowing as bases for recognition: Effects of size

congruency. Journal of Memory And Language, 34(5), 622–643. doi:10.1006/jmla.1995.1028

Zeitlin, S. B., & McNally, R. J. (1991). Implicit and explicit memory bias for threat in post- traumatic stress disorder. Behaviour Research And Therapy, 29(5), 451–457. doi:10.1016/0005-7967 (91)90129-Q

Zoellner, L. A., Sacks, M. B., & Foa, E. B. (2003). Directed forgetting fol- lowing mood induction in chronic posttraumatic stress disorder patients. Journal of Abnormal Psychology, 112(3), 508–514. doi:10. 1037/0021-843X.112.3.508

Zwissler, B., Hauswald, A., Koessler, S., Ertl, V., Pfeiffer, A., Wöhrmann, C., … Kissler, J. (2012). Memory control in post-traumatic stress dis- order: Evidence from item method directed forgetting in civil war victims in Northern Uganda. Psychological Medicine, 42(6), 1283– 1291. doi:10.1017/S0033291711002273

MEMORY 1413

Appendix A

Mean proportions (and standard errors) of total recognitions, Remember responses and IRK scores by group, encoding condition and emotional valence.

Note: * p < .05; ** p < .01; *** p < .001M: mean; SE: standard error. IRK stands for Independent Know responses calculation according the Inde- pendent Remember-Know procedure (Yonelinas & Jacoby, 1995)

1414 A.-C. TUDORACHE ET AL.