Involuntary Memories of War-Related Scenes in Veterans With PTSD

Abstract

Studies suggest that general control deficits and elevated affect intensity in posttraumatic stress disorder (PTSD) extend beyond memory for the index trauma. However, few researchers have pursued this possibility experimentally by examining memory for novel events. We used an experimental design to measure the frequency and characteristics of involuntary memories over time. Veterans with and without PTSD saw pictures of neutral and war-related scenes. Half of the participants completed an involuntary-retrieval task immediately after encoding, whereas the other half completed the retrieval task after 1 week. Veterans with PTSD had stronger emotional reactions to their involuntary memories of the scenes regardless of their original valence. The emotional impact and specificity of the memories did not diminish over time in PTSD veterans but did so in the control group. The findings are consistent with an increased emotional response to a range of memories that include—but are not limited to—memories of traumatic events.

Keywords

involuntary memory intrusions posttraumatic stress disorder trauma memory

Repeated and unwanted involuntary memories of a traumatic event is a hallmark symptom of posttraumatic stress disorder (PTSD; American Psychiatric Association, 2013; Brewin et al., 2010). Such intrusive memories are common among trauma survivors during the days and weeks immediately following the traumatic event, after which their frequency and intensity in most cases will subside. However, for individuals diagnosed with PTSD, intrusive memories of the trauma are persistent and may cause significant impairment of the person’s life even years after the event (American Psychiatric Association, 2013). Intrusions in PTSD can be very vivid and emotionally intense (Marks et al., 2018), and because they enter awareness unintended, their appearance is experienced as sudden and uncontrollable. In addition, individuals may not always be aware of the cues triggering the memories, making the intrusion seem to appear out of nowhere (Ehlers et al., 2004).

Despite their importance, the causal mechanisms underlying intrusive memories in PTSD are unresolved. According to one influential group of theories, intrusive memories primarily owe their existence to specific cognitive processes operating during trauma exposure, leading to a faulty encoding of the event. This, in turn, renders the traumatic memory highly accessible for involuntary and repetitive remembering (e.g., Brewin et al., 2010; Ehlers et al., 2004). Researchers have attempted to clarify such underlying mechanisms by simulating exposure to real-life trauma in healthy individuals, using the trauma film paradigm, and prospectively examining factors during exposure that may lead to subsequent intrusions of the trauma film. In a recent review of this literature, Marks et al. (2018) proposed that maladaptive processing at the time of retrieval, in terms of negative appraisals and lack of conceptual processing, may be as important for the persistence of intrusions as processes occurring during the encoding of the trauma.

These approaches share a common focus on memory for the index trauma. In their view, the processing of the traumatic event generates the persistent, intrusive memories. However, an exclusive focus on memory processes related to the index trauma has limitations (e.g., Boykin & Teng, 2019). Recent studies suggest that more global cognitive deficits, such as a general inability to inhibit aversive memories (Catarino et al., 2015); global, deficient memory control abilities (Mary et al., 2020); or increased affect intensity (Rubin et al., 2011) may be at least as important as how the index trauma is processed. For example, Rubin and colleagues (2008) showed increased ratings of sensory and emotional qualities of a range of autobiographical memories, including but not limited to memory for the traumatic event, in students with elevated levels of PTSD symptoms (Rubin et al., 2008) and in patients with PTSD relative to matched control participants (Rubin et al., 2011). According to these views, not only memory for the index trauma is affected in PTSD but also memory for other events in the person’s life; novel events will also be influenced.

There is a surprising dearth of experimental studies investigating intrusive memories and their associated emotional reactions for novel events in patients with PTSD. Although many studies have examined reactions to traumatic or stressful material using a trauma-analogue design (for reviews, see James et al., 2016; Marks et al., 2018), participants in these experimental studies are almost exclusively drawn from healthy populations. We were able to identify only two experimental studies investigating intrusions in clinical participants by using a trauma-analogue design: Nicholson et al. (2014) and Hall et al. (2018). We further identified two studies that examined memory suppression of intrusions using either a voluntary-retrieval task (Catarino et al., 2015) or memory for neutral material (Mary et al., 2020) in individuals with or without elevated levels of PTSD. In the following, we briefly review these four studies.

Nicholson et al. (2014) presented participants with and without PTSD with negative, neutral, and positive stimuli from the International Affective Picture System (IAPS; Lang et al., 2008). After 48 hr, participants returned for a retrieval task in which they retrospectively recorded the number of intrusions they had experienced during the preceding 2 days. Participants with PTSD reported having had significantly more intrusions compared with participants without PTSD.

Hall et al. (2018) presented participants with and without PTSD with negative and neutral pictures of scenes. Each picture was paired with a sound. The participants then listened to the sounds from encoding in a functional MRI (fMRI) scanner while performing a simple sound-location task. After the conclusion of this task, participants retrospectively reported whether they had experienced an involuntary memory in response to each sound. Compared with control participants, participants with PTSD had delayed activation of the memory network in general and increased activity in prefrontal areas for negative memories compared with neutral memories. In addition, participants with PTSD reported more involuntary memories (of both negative and neutral scenes) and reported higher emotional intensity and vividness when asked immediately after the scanning phase.

Catarino et al. (2015) used an adapted version of the think/no-think paradigm to examine memory suppression in participants with PTSD and participants without PTSD. The participants saw a series of aversive scenes, each paired with a thematically related visual cue. They next saw the cues together with an instruction to either retrieve or suppress the associated scene-memory. In the final test phase, the cues were presented once again, and participants were asked to recall the associated scenes. The control participants showed the expected memory reduction for scenes they had been instructed to suppress during the think/no-think phase, whereas the PTSD group did not, leading the authors to conclude the presence of generic inhibitory-control deficits in PTSD patients.

Following up on this work, Mary et al. (2020) used neutral pictures paired with neutral words to examine memory control deficits and their neural basis in survivors of the 2015 Paris terrorist attack. Participants were classified into three groups: exposed individuals with PTSD, exposed individuals without PTSD, and nonexposed individuals. Using fMRI, they showed that voluntary attempts to suppress the memories of the neutral pictures in response to the reminder cues were associated with different neural connectivity in individuals with PTSD relative to control participants, suggesting a reduced ability of the prefrontal control system to downregulate memory activity in the PTSD group. At the behavioral level, individuals with PTSD showed no reduction in priming effects after the suppression task, in contrast to the two other groups. Mary et al. interpreted the findings as reduced inhibitory control in the PTSD group, consistent with Catarino et al. (2015).

Despite their strengths, these previous studies had limitations that we aimed to meet in the present study. First, the previous studies used either retrospective recording of involuntary memories or a voluntary memory task. In contrast, in the present study, we used an online (nonretrospective) recording procedure, allowing intrusive memories to be recorded as they occur and thus reducing the need for retrospection. Second, none of the previous studies examined the effect of retention time, whereas we compared memories retrieved immediately with memories retrieved after a delay to detect more long-term effects. Third, the participants in the present study were sampled from a population of Danish veterans, and the negative stimuli used in the present experiment were all war related. Thereby the present study adds to previous work by focusing on a more homogeneous population and more specific negative stimuli.

The Present Study

The main purpose of the study was to examine the persistence and emotional intensity of intrusive memories in veterans with PTSD compared with veterans without a PTSD diagnosis by using an experimental approach. To this end, we created a trauma analogue in which negative scenes appeared unpredictably within a stream of neutral scenes. Each scene was paired with a neutral sound cue, and some sound cues were associated with both threatening and nonthreatening scenes. The aim of this manipulation was to simulate real-world traumatic situations, in which traumatic events usually happen in a neutral or benign setting if not for the traumatic episode (e.g., an assault in an otherwise peaceful forest or an accident on an otherwise ordinary road). This approach reflects numerous clinical anecdotes of veterans with PTSD experiencing intrusions in response to cues that would appear neutral to most people, such as the scent of lilies or a pot roast, the sound of a helicopter or fireworks, or the sight of a cornfield (all authentic examples from patients at the Clinic for PTSD and Anxiety, Central Region Denmark; see also Kleim et al., 2013; Pitman, 1988).

In previous research, we developed a task that reliably induces involuntary memories in control participants (Berntsen et al., 2013). A modified version was also used in the brain-imaging design by Hall et al. (2018) described above. By introducing highly negative war-related scenes into this design, we made a trauma analogue suitable for combat veterans. Veterans were presented with negative and neutral scenes paired with sounds. After a delay of either a few minutes or a week, depending on condition, the veterans completed a simple visual attention task while listening to the sound cues from the earlier task as well as sounds not previously heard. Participants then recorded any memories that came to mind involuntarily. In addition to these direct measures of involuntary memories as they occurred, we also included self-report measures of involuntary memories during daily living and during the study’s retention interval. This allowed us to compare on-line measures with retrospective measures.

In sum, the present study investigated the persistence and emotional intensity of intrusions as they occurred in a clinical sample using trauma-relevant stimuli. Following the view of general elevated affect intensity in PTSD (e.g., Rubin et al., 2008, 2011), we expected veterans with PTSD to report higher emotional intensity of involuntary memories of both negative and neutral scenes. In addition, we expected that the emotional intensity of involuntary memories would persist over a delay of 1 week in veterans with PTSD but not in veterans without PTSD.

Following the idea of a general reduction of inhibitory control in PTSD (e.g., Catarino et al., 2015; Mary et al., 2020), we expected a higher frequency of involuntary memories overall in veterans with PTSD. We also expected a higher frequency of involuntary memories of negative scenes in response to neutral cues associated with both negative and neutral events in this group compared with veterans without PTSD.

Method

Recruitment and ethics

Veterans with PTSD

Ethical approval was obtained from the Regional Committee on Health Research Ethics for the Central Denmark Region. Forty-six veterans with PTSD were recruited from the Clinic for PTSD and Anxiety, Central Region Denmark, and the Unit for Trauma- and Torture Survivors, Region of Southern Denmark (additional information is provided in the Supplemental Material available online). All veterans with PTSD were diagnosed by experienced clinical psychologists or psychiatrists according to the criteria of the International Statistical Classification of Diseases and Related Health Problems (10th Revision; World Health Organization, 2004). Veterans with PTSD were invited either by the therapist at their treatment center or through a letter asking for their consent to contact them. As part of the ethical requirements from the participating treatment centers, all veterans with PTSD had completed their treatment before being invited to the study. Treatment was tailored to the individual needs and personal resources of each veteran but generally included prolonged exposure and selective serotonin reuptake inhibitors. In addition, some veterans received physical therapy, eye-movement desensitization and reprocessing therapy, or mentalization-based therapy.

Veterans with PTSD were informed about the study and gave consent to be contacted by the research team. During this second contact, they were again informed of the purpose and procedure of the study as well as possible risks and benefits. When they arrived for testing, they were informed a third time and signed the consent form. It was explained that the study involved potentially unpleasant, offensive, or frightening pictures of war. Veterans were told they could stop participation at any time without any consequences. A clinical psychologist was always present during testing to ensure participants’ safety. Two weeks following participation, participants were contacted by phone and asked whether their participation had led to any unexpected side effects. No participant reported side effects from the study, but a few seemed uncomfortable during testing. In each case, the psychologist gently examined the reason for their distress.

Veterans without PTSD

Fifty-four veterans without PTSD were recruited through advertising on Facebook, the soldiers’ union, and the Jutland Dragoon Regiment at Holstebro, Denmark. In the recruitment poster, potential participants were asked to contact the research team if they were interested in participating in the study. Inclusion criterion for the study was at least one previous deployment to a war zone. From that point on, the recruitment procedure was identical to the one described above for veterans with PTSD. The ethical procedures were also the same except that a clinical psychologist was not necessarily present during testing of veterans without PTSD.

Design

The design and materials were based on a paradigm introduced in Berntsen et al. (2013) with modifications. The current experiment consisted of two tasks: encoding and retrieval. During encoding, participants were presented with visual scenes paired with sounds. The scenes were either negative or neutral, whereas all the sounds were neutral. Sounds and scenes were either unique (belonging to a category presented only once during encoding) or repeated (belonging to a category presented several times). Although repeated sounds were not identical, they were similar enough to be difficult to separate (e.g., highly similar sounds of birds singing). This created four types of trials: one in which sound and scene were both unique (unique cue, unique scene), one in which sound was unique and scene was repeated (unique cue, repeated scene), one in which sound was repeated and scene was unique (repeated cue, unique scene), and finally, one in which sound and scene were both repeated (repeated cue, repeated scene). Note that, as modified from the original paradigm, the repeated-cue, unique-scene trials contained only negative scenes, whereas the other three trials contained only neutral scenes (see Fig. 1).

Fig. 1.

Overview of the encoding procedure in one of the four versions of the task. A = repeated sound, repeated neutral scene; B = unique sound, unique neutral scene; C = unique sound, repeated neutral scene; D = repeated sound, unique negative scene.

This created a trauma-analogue experience in which negative events would happen unpredictably within a flow of neutral events and cues could refer to either the negative or the neutral aspects of the experience. The design allowed us to investigate whether veterans with PTSD reported a higher frequency of negative scenes than veterans without PTSD in response to cues associated with both negative and neutral scenes. Of the 40 trials presented during encoding, 28 were repeated cue, repeated scene, four were repeated cue, unique scene, four were unique cue, repeated scene, and four were unique cue, unique scene.

The study had four different versions of the encoding task. In one version, different pictures of trees and different sound clips of bird songs were the neutral repeated stimuli, and pictures of a dog, a car, and a street with their respective sounds were the neutral unique stimuli. In the second version, a street, a car, and trees (and their respective sounds) acted as the neutral unique stimuli, and dogs (with their sounds) were the neutral repeated stimuli. In the third and fourth versions, streets and cars were the neutral repeated stimuli, respectively. All four versions included a fourth neutral unique scene featuring a man playing a trumpet accompanied by the unique sound of jazz music. This randomization was used to control for the content of the scenes and sounds. Finally, the negative scenes were the same across every version of the task and consisted of violent war-related scenes (see Table S1 in the Supplemental Material).

During retrieval, participants watched a yellow star, which appeared either on the left or the right side of the monitor in an unpredictable pattern. With each presentation of the star, a sound was played. Sounds were either from the encoding task, and therefore associated with scenes from encoding, or new and therefore not associated with any scene from encoding (see Fig. 1).

Materials

We asked the veterans about age, gender, education, military rank, and number of deployments (Table 1). We used the following standardized questionnaires: (a) Beck Depression Inventory–II (BDI-II; Beck et al., 1996) measures symptoms related to depression. The internal consistency was α = .95 in our sample. (b) The Traumatic Life Events Questionnaire (TLEQ; Kubany et al., 2000) measures a wide range of possible traumatic life experiences along with their impact and frequency. (c) The PTSD Checklist for DSM-5 (PCL-5; Blevins et al., 2015) measures a range of symptoms related to PTSD. In the present sample, internal consistency was α = .91.

Table 1.

Demographic and Questionnaire Data

Characteristic	Veterans with PTSD		Veterans without PTSD		Effect of group	Effect of time	Group × Time interaction
Characteristic	Immediate	1 week	Immediate	1 week	F(1, 85)	F(1, 85)	F(1, 85)
N	19	19	29	22	—	—	—
Age (years)	40.8 (7.6)	41.1 (6.7)	35.6 (9.3)	33.0 (6.6)	15.67**	0.51	0.76
Gender (male/female)	22/0	20/0	27/2	22/0	—	—	—
Education (years)	14.3 (2.4)	13.2 (2.2)	14.3 (2.2)	14.5 (1.5)	2.10	0.85	2.01
Military rank^a	1.1 (0.3)	1.2 (0.5)	1.3 (0.5)	1.3 (0.6)	1.28	0.10	0.51
Deployments	2.8 (2.3)	2.8 (2.3)	3.3 (2.2)	2.3 (1.3)	0.00	1.24	1.24
Encoding intensity (negative)	3.7 (1.2)	3.8 (1.2)	3.3 (1.0)	2.9 (1.0)	7.75*	0.43	0.59
Encoding intensity (neutral)	1.5 (0.6)	1.7 (0.6)	1.4 (0.8)	1.4 (0.6)	3.37	0.48	0.56
Questionnaire scores
BDI-II	22.5 (7.9)	28.7 (12.0)	4.6 (4.8)	4.4 (4.0)	174.65**	3.51	3.93
TLEQ	21.3 (10.5)	19.1 (11.4)	16.3 (15.4)	10.0 (10.4)	6.80*	2.49	0.57
Combat exposure	3.9 (2.5)	4.2 (2.4)	5.0 (2.0)	4.4 (2.5)	1.63	0.15	0.64
PCL-5	37.1 (17.5)	43.3 (14.6)	6.3 (5.5)	4.7 (5.4)	203.36**	0.92	2.55
CES	27.7 (5.2)	27.9 (8.4)	15.3 (7.2)	14.0 (6.7)	76.54**	0.14	0.26
IAMI
Total	58.6 (16.0)	58.0 (14.6)	41.9 (13.8)	41.2 (10.0)	32.40**	0.49	0.00
Past	31.9 (8.6)	31.3 (6.3)	21.2 (6.8)	23.0 (6.0)	40.31**	0.16	0.68
Future	26.7 (8.0)	26.7 (9.2)	20.7 (7.8)	18.2 (4.9)	19.69**	0.59	0.59

Note: Values are means; values in parentheses are standard deviations. BDI = Beck Depression Inventory–II (Beck et al., 1996); TLEQ = Traumatic Life Events Questionnaire (Kubany et al., 2000); PCL-5 = PTSD Checklist for DSM-5 (Blevins et al., 2015); CES = Centrality of Event Scale (Berntsen & Rubin, 2006); IAMI = Involuntary Autobiographical Memory Inventory (Berntsen et al., 2015).

Military rank was coded as 1 for junior enlisted personnel, 2 for noncommissioned officers, and 3 for officers.

p < .05. **p < .001.

We used three scales to measure the importance and frequency of autobiographical memories. (d) The short form of the Centrality of Events Scale (CES; Berntsen & Rubin, 2006) measures the extent to which an event has become central to a person’s life story and identity. When filling out the CES, participants were instructed to think about the stressful event they had nominated for the PCL-5. Internal consistency in the present sample as measured by Cronbach’s α was .86. (e) The Involuntary Autobiographical Memory Inventory (IAMI; Berntsen et al., 2015) measures the self-reported frequency of involuntary memories and involuntary future thoughts during daily activities without specifying their valence or content (e.g., “When I am relaxing or doing routine work, memories of past events come to my mind by themselves—without me consciously trying to remember them”). Each item is scored on a 5-point scale from 0 = never to 4 = once an hour or more. In the present sample, internal consistency as measured by Cronbach’s α was .95. Finally, we measured rehearsal with the following three items: “During the past week, did you experience that the pictures from the task appeared in your memory?” (1 = not at all, 5 = all the time); “Which pictures appeared in your memory?” (1 = only the negative, 2 = mostly the negative, 3 = a bit of both, 4 = mostly the neutral, 5 = only the neutral); “When the pictures appeared in your memory, did it happen” (1 = only spontaneously, I did not try to remember; 2 = mostly spontaneously, I mostly did not try to remember; 3 = a bit of both; 4 = mostly voluntary, I mostly tried to actively remember; 5 = only voluntary, I tried to actively remember).

The experiment was run with E-Prime (Version 2.0; Psychology Software Tools, 2017). Neutral pictures were taken from various sources on the Internet. The negative pictures were taken from IAPS and were chosen to be war related (i.e., a dead soldier). All pictures were resized to 675 × 506 pixels. Sounds were a combination of royalty-free sound files from the Internet as well as from a large library of sounds. All sounds were normalized, and additional versions were created of each sound by panning it 75% to the left and right, respectively. Trials were run either on desktop or laptop computers, depending on the site.

Procedure

Regardless of diagnostic status, participants were randomly assigned to one of the four versions of the encoding task described above. Because of practical and geographical constraints, randomization to retrieval condition was not possible. To minimize the impact of a possible selection bias, we took care to match the control participants with the patients across retrieval conditions. Veterans with PTSD were tested mostly individually at their treatment clinics, although a few were tested in their own homes. Veterans without PTSD were tested either in a computer lab; in a meeting room at a conference center in Copenhagen, Denmark; or in a meeting room at a library in Vejle, Denmark. By keeping to a strict protocol, we strived for every testing session to be as similar as possible.

All participants were seated in front of a computer with headphones. All locations were quiet, and participants were not disturbed during the task. First, participants filled out the demographics and clinical questionnaires. Then, participants were told, “You will now be presented with a number of trials featuring some sounds and some pictures. Please pay close attention to both sounds and pictures.” The encoding task presented participants with 40 trials, each of which consisted of a scene-sound pair presented for 4 s (see Fig. 1). Following each trial, participants rated the emotional intensity of the scene on a 5-point scale (1 = not at all intense, 5 = very emotionally intense).

After completing the encoding task, half of the participants continued to the retrieval task immediately (the immediate group), whereas the other half left the lab and returned after 1 week (the 1-week group). In the retrieval task, participants completed 160 trials, each consisting of a sound presented either in the left ear or the right ear and a yellow star presented in either the left side or the right side of the screen. The star was presented 1.5 s after the initiation of the sound. The sounds were played for 4 s or until the participants pressed a button. Participants were told to press “1” when the star was to the left and “2” when the star was to the right. At the same time, they were told to press “3” if they experienced an image spontaneously appearing in their mind. It was not specified in advance whether the image should be one previously seen. If participants asked for clarification, they were told, “any image that enters your mind in response to a sound.” Pressing “3” paused the experiment and presented a digital questionnaire. Participants were required to describe their memory and rate it on three 5-point scales: “Did you experience a bodily reaction to the image, such as increased heart rate, nervousness or tension?” (1 = not at all, 5 = very much); “How specific is the image?” (1 = not specific at all, 5 = highly specific); “How emotionally intense is the image?” (1 = not intense at all, 5 = very intense).

Eighty of the 160 sounds were the 40 sounds from encoding presented once to each ear, and the remaining 80 sounds were 40 new sounds presented once to each ear. New sounds were also repeated or unique.

After the retrieval task, participants filled out the IAMI. Participants in the 1-week condition also filled out the rehearsal questionnaire. Participants were then awarded a gift card worth DKK 150 ($23).

Coding

The keyword descriptions written by the participants in response to the sound cues were coded for their correspondence to the scenes presented during encoding. If one or more words described a central component of the relevant scene, it was coded as correct. If a neutral scene was retrieved in response to either a wrong (unpaired) sound from encoding or to a new sound, this was coded as an error. If a negative scene was retrieved in response to a wrong sound from encoding or to a new sound, this was coded as an intrusion. If a memory clearly referred to a deployment event in the participant’s personal past, it was coded as deployment. Finally, if the keyword description contained no clear reference to a picture presented during encoding or to an event during deployment (but instead, e.g., appeared to be autobiographical), it was coded as other. Note that we did not expect participants to correctly remember which specific repeated scene they had seen in response to a specific repeated sound as long as they were able to correctly describe the central component of the scene (e.g., trees).

Two judges independently coded 20% of the keyword phrases and agreed in 89% of the cases. The remaining keyword phrases were coded by one judge.

Results

Demographics and questionnaires

Despite having completed their treatment, veterans with PTSD as a group still scored above the clinical cutoff on the PCL-5. At the same time, some veterans with PTSD had very low scores on the PCL-5. We excluded three veterans with PTSD in the immediate-retrieval condition with PCL-5 scores less than 15. One PTSD veteran was excluded from the 1-week-retrieval condition for being a statistical outlier on the BDI-II. Finally, two veterans without PTSD were excluded from the immediate-retrieval condition for having PCL-5 scores in the clinical range, defined as more than 33 (Weathers et al., 2013). We compared all memory analyses with and without these participants and found no statistically significant differences (analyses available on request).

The means and standard deviations for demographic and questionnaire data can be seen in Table 1. Education was estimated as the number of years it would take a Danish citizen to reach the level of education reported. Military rank was coded as 1 for junior enlisted personnel, 2 for noncommissioned officers, and 3 for officers. Three participants were civilian personnel and not assigned a score. Combat exposure was based on the corresponding item from the TLEQ, whereas deployments referred to the mean number of deployments reported by participants. In order of highest to lowest, the most frequent areas of deployment were Afghanistan, former Yugoslavia, Kosovo, and Iraq. We also used the item from the TLEQ asking whether participants with combat exposure had experienced intense fear during combat. Participants indicated their answer as yes or no. A χ² test showed that PTSD veterans more often reported intense fear during combat than veterans without PTSD (veterans with PTSD: yes = 21, no = 11; veterans without PTSD: yes = 8, no = 35, χ² = 17.1, p < .001). In addition, we found that veterans with PTSD in the immediate-retrieval condition reported a lower frequency of intense fear during combat (yes = 8, no = 9) than the veterans with PTSD in the 1-week-retrieval condition (yes = 13, no = 2), χ² = 5.5, p = .028. However, note that only a subset of participants gave answers to the combat-fear item because this item was answered only if the person had indicated exposure to combat. If we replaced all missing replies with “no” (because these participants did not report combat exposure and therefore could not have experienced combat-related fear), the χ² was no longer statistically significant, χ² = 2.7, p = .103. In short, combat fear did not appear to differentiate the immediate- and 1-week-retrieval groups within the PTSD sample.

We conducted a series of factorial analyses of variance (ANOVAs) with group (veterans with PTSD vs. veterans without PTSD) and retention interval (immediate vs. 1 week) as between-subjects variables (Table 1). Veterans with PTSD were older than veterans without PTSD (η_p² = .16; 90% confidence interval [CI] = [.05, .27]), and veterans with PTSD scored higher on the BDI-II (η_p² = .67; 90% CI = [.57, .73]), PCL-5 (η_p² = .71; 90% CI = [.62, .76]), TLEQ (η_p² = .07; 90% CI = [.01, .17]), CES (η_p² = .47; 90% CI = [.34, .57]), and IAMI (η_p² = .28; 90% CI = [.15, .39]). Finally, veterans with PTSD rated the war-related scenes as more emotionally intense than did veterans without PTSD (η_p² = .08; 90% CI = [.01, .19]), whereas the difference for neutral scenes was not statistically significant (η_p² = .04; 90% CI = [.0, .12]), although in the same direction numerically (see Table 1).

Memory frequency across groups, time, and cuing

We calculated frequency of memories as the percentage of correct memories of a certain type (e.g., unique cue paired with unique scene) relative to the total number of cues for that particular type of memory. We then entered the mean percentages for each of the four types of memory into a 2 × 2 × 4 mixed ANOVA with group (veterans with PTSD vs. veterans without PTSD) and retention interval (immediate vs. 1 week) as between-subjects variables and cue-item type (unique cue, unique scene vs. unique cue, repeated scene vs. repeated cue, unique scene vs. repeated cue, repeated scene) as a repeated measure. To reiterate, the repeated-cue, unique-scene category was the only one with negative scenes, and the cues in this condition were highly similar to the cues in the repeated-cue, repeated-scene condition.

Consistent with previous work, unique cues (M = 9.97, SD = 15.67) were more effective than repeated cues (M = 2.63, SD = 7.55) in triggering involuntary memories, as shown by a main effect of cue-item type, F(3, 255) = 10.51, p < .001, η_p² = .11; 90% CI = [.06, .18]. However, the proportion of memories in response to unique cues was reduced after 1 week (M = 12.89, SD = 18.79 vs. M = 6.55, SD = 10.17), whereas it increased in response to repeated cues (M = 1.34, SD = 2.77 vs. M = 4.14, SD = 10.58), as demonstrated by an interaction between cue-item type and retention interval, F(3, 255) = 3.99, p = .008, η_p² = .05; 90% CI = [.01, .09]. No other effects were statistically significant (ps > .167, η_p² < .03), including the expected interaction effect between group and cue-item type. Thus, counter to predictions deriving from the assumption of reduced inhibitory control in PTSD, the veterans with PTSD did not have a greater tendency for experiencing memories of war-related scenes in response to the neutral sound cues, nor did they report more involuntary memories overall.

Memory characteristics

We calculated mean ratings of bodily reaction, emotional intensity, memory specificity, and retrieval time for each individual participant’s correct memories. Because only a few memories were recorded in response to repeated cues, we did not include cuing as a factor in these analyses. We then ran separate factorial ANOVAs for each characteristic with group (veterans with PTSD vs. veterans without PTSD) and retention interval (immediate vs. 1 week) as between-subjects variables. Veterans with PTSD had higher bodily reaction than veterans without PTSD at both time points (Fig. 2a; see Table 2 for statistics). Although veterans without PTSD reported similar emotional intensity at both intervals (p = .134, η_p² = .08¹; 90% CI = [.0, .26]), veterans with PTSD reported higher emotional intensity after a week, F(1, 22) = 8.87, p = .007, η_p² = .29; 90% CI = [.05, .48] (Fig. 2b).

Fig. 2.

Characteristics of involuntary memories across the two time points and diagnostic status of participants. Bodily reaction (a), emotional intensity (b), and memory specificity (c) are presented for the immediate-retrieval and 1-week-retrieval groups. Error bars indicate ±1 SE.

Table 2.

Memory Characteristics Across Groups

	Veterans with PTSD		Veterans without PTSD		Effect of group	Effect of time	Group × Time interaction
Characteristic	Immediate	1 week	Immediate	1 week	F(49)	F(49)	F(49)
Bodily reaction	1.6 (0.8)	2.1 (0.8)	1.1 (0.3)	1.1 (0.4)	16.40**	2.07	2.05
Emotional intensity	2.0 (1.0)_a	3.2 (0.9)_b	1.9 (1.2)_a	1.3 (0.6)_a	13.54*	1.14	10.27*
Memory specificity	3.1 (1.1)_a	3.3 (0.7)_a	3.5 (1.2)_a	2.4 (1.1)_b	0.60	1.92	5.52*
Retrieval time (ms)	4,178 (1,533)	4,904 (3,123)	3,628 (1,226)	3,622 (1,587)	2.88	0.45	0.46
Intrusions	0.2 (0.5)	0.2 (0.7)	0.2 (0.5)	0.6 (1.7)	1.05	0.63	0.63
Negative memory index	0.3 (0.7)	0.3 (0.8)	0.4 (0.8)	1.0 (2.2)	2.29	1.65	1.20
Deployment memories	1.1 (2.1)	0.4 (1.0)	0.1 (0.3)	0.2 (0.8)	5.90*	1.44	3.42

Note: Values are the results of post hoc tests. Within the same row, values sharing a common subscript are not statistically significantly different (p < .05).

p < .05. **p < .001.

Memory specificity showed an interaction between group and time (p = .023, η_p² = .10; 90% CI = [.008, .24]), which was due to lower specificity in veterans without PTSD in the 1-week-retrieval relative to the immediate-retrieval condition, whereas veterans with PTSD did not show an effect of retention interval (Fig. 2c). Finally, retrieval time showed no statistically significant effects (Table 2). The largest effect for retrieval time was a medium difference showing that veterans with PTSD were potentially a bit slower than veterans without PTSD (η_p² = .06; 90% CI = [.00, .18]). We repeated this last analysis with logarithmic transformation of the retrieval-time data and replicated the nonsignificant results (ps > .123, η_p² < .049).

Negative memories and deployment memories

There were few negative memories and intrusions across groups (see Table 2), so we created a negative memory index by calculating, for each participant, the mean number of memories coded as describing a negative scene regardless of whether they were retrieved in response to the same sound as they were paired with during encoding. We analyzed the effects of time and diagnostic status on the negative memory index in a factorial ANOVA with retention interval (immediate vs. 1 week) and group (veterans with PTSD vs. non-PTSD Veterans) as between-subjects variables. No effects of this analysis were statistically significant (ps > .133, η_p² < .03; Table 2).

For deployment memories, we followed the same strategy as above. One PTSD veteran in the immediate retention interval had a very high number of deployment memories (k = 21) and was excluded from this analysis. Veterans with PTSD reported significantly more deployment memories compared with veterans without PTSD (p = .017, η_p² = .07; 90% CI = [.01, .16]; Table 2). In addition, these deployment memories were rated as more intense by veterans with PTSD (veterans with PTSD: M = 3.5, SD = 1.4; veterans without PTSD: M = 2.6, SD = 0.8), F(1, 39) = 4.80, p = .035, η_p² = .11; 90% CI = [.004, .27], and as accompanied with more bodily arousal (veterans with PTSD: M = 3.1, SD = 1.3; veterans without PTSD: M = 1.9, SD = 0.7), F(1, 39) = 8.66, p = .005, η_p² = .18; 90% CI = [.03, .35].

Retrospective rehearsal reports

We conducted a one-way ANOVA for each individual item from the rehearsal questionnaire, with group (veterans with PTSD vs. veterans without PTSD) as a between-subjects variable, for participants in the 1-week-retrieval condition. In general, the two groups did not differ significantly on self-reported frequency of memories for scenes during the 1-week retention interval (PTSD veterans: M = 1.3, SD = 0.8; veterans without PTSD: M = 1.1, SD = 0.7; p = .344, η_p² = .02; 90% CI = [.00, .14]). However, veterans with PTSD reported retrieving memories that almost exclusively referred to the negative war-related scenes (M = 1.1, SD = 0.8), whereas the responses from the veterans without PTSD suggested more mixed memories (M = 1.8, SD = 1.0; p = .017, η_p² = .16; 90% CI = [.01, .30]). In addition, veterans with PTSD reported their memories of the scenes to be more spontaneous (M = 1.0, SD = 1.3) than veterans without PTSD (M = 2.0, SD = 1.4; p = .034, η_p² = .13; 90% CI = [.01, .27]). This suggests that they experienced a higher proportion of involuntary memories of negative scenes during the 1-week delay.

Discussion

The aim of the present study was to examine the frequency, persistence, and emotional impact of involuntary memories of war-related and neutral scenes in veterans with and without PTSD using an experimental design. Compared with veterans without PTSD, we expected veterans with PTSD to report higher emotional intensity for involuntary memories of both war-related and neutral scenes (e.g., Rubin et al., 2008, 2011). We also expected the emotional intensity to persist over a delay of 1 week in veterans with PTSD only. Because of reduced inhibitory control, we expected veterans with PTSD to report a higher frequency of involuntary memories than veterans without PTSD (Mary et al., 2020). Finally, we expected veterans with PTSD to report a higher frequency of involuntary memories of war-related scenes in response to neutral cues associated with both war-related and neutral events (Catarino et al., 2015).

We found that veterans with PTSD reported higher ratings of both bodily reaction and emotional intensity in response to retrieval of involuntary memories. For emotional intensity, this main effect was qualified by an interaction with retention interval, reflecting a considerably larger group difference after a week. No overall group difference was observed for specificity ratings, but veterans without PTSD reported significantly lower memory specificity after 1 week, whereas veterans with PTSD reported the same level of memory specificity at both retention intervals, reflected in a significant Group × Delay interaction (Fig. 2).

Consistent with our first two predictions, veterans with PTSD showed persistent and increased emotional impact of war-related and neutral involuntary memories even after a delay of 1 week. Given that there were only few negative memories reported overall, the increased emotional impact of memories reported by veterans with PTSD was based in large part on memories for neutral scenes, consistent with the idea that increased affect intensity in PTSD influences memory for both neutral and trauma-related events (Rubin et al., 2008). Rubin et al. (2008) further suggested that because involuntary retrieval is uncontrolled and therefore leaves little to no room for emotion regulation, the memories are experienced as emotionally intense and intrusive. The effect of group on emotional intensity became apparent only after a week, when ratings of emotional intensity were particularly low in veterans without PTSD and particularly high in veterans with PTSD. The elevated intensity after a week was consistent with our prediction of general elevated affect intensity in PTSD (e.g., Rubin et al., 2008, 2011) reflected in intensity ratings of the memories. However, we had expected this effect also to be present at the immediate test, in which only a small numerical difference in the expected direction was seen. This pattern of results is consistent with a similar pattern of findings in an earlier study in which a variant of the present design in healthy individuals was used and may be related to consolidation processes operating at longer delays (Staugaard & Berntsen, 2014).

We did not find any effects of PTSD diagnosis on the overall frequency on involuntary memories or specifically on the proportion of involuntary memories of war-related scenes retrieved in response to neutral cues. The frequency of such negative memories did not change between the two retention intervals in either group. Consistent with the idea of inhibition deficits (e.g., Catarino et al., 2015), we did find that veterans with PTSD reported a higher frequency of autobiographical memories related to their own prior deployments while they took part in the experiment, and these autobiographical memories were also more emotionally intense and associated with higher bodily arousal compared with veterans without PTSD.

In addition to inhibitory deficits, this latter finding could also reflect that deployment events are highly accessible for veterans with PTSD. For example, McNally et al. (1995) described how veterans with PTSD who wore fatigues and medals were more likely to retrieve war-related memories in response to cue words compared with veterans who did not wear regalia. McNally et al. interpreted this as resulting from a psychological fixation on the veterans’ wartime experiences. Berntsen and Rubin (2006, 2007) presented a similar view but identified the specific mechanism behind this psychological fixation as “centrality of event,” which refers to how past events are integrated into a person’s life story. The more meaningful and important individuals perceive a past event to be, the more central to their life story it becomes. This again increases the accessibility of the event in memory, making it very likely to appear in response to a variety of cues, even those with weaker associative strength. In the present study, we used the nondistinctive sounds of dogs barking, a busy street, and rainfall, which cued, for example, the following deployment memories: “the homeless and wild dogs that often haunted the camp looking for something to eat,” “market in the Middle-East,” and “a rainy patrol.” Dogs barking, busy streets, and rain falling are very commonly occurring sounds (both in the real world and in our experimental design) and are likely associated with numerous neutral or even positive personal experiences, many of which are probably more recent than the deployment experiences. However, the deployment memories are likely to be much more central to the life story and identity of veterans with PTSD compared with veterans without PTSD, which could explain why they “intrude” into the experimental setting for veterans with PTSD in particular (e.g., Berntsen & Rubin, 2006). In support of this interpretation, we did find that veterans with PTSD reported much higher scores on the CES for their trauma than veterans without PTSD.

To further test the frequency and valence of involuntary memories, we included two retrospective measures of frequency. One was the IAMI, measuring the general frequency of involuntary memories during daily living, in which veterans with PTSD reported considerably higher frequencies of involuntary memories and future projections, consistent with previous findings showing positive correlations between the IAMI and measures of PTSD symptoms (Berntsen et al., 2015). The other was a rehearsal questionnaire, specifically measuring frequency, valence, and retrieval effort of memories of the scenes from the encoding task during the 1-week delay. This was answered only by the participants in the 1-week condition. Here, both veterans with PTSD and those without PTSD reported similar memory frequency overall, but the memories of veterans with PTSD were rated more negative and spontaneous. This suggests that veterans with PTSD retrospectively reported a higher proportion of involuntary memories of the war-related scenes compared with veterans without PTSD. As above, these findings can be discussed in relation to reduced inhibition. Because the IAMI measures self-reported frequency of a wide range of involuntary memories, the finding that veterans with PTSD reported higher frequencies is consistent with the notion of a general lack of inhibitory control (Catarino et al., 2015; Mary et al., 2020).

Taken together, the pattern of findings suggests that inhibitory failure may be particularly prevalent for personal memories perceived as highly central to identity (i.e., deployment memories) and for retrospectively reported memories perceived as less central to identity (i.e., experimentally induced memories and everyday involuntary memories). However, we did not observe a similar pattern for experimentally induced memories of trauma-related scenes, which could indicate that veterans with PTSD may have been able to inhibit these stimuli at least to some extent. Albeit speculative, one explanation could be that veterans in our design were expecting to confront reminders of the war-related scenes during retrieval, so they were prepared to inhibit them. In contrast, the deployment memories could have entered awareness unexpectedly (as intrusions) during the retrieval task, overruling any conscious attempt at inhibition.

A strength of the present study is the reporting of involuntary memories as they happen in an experimental design, manipulating both encoding and retrieval. This allowed us to provide a direct test of the theoretical position that PTSD is associated with stronger emotional responses to a variety of memories and, thus, that part of the mechanisms underlying intrusive memories of traumatic events is a general elevated tendency to respond to memories with strong emotion. By using an experimental design, the present study effectively rules out alternative explanations related to types of events or types of cues.

However, the present study also has some limitations that could affect the interpretation of the results. It is possible that memories retrieved during the experimental task are not directly comparable with autobiographical memories happening during daily living. For one, the experimentally induced memories are not personally relevant and do not share many of the characteristics of real-world memories. Future research might be able to further increase ecological validity by using technologies such as virtual reality, which enables an immersive experience closer to real life.

Most veterans with PTSD were tested at their treatment clinic, and this context could have cued deployment memories. On the other hand—as discussed above—the main source of deployment memories appeared to be cue-overlap between the sounds in the retrieval task and elements of the deployment memories (e.g., dogs or cars). In addition, the fact that veterans with PTSD had all completed their treatment could mean that their responses to the task does not directly compare with nontreated veterans. Specifically, treated veterans could have developed increased awareness of intrusions and learned strategies to reduce their frequency. It is also possible that some veterans with PTSD had recovered enough to no longer meet criteria for a PTSD diagnosis. However, their depression and posttraumatic symptom scores as a group were still in the clinical range, reflecting the often chronic nature of PTSD. Finally, we did not collect data on ethnicity, income, or socioeconomic status and are therefore unable to test the influence of these factors. However, the present sample was a generally homogeneous group consisting mostly of White men of similar educational background.

We also did not have enough women in the sample to consider gender differences. This could be a very interesting avenue for future research given that we recently found a very similar pattern of results using a similar design in a healthy sample of young men and women. In that study, women reported higher emotional intensity and bodily reaction to negative—but not neutral—scenes, but there were no differences on memory frequency (Staugaard & Berntsen, 2021).

In summary, veterans with PTSD reported increased emotional impact and maintained specificity after a retention interval of 1 week for memories of both war-related and neutral scenes. We did not find an effect of diagnostic status on frequency of involuntary memories in general or specifically regarding negative involuntary memories in the experimental study. However, we did find a higher incidence of deployment memories reported during the experimental task in veterans with PTSD, consistent with the centrality of events theory (Berntsen & Rubin, 2006) and the notion of inhibitory failure (Catarino et al., 2015; Mary et al., 2020). In addition, veterans with PTSD retrospectively reported memories that were more negative and spontaneous after a 1-week delay. The findings are consistent with PTSD being associated with stronger emotional responses to a range of memories, not limited to memories of the traumatic event or to trauma-related material. The findings suggest that targeting any type of personal memory—and not just the traumatic memories—could have clinical relevance in PTSD. The findings show that not just intrusions related to the index trauma are central in PTSD. In addition, involuntary memories of more mundane events are associated with increased emotional response, for which reason retrieval processes in general, such as appraisal, emotion regulation, and monitoring, might be key targets for intervention.

Supplemental Material

sj-pdf-1-cpx-10.1177_2167702621994544 – Supplemental material for Involuntary Memories of War-Related Scenes in Veterans With PTSD

Supplemental material, sj-pdf-1-cpx-10.1177_2167702621994544 for Involuntary Memories of War-Related Scenes in Veterans With PTSD by Søren Risløv Staugaard, Annette Kjær Fuglsang and Dorthe Berntsen in Clinical Psychological Science

Footnotes

Acknowledgements

We are grateful to Nina Bylod Staugaard, Cecilie Moesgaard, Mette Sørensen, Heidi Kamp Lærke Madsen, Meike Bohn, Marie Lund, and Astrid Lilhav Riis for their assistance with recruitment and data collection. The data of the present study were presented at two internal meetings at the Center on Autobiographical Memory Research.

Transparency

Action Editor: Erin B. Tone

Editor: Kenneth J. Sher

Author Contributions

S. R. Staugaard and D. Berntsen developed the study design and hypotheses. A. K. Fuglsang was clinically responsible for patients at the clinic for posttraumatic stress disorder and anxiety in Aarhus and assisted with recruitment and materials. S. R. Staugaard performed the data analysis and interpretation in collaboration with D. Berntsen. S. R. Staugaard and D. Berntsen drafted the manuscript, and D. Berntsen and A. K. Fuglsang provided critical revisions. All of the authors approved the final manuscript for submission.

ORCID iD

Søren Risløv Staugaard

Supplemental Material

Additional supporting information can be found at

Notes

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.).

Beck

A. T.

Steer

R. A.

Brown

G. K.

(1996). Manual for the Beck Depression Inventory-II. Psychological Corporation.

Berntsen

Rubin

D. C.

(2006). Emotion and vantage point in autobiographical memory. Cognition and Emotion, 20, 1193–1215. https://doi.org/10.1080/02699930500371190

Berntsen

Rubin

D. C.

(2007). When a trauma becomes a key to identity: Enhanced integration of trauma memories predicts posttraumatic stress disorder symptoms. Applied Cognitive Psychology, 21, 417–431. https://doi.org/10.1002/acp.1290

Berntsen

Rubin

D. C.

Salgado

(2015). The frequency of involuntary autobiographical memories and future thoughts in relation to daydreaming, emotional distress, and age. Consciousness and Cognition, 36, 352–372. https://doi.org/10.1016/j.concog.2015.07.007

Berntsen

Staugaard

S. R.

Sørensen

L. M. T.

(2013). Why am I remembering this now? Predicting the occurrence of involuntary (spontaneous) episodic memories. Journal of Experimental Psychology: General, 142, 426–444. https://doi.org/10.1037/a0029128

Blevins

C. A.

Weathers

F. W.

Davis

M. T.

Witte

T. K.

Domino

J. L.

(2015). The Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5): Development and initial psychometric evaluation. Journal of Traumatic Stress, 28, 489–498. https://doi.org/10.1002/jts.22059

Boykin

D. M.

Teng

E. J.

(2019). A proposal for augmenting the measurement of index events in PTSD assessment using event centrality. Anxiety, Stress & Coping, 32, 559–567. https://doi.org/10.1080/10615806.2019.1638682

Brewin

C. R.

Gregory

J. D.

Lipton

Burgess

(2010). Intrusive images and memories in psychological disorders: Characteristics, neural basis, and treatment implications. Psychological Review, 117, 210–232. https://doi.org/10.1037/a0018113

10.

Catarino

Küpper

C. S.

Werner-Seidler

Dalgleish

Anderson

M. C.

(2015). Failing to forget: Inhibitory-control deficits compromise memory suppression in posttraumatic stress disorder. Psychological Science, 26(5), 604–616. https://doi.org/10.1177/0956797615569889

11.

Ehlers

Hackmann

Michael

(2004). Intrusive re-experiencing in post-traumatic stress disorder: Phenomenology, theory, and therapy. Memory, 12, 403–415. https://doi.org/10.1080/09658210444000025

12.

Hall

S. A.

Brodar

K. E.

LaBar

K. S.

Berntsen

Rubin

D. C.

(2018). Neural responses to emotional involuntary memories in posttraumatic stress disorder: Differences in timing and activity. NeuroImage: Clinical, 19, 793–804. https://doi.org/10.1016/j.nicl.2018.05.009

13.

James

E. L.

Lau-Zhua

Clark

I. A.

Vissera

R. M.

Hagenaars

M. A.

Holmes

E. A.

(2016). The traumafilm paradigm as an experimental psychopathology model of psychological trauma: Intrusive memories and beyond. Clinical Psychology Review, 47, 106–142. https://doi.org/10.1016/j.cpr.2016.04.010

14.

Kleim

Graham

Bryant

R. A.

Ehlers

(2013). Capturing intrusive re-experiencing in trauma survivors’ daily lives using ecological momentary assessment. Journal of Abnormal Psychology, 122, 998–1009. https://doi.org/10.1037/a0034957

15.

Kubany

E. S.

Haynes

S. N.

Leisen

M. B.

Owens

J. A.

Kaplan

A. S.

Watson

S. B.

Burns

(2000). Development and preliminary validation of a brief broad-spectrum measure of trauma exposure: The Traumatic Life Events Questionnaire. Psychological Assessment, 12, 210–224. https://doi.org/10.1037/1040-3590.12.2.210

16.

Lang

P. J.

Bradley

M. M.

Cuthbert

B. N.

(2008). International affective picture system (IAPS): Affective ratings of pictures and instruction manual (Technical report A-8). The Center for Research in Psychophysiology, University of Florida.

17.

Marks

E. H.

Franklin

A. R.

Zoellner

L. A.

(2018). Can’t get it out of my mind: A systematic review of predictors of intrusive memories of distressing events. Psychological Bulletin, 144, 584–640. https://doi.org/10.1037/bul0000132

18.

Mary

Dayan

Leone

Postel

Fraisse

Malle

Vallée

Klein-Peschanski

Viader

Sayette

Peschanski

Eustache

Gagnepain

(2020). Resilience after trauma: The role of memory suppression. Science, 367(6479), Article eaay8477. https://doi.org/10.1126/science.aay8477

19.

McNally

R. J.

Lasko

N. B.

Macklin

M. L.

Pitman

R. K.

(1995). Autobiographical memory disturbance in combat-related posttraumatic stress disorder. Behaviour Research and Therapy, 33(6), 619–663. https://doi.org/10.1016/0005-7967(95)00007-k

20.

Nicholson

E. L.

Bryant

R. A.

Felmingham

K. L.

(2014). Interaction of noradrenaline and cortisol predicts negative intrusive memories in posttraumatic stress disorder. Neurobiology of Learning and Memory, 112, 204–211. https://doi.org/10.1016/j.nlm.2013.11.018

21.

Pitman

R. K.

(1988). Post-traumatic stress disorder, conditioning, and network theory. Psychiatric Annals, 18(3), 182–189. https://doi.org/10.3928/0048-5713-19880301-11

22.

Psychology Software Tools (2017). E-Prime: Documentation article. https://support.pstnet.com.

23.

Rubin

D. C.

Boals

Berntsen

(2008). Memory in posttraumatic stress disorder: Properties of voluntary and involuntary, traumatic and nontraumatic autobiographical memories in people with and without posttraumatic stress disorder symptoms. Journal of Experimental Psychology: General, 137(4), 591–614. https://doi.org/10.1037/a0013165

24.

Rubin

D. C.

Dennis

M. F.

Beckham

J. C.

(2011). Autobiographical memory for stressful events: The role of autobiographical memory in posttraumatic stress disorder. Consciousness and Cognition, 20, 840–856. https://doi.org/10.1016/j.concog.2011.03.015

25.

Staugaard

S. R.

Berntsen

(2014). Involuntary memories of emotional scenes: The effects of cue discriminability and emotion over time. Journal of Experimental Psychology: General, 143(5), 1939–1957. https://doi.org/10.1037/a0037185

26.

Staugaard

S. R.

Berntsen

(2021). Gender differences in the experienced emotional intensity of experimentally induced memories of negative scenes. Psychological Research, 85, 1732–1747. https://doi.org/10.1007/s00426-020-01334-z

27.

Weathers

F. W.

Litz

B. T.

Keane

T. M.

Palmieri

P. A.

Marx

B. P.

Schnurr

P. P.

(2013). The PTSD Checklist for DSM-5 (PCL-5). U.S. Department of Veteran Affairs, National Center for PTSD. https://www.ptsd.va.gov/professional/assessment/adult-sr/ptsd-checklist.asp

28.

World Health Organization. (2004). ICD-10: International statistical classification of diseases and related health problems: Tenth revision (2nd ed.). https://apps.who.int/iris/bitstream/handle/10665/42980/9241546530_eng.pdf

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB