Changing beliefs in repressed memory and dissociative amnesia

In three studies, we examined whether beliefs in repressed memory and dissociative amnesia could be changed. Participants provided agreement ratings to statements related to repressed memory and dissociative amnesia. Then, they received a university course which included education on the science of memory. Following this, participants had to re-rate the statements. In Study 3, at Times 1 and 2, participants also received a case vignette on a therapy-induced recovered memory and rated several statements related to this case. Participants who received education on the science of memory were less likely to agree with statements endorsing repressed memory and dissociative amnesia — and participants were more likely to state that the case vignette involved a false memory. Providing education on the science of memory can help people (e.g., legal professionals, people from the general population) to use critical thinking on the topic of repressed memory and dissociative amnesia.

some clinicians continued to claimed to see patients with unconsciously repressed memory.Some clinicians reasoned-along the lines of the Freudian tradition-that some patients' symptoms (e.g., sexual dysfunctions, anxiety, insomnia) were caused by hidden memories of trauma.For example, Van der Kolk and Fisler (1995, p. 512) argued that trauma can be "entirely organized on an implicit or perceptual level, without an accompanying narrative about what happened."To achieve clinical improvement, the argument went that the hidden memories of trauma should be recovered using therapeutic techniques.In contrast, some memory researchers argued that in attempting to exhume memories therapists could apply suggestive interview techniques that could lead to false memories (Loftus & Ketcham, 1994;Otgaar et al., 2022).In certain situations, these false memories resulted in false accusations and wrongful convictions, tearing families apart (Loftus, 1994;Otgaar et al., 2022).
Not only is there a concern that trying to unlock repressed memories could lead to false memories (Otgaar et al., 2019), there is also a concern that the empirical support for the existence of repressed memory is lacking.Some scholars have argued that claims of repressed memory could be explained by more plausible mechanisms.
For example, victims oftentimes do not want to think or talk about trauma and that this is sometimes mislabeled as a repressed memory (e.g., McNally, 2005).Also, claims of repressed memory are oftentimes quite easily explained with ordinary forgetting mechanisms or failures to encode (part of) the events (Otgaar et al., 2019).Concerning recovered memories, research has shown that people incorrectly claim that prior to a memory recovery they never had previously remembered the event, while they actually had remembered it previously, a phenomenon called the forget-it-all-along effect (e.g., Janssen et al., 2022).Finally, there is evidence showing that sometimes individuals do not fully realize that an inappropriate event is abusive at the time (especially in childhood), and only later reinterpret it as emotionally disturbing-this is sometimes mislabeled as a repressed memory (McNally & Geraerts, 2009).
The debate over whether memories were falsely recovered in therapy has been called the memory wars (Crews, 1995).Although scholars had argued that the memory wars were over (e.g., McHugh, 2003), recent evidence shows that the controversy surrounding the topic of repressed memory continues to circulate in academic, clinical, and legal contexts (Brewin, 2021;Otgaar et al., 2019Otgaar et al., , 2021)).For instance, research has shown that many populations ranging from students (Dodier et al., 2021) to professionals (e.g., clinical psychologists) strongly believe in the existence of repressed memory (Otgaar et al., 2019;Sumampouw et al., 2022).Specifically, Otgaar et al. (2019) reviewed studies in which people were surveyed regarding their belief in repressed memory.They found that 58% (n = 4745) indicated to believe to some extent in the concept of repressed memory.Furthermore, 70% (n = 2305) of surveyed clinical psychologists indicated a belief in the existence of repressed memories.This endorsement of repressed memories was lower in the 1990s (61%; n = 719) and it increased to 76% (n = 1586) from 2010 onward showing that the belief in repressed memory may have become more pronounced.To put these percentages in broader perspective, Patihis et al. (2021) asked 17 memory experts a series of statements on the functioning of memory (e.g., "Traumatic memories are often repressed").Memory experts were for the most part skeptical toward the concept of repressed memory and, on average, did not agree with the idea of repressed memory (mean agreement rating: 2.30 on a scale from 1 = strongly disagree and 6 = strongly agree; SD = 1.40).These discrepant views on repressed memory between memory experts and clinicians suggest that the memory wars debate continues today.
It is equally important to stress that in questionnaires that asked specifically about a belief in unconscious repressed memories, many people endorsed the idea (for a discussion on this see Otgaar et al., 2021).For example, in Otgaar et al. (2020), participants received the following statement "traumatic memories are often repressed."If people agreed with this statement, additional questions were posed that investigated whether they meant those traumatic memories are accessible during repression, and another question on whether they are unconscious during repression, or not.The results demonstrated that 89.5% (n = 909) agreed to some extent that traumatic memories can be repressed and, of those, 73.7% (n = 670) agreed that such memories are inaccessible, and 80.9% (n = 735) agreed that such memories are unconscious.Taken together, these results suggest that many people believe in the controversial concept of unconscious repression.
Relatedly, some scholars also seem to hold controversial beliefs on unconscious repression in academic writings (for an overview, see Otgaar et al., 2021).For example, Van der Hart and Nijenhuis (1995) described that memory loss due to trauma "involves a reversible memory impairment in which memories of personal experience cannot be retrieved in a verbal form, or, if temporarily retrieved, cannot be wholly retained in consciousness" (p. 1).To give another example, Kessler et al. (2017) wrote that "repressed mental contents are less accessible to conscious processing and still underlie dynamic processes.They might, for instance, generate associated material and intrude into awareness.In the case, that repressed material gains access to consciousness, it elicits a secondary repression excluding it once more from consciousness" (p. 1).These examples show that the concept of unconscious repression continues to be discussed within academic circles (see also Brand et al., 2018;Dodier, 2019;Merckelbach & Patihis, 2018;Patihis, Otgaar, et al., 2019).Although it is problematic that the belief in repressed memory is prevalent among many different populations, it is especially perilous among clinical psychologists.Clinical psychologists who endorse the notion of repressed memory may suggestively search for hidden traumas when treating their patients-which could lead to the formation of false memories.
Another worrisome issue in the resurgence of repressed memory is that the concept is sometimes disguised under a different name: Dissociative amnesia (Mangiulli et al., 2022;Radcliffe & Rix, 2019).
Dissociative amnesia is defined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013, p. 298) as an "inability to recall autobiographical information" that (1) is "usually of a traumatic or stressful nature," (2) is "inconsistent with ordinary forgetting," (3) should be "successfully stored", (4) involves a period of time when there is an "inability to recall," (5) is not caused by "a substance" or "neurological…condition," and ( 6) is "always potentially reversible because the memory has been successfully stored."The definition of dissociative amnesia fits seamlessly with the definition of repressed memory (see Otgaar et al., 2019 for a detailed comparison).For example, dissociative amnesia contains similar problematic components to repressed memory-central to which being the idea that autobiographical memories can be blocked due to traumatic experiences.Furthermore, and in line with studies on beliefs on repressed memory (e.g., Dodier et al., 2021;Otgaar et al., 2019;Sumampouw et al., 2022), people tend to agree with controversial statements regarding both the cause and the underlying mechanism of dissociative amnesia.Specifically, Mangiulli et al. (2021) showed that more than 60% of their sample (N = 1017) agreed with dubious notions such as that being exposed to trauma during childhood would increase the likelihood of developing amnesia for those events.Hence, a critical eye should be given to the concept of dissociative amnesia, as well as repressed memory.
The problematic status of dissociative amnesia was exemplified in a recent study by Mangiulli et al. (2022) who reviewed 128 case studies on dissociative amnesia in the period 2000-2020.They found that none of the published case studies actually met the diagnostic criteria of dissociative amnesia as described in the DSM-5.The heterogenous nature of these case studies might be because dissociative amnesia lacks precise and objectively measurable signs and symptoms in the DSM-5.Mangiulli et al. (2022) found that more plausible alternative explanations for the claimed dissociative amnesia, such as ordinary forgetting or malingering, were frequently not considered in these case studies.Even more interesting, in some of the cases no traumatic precursor was even indicated or found prior to the purported dissociative amnesia.
Overall, there is a continued belief in the concept of repressed memory which might be damaging in the courtroom, thereby potentially contributing to wrongful convictions.As well as demonstrating that the debate on repressed memory lingers on, recently, scholars have become interested in investigating whether controversial beliefs in repressed memory can be corrected (Sauerland & Otgaar, 2022).

| CHANGING BELIEFS IN REPRESSED MEMORY
One way to view the concept of repressed memory is to regard its prevalence in society as being caused by some form of misinformation.Specifically, the type of publicly spread misinformation that contains information that is not supported by science (often called fake news: e.g., that vaccination is the major cause of autism).Because the idea underpinning repressed memory contains controversial elements that also are not supported by science (e.g., unconscious memories of trauma that can accurately be retrieved after many years), one way is to treat the idea of repressed memory as misinformation.Substantial inroads have been made in the area of correcting misinformation such as remedying fake news on the effectiveness of vaccines (e.g., Lewandowsky et al., 2012).One recurrent finding is that although misinformation can be corrected (see Greenspan & Loftus, 2021;Oeberst et al., 2021), it sometimes is not completely eliminated; a phenomenon also known as the continued influence effect (Ecker et al., 2022;Walter & Tukachinsky, 2020).
Several interventions have been devised that might potentially abolish or reduce the dire effects of misinformation (also called debiasing methods; Lilienfeld et al., 2009) such as providing preexposure warnings and repeatedly debunking the misinformation (Ecker et al., 2010;Ecker et al., 2011).Many of these interventions have been designed to switch cognitive processing from an automatic, heuristic style of thinking (e.g., System 1) to a more controlled style of thinking (e.g., System 2; Croskerry et al., 2013;Kahneman, 2003;Stanovich & West, 2000).The rationale behind this switching is that certain biases, like the belief in repressed memory, operate relatively automatically.In order to reduce those biases, people should become more aware of the intricacies of these biases.Such awareness might inform them about the erroneous nature of these biases.
Of interest for the current studies is research showing that education might affect people to shift to a more controlled mode of thinking.That is, research has revealed that education concerning specific biases can make people less vulnerable for certain cognitive errors such as the confirmation bias (e.g., van Stekelenburg et al., 2021).For example, Evans et al. (1994;Experiment 3) provided participants with elaborate instructions on principles in reasoning.Participants receiving such instructions were least likely to accept invalid conclusions based on syllogisms.Similarly, in the area of repressed memory, education could involve information regarding the science of memory.The idea of providing education to diminish the intensity of biases runs parallel to related effective debiasing methods such as providing alternative accounts and emphasizing the facts of a certain topic (Lewandowsky et al., 2012).Furthermore, this idea is linked to research showing that in order to correct misconceptions among students, scientific information should be provided that refutes these misconceptions (Guzzetti, 2000;e.g., lectures on the science of memory).
In a recent study, Sauerland and Otgaar (2022) provided students involved in a Legal or Forensic Psychology master program with an educational course on the science of memory.In this course, lectures and readings were included involving-among others-the fallibility of memory, eyewitness memory, and the effects of trauma on memory.
At the start and end of the course, students were surveyed on their beliefs about different aspects of memory, including their beliefs in repressed memory.The most striking finding was that students became more skeptical toward the concept of repressed memory at the end of course compared with at the start.This belief change persisted after a long delay (6 or 18 months later).
Although promising, the study was limited in several aspects.For one, in Sauerland and Otgaar's (2022) study, there were only 2 statements related to repressed memory that were surveyed at the start and end of the course (i.e., "The mind is capable of unconsciously "blocking out" memories of traumatic events," "A poor memory for childhood events is indicative of a traumatic childhood").Scholars have criticized past survey work on repressed memory because oftentimes only a small set of belief statements was used to examine repressed memory (Brewin et al., 2020).This is problematic because repressed memory is a complex concept which is difficult to capture in merely one or two statements.Second, Sauerland and Otgaar's (2022) study primarily focused on changing beliefs in repressed memory.Dissociative amnesia encompasses similar problematic connotations as repressed memory and is especially popular in clinical settings and is in the DSM-5.It is presently unknown whether beliefs in dissociative amnesia can also be changed when basic education on the science of memory is provided.Hence it is of interest to examine whether beliefs in repressed memory and dissociative amnesia can be changed when knowledge is presented concerning the functioning of memory.
From a practical perspective, research on the effect of scientific education on beliefs about repressed memory and dissociative amnesia is vital as it might lead to important insights in how these two ideas can be debunked among many professionals.For example, both debunking and correction might be crucial to lower the chance that therapists might engage in suggestive treatment practices to uncover ostensibly hidden traumatic memories.If so, false accusations and wrongful convictions might be prevented.Furthermore, such corrections are not only relevant for professionals but can also be important for the general public.Raising awareness about the controversies surrounding the concept of repressed memory and dissociative amnesia can be beneficial for people from the general public who might know patients undergoing (suggestive) therapy, who themselves undergo treatment, or who might become jurors in legal cases involving repressed memory recovery.Developing a more critical attitude toward the concepts of repressed memory and dissociative amnesia might even make people more reluctant to go along with suggestions that they have hidden memories of trauma.

| THE PRESENT STUDIES
In the current studies, the goal was to investigate whether beliefs in repressed memory and dissociative amnesia could be corrected when educating participants on the science of memory.Specifically, we tested students' beliefs on repressed memory and dissociative amnesia at the start and end of a course which included education on the science of memory.Importantly, we used a wide variety of statements covering different aspects of repressed memory and dissociative amnesia.For example, while in previous research a focus was put on whether traumatic memories could be repressed, we also focused on other problematic aspects of repressed memory and dissociative amnesia.
That is, we included statements such as whether unconscious memories of trauma can lead to depressive symptoms and whether people can develop multiple identities.Our chief prediction was that although students would display a strong belief in repressed memory and dissociative amnesia at the start of the course, this belief would statistically significantly drop at the end of it.Specifically, we predicted that belief ratings would be lower for statements on repressed memory and dissociative amnesia after participants received knowledge about the science of memory, compared to before.

Statements
In the current study, we used 35 statements on topics related to the field of psychology (i.e., memory and lie detection).Of these 35 statements, 23 statements were directly related to the topic of repressed memory and dissociative amnesia (see below) and only these 23 statements were used for our analyses.The statements were adopted from previous studies (e.g., Otgaar et al., 2021;Patihis et al., 2014;Sauerland & Otgaar, 2022; see https://osf.io/mg2kz).Statements referring to various aspects related dissociative amnesia were newly created.
Specifically, the first two authors (HO, IM) devised several statements related to dissociative amnesia and this list was then assessed by the last author (LP).After discussions and revisions among the three authors, a final list of statements was developed.Statements had to be rated on 7-points Likert scales (1 = completely disagree, 7 = completely agree).

Course on psychology and law
Participants were enrolled in a course (in English) on psychology and law that starts every semester at the end of September/beginning of October to the end of December 2020.Students received weekly lectures of 2 h on diverse topics related to psychology and law (e.g., expert witnesses, biases).Of importance for the current study were lectures specifically devoted to eyewitness memory, false memory, trauma, and memory.Specifically, related to trauma and memory, lectures were provided that discussed the general functioning of memory (e.g., reconstructive nature, forgetting curve), the formation of false memories, the memory wars, and the controversial elements behind repressed memory and dissociative amnesia.These themes were also taught in Studies 2 and 3.Because of the COVID-19 pandemic, some lectures were given online.Of note, in all studies, students did not receive any psychology-related courses at the Faculty of Law and Criminology which could have affected their belief ratings.In the Appendix A, we show the specific topics provided to the students.

| Design and procedure
Participants were involved in a within-subjects design.During the first lecture of the course, students were asked to participate in a study on memory (i.e., Time 1).A link to a Qualtrics survey was shared on a digital learning page of the course.If students clicked on the link, they were directed to the survey and asked for their informed consent.If they agreed to participate, they received several (demographic) questions (i.e., age, gender, ethnicity, level of education, and English proficiency).Also, to assure anonymity and link participants' responses, in all studies, participants were asked to generate a personal code by filling in their date of birth and the last two letters of their first name.
Following this, they were presented with several statements and were asked to provide their level of agreement.Intermixed between the statements were two attention check questions 4 (i.e., "What is 2 + 2?," "Please respond 'black' to the following").During the last lecture of the course, they were again asked to complete the online survey (i.e., Time 2), approximately after 11 weeks from the first lecture.The procedure was identical to the first one.When all statements were rated, they received an online debriefing.

| RESULTS AND DISCUSSION
We only included participants that rated all statements at Time 1 and 2. A total of 20 students were included in the final analyses.Paired-samples t-tests were conducted on the agreement ratings (twotailed) 5 .The results can be found in Tables 1 and 2. As can be seen, except for the statement that dissociative amnesia can lead to new identities, all comparisons were statistically significant with Cohen's d effect sizes ranging between .43-1.59.When we applied a more stringent significance level because of the multiple comparisons (.05/23 = .002),11 comparisons were statistically significant.
Taken together, data of Study 1 clearly demonstrated that students became more critical toward the concept of repressed memory and dissociative amnesia and their underlying mechanisms when they received education on the science of memory.Specifically, students were less likely to agree with statements on repressed memory and dissociative amnesia after they received extensive education on the science of memory.Our results are in line with previous research by Sauerland and Otgaar (2022), albeit with the exception that in the current study we included more statements on repressed memory and used-for the first time-a wide variety of statements on dissociative amnesia.Although the current findings are promising, Study 2 was conducted to replicate the findings but now with a larger sample size.
As in Study 1, we hypothesized that providing participants with knowledge on the science on memory would make them more critical toward the existence of repressed memory and dissociative amnesia.Of the 555 registered responses, 237 completed the first survey.

Statements
The same statements were used as in Study 1.

Course on legal psychology
Participants followed a course (in Dutch) on legal psychology that starts every semester at the end of September/ beginning of October to the end of December 2020.Students received 3 h of weekly lectures and workshops on diverse topics related to legal psychology (e.g., investigative interviewing, expert witnesses, biases).Of importance for the current study were lectures specifically focused on eyewitness memory, false memory, and trauma and memory functioning.
Because of the COVID-19 pandemic, some lectures were provided online.

| Design and procedure
The same design and procedure were applied as in Study 1.

| RESULTS AND DISCUSSION
Paired samples t-tests were conducted to examine whether students became more critical regarding repressed memory and dissociative amnesia after receiving education on these topics.Tables 3 and 4 provide the statistics (e.g., means, effect sizes) of these analyses.As can be noticed, all comparisons were statistically significant with effect sizes ranging between .75-2.48.These effects remained statistically significant when a stricter significance level was used to account for multiple comparisons (=.002).Thus, using a larger sample size, we found a similar pattern of data as in Study 1.That is, when students did not have any detailed knowledge on repressed memory and dissociative amnesia, they were more likely to agree that repressed memories and dissociative amnesia exist.However, after they received education on the science of memory, they were less likely to endorse statements on repressed memory and dissociative amnesia.Thus, the aims of Study 3 were the following.First, we wanted to replicate the results obtained in Studies 1 and 2. Second, our interest was to dig more deeply in people's beliefs concerning repressed memory and dissociative amnesia by asking additional follow-up questions.
As in the previous studies, we expected that participants would believe less in repressed memory and dissociative amnesia after receiving scientific knowledge on the functioning of memory.Third, we included a case vignette on therapy-induced recovered memories and predicted that students would be more likely to label such memories as potentially false after receiving teaching in memory science.

Statements
Although the same statements were used as in Studies 1 and 2, for some statements, we included additional follow up questions (see Tables 5 and 6).Also, for exploratory purposes, we included additional statements related to repressed memory and appraising memories of emotion.These new statements were added because of recent work showing that cognitive reappraisals can alter memories for emotions (Patihis, Cruz, et al., 2019).

Case vignette
Participants received a case vignette (193  whether (1) the young woman's memories were repressed and accurately recovered in therapy, (2) the memories of abuse reflected an event that she truly experienced, (3) the memories of abuse reflected an event that she did not experience and was false, and (4) she unconsciously forgot memories of the abuse to cope with the trauma.
Participants had to answer these questions by using 7-point Likert scales (1 = strongly disagree, 7 = strongly agree).

Course on criminological psychology
Participants followed a course (in Dutch) on Criminological Psychology that starts every second semester at the end of January/ beginning of February to the end of May 7 2021.Students received weekly lectures and workshops of 4 h on diverse topics related to legal psychology (e.g., false confessions, investigative interviewing, expert witnesses, and criminal behavior).Of importance for the current studies were lectures specifically focused on eyewitness memory, false memory, and trauma and memory.Because of the COVID-19 pandemic, some lectures were online.

| Design and procedure
The same design and procedure were applied as in Studies 1 and 2, except for the case vignette provided to participants after having received the statements (prior to the course and after the course).
Also, participants only received the follow-up questions when they agreed with the respective statements (i.e., they provided a rating of 5 (somewhat agree), 6 (agree), or 7 (strongly agree)).

| Memory statements
Paired samples t-tests were performed to examine the change in statement ratings before and after receiving a course on Criminological Psychology.Tables 5 and 6 provide the statistics of these changes.
Two important findings emerged.First, as in previous studies, most of the comparisons between Time 1 and 2 were statistically significant when a stricter significance level was applied (=.002).Second, 5 of the 7 follow up questions also produced statistically significant effects.Importantly, the number of participants receiving the follow up questions was substantially lower at Time 2 (i.e., more than half of a reduction) than at Time 1.

| Case vignette
We also examined participants' responses to the case vignette statements (see Tables 7 and 8).All statements showed statistically signifi-

| Exploratory correlational and joint analysis
To explore whether ratings on statements on repressed memory and dissociative amnesia were correlated, we created a composite score of statements on repressed memory and a composite score of dissociative amnesia statements.Specifically, the composite score was calculated by averaging all respective statements.For Study 1, we found that at Time 1, no statistically significant correlation emerged between the repressed memory composite and the dissociative amnesia composite (r(19) = .41,p = .07).However, at Time 2, a statistically significant correlation was detected (r(19) = .75,p < .001).For Study 2, at Time 1 and Time 2, a statistically significant correlation emerged (Time 1: r(107) = .55,p < .001;Time 2: r(107) = .77,p < .001).A similar pattern of correlations was also observed in Study 3 (Time 1: r (57) = .27,p = .04;Time 2: r(57) = .73,p < .001).
Also, to explore the size of the obtained effects in the present studies, we conducted a joint analysis on the observed effect sizes of the studies.When we averaged all effect sizes (n = 69) for the same statements that were provided across studies, we obtained a Cohen's d of 1.13 (SD = .47).We also compared the mean effect sizes between belief scores regarding statements on repressed memory (n = 24) and statements on dissociative amnesia (n = 45).An independent samples t-test was performed with effect sizes as dependent variable and whether they referred to statements on repressed memory or dissociative amnesia as between-subjects factor.A statistically significant effect was detected, (t(67) = 2.92, p = .005,Cohen's d = .74)with Cohen's ds being higher for statements on repressed memory (M = 1.34,SD = .48)than for statements on dissociative amnesia (M = 1.02,SD = .42).

| GENERAL DISCUSSION
Can controversial beliefs in repressed memory and dissociative amnesia be changed?The answer to this question is yes.In the current line of studies, students had to rate their beliefs concerning a diverse set  Dodier et al., 2021;Sumampouw et al., 2022).
Such studies present informative data on the widespread nature of repressed memory but are insofar limited as they merely present a static pattern of belief rates.
In the current studies, we replicated the finding that students strongly agree with the idea of repressed memory and dissociative amnesia.Most importantly, however, our findings offer some new contributions to the controversy on repressed memory.First, we showed that the belief in repressed memory and dissociative amnesia is not static but amenable to change.Specifically, across three studies, when students received scientific knowledge on the functioning of memory, they were less likely to agree with several dimensions related to repressed memory and dissociative amnesia.
The result that belief ratings dropped at Time 2 after receiving scientific knowledge on the science of memory might be due to the fact that this knowledge made students more informed and critical toward repressed memory, thereby perhaps shifting from an automatic to a more controlled mode of thinking (see e.g., Lilienfeld et al., 2009).However, this is speculation because we did not have any measurement to support the proposition that there was a switch from an automatic to a more controlled thinking.What could have happened is that after receiving education on the science of memory, participants could mobilize more knowledge concerning the topic of repressed memory and dissociative amnesia and therefore make better judgments.Second, our main finding is well in line with the results obtained in Sauerland and Otgaar's study (2022)  Taken together, these data showed that when more specific and detailed questions concerning repressed memory and dissociative amnesia were asked, students became more critical toward agreeing with them after receiving scientific knowledge on memory functioning.
Fourth, we also included a case vignette (Study 3) in which students had to read a case about a young person recovering a memory of sexual abuse during therapy.The rationale of inserting this case vignette was to examine whether students could apply their knowledge on memory in order to more critically evaluate this case.So, for example, at Time 1, students reading this case highly agreed with the statement that the memories of the person (i.e., Kelly) were repressed and accurately recovered in therapy.However, at Time 2, on average, students were less likely to agree with this statement than at Time 1.This shows that providing students with scientific sources on memory could potentially help them in making better judgments in recovered memory cases.
We also conducted an exploratory joint analysis on the effect sizes obtained in the current studies.We observed a Cohen's d of 1.13 across all studies and statements.This implies that about 85% of the Time 2 data is above the average of the mean of the Time 1 data (i.e., Cohen's U 3 ; Magnusson, 2021).Notably, we also found that when we compared the effect sizes of repressed memory statements with dissociative amnesia statements, the mean effect size was larger for the repressed memory (Cohen's d = 1.34) than dissociative amnesia statements (Cohen's d = 1.02).This suggests that the offered education was more effective in making students critical about repressed memory than dissociative amnesia.There are two explanations for this finding.First, students received more information concerning the controversial topic of repressed memory than of dissociative amnesia.Second, and relatedly, the concept of dissociative amnesia might have been more difficult to grasp than repressed memory because it is also strongly related to the field of mental disorders (e.g., Staniloiu & Markowitsch, 2014) which might have been more difficult to comprehend for students following a course at a law faculty.
We also performed exploratory correlational analyses between ratings on repressed memory and dissociative amnesia statements.An underlying reason for these analyses was to examine whether people regard the concept of repressed memory equivalent to dissociative amnesia; something that has been foreshowed by several scholars (e.g., Mangiulli et al., 2022;McNally, 2005;Otgaar et al., 2019;Porter et al., 2001).By creating composite scores, we indeed found evidence for this.Specifically, overall, we found that people who believed in the concept of repressed memory also believed in the concept of dissociative amnesia.Although the correlations of this pattern were higher at Time 2 and thus after receiving knowledge about the link between repressed memory and dissociative amnesia, this link was also evident at Time 1. Also, it might be fruitful to examine whether similar results occur when at Time 2 somewhat different variations of the same statements were applied.For example, instead of stating "Traumatic memories can be unconsciously repressed," one could also include "Traumatic memories cannot be unconsciously repressed."These variations might test whether participants are able to use their acquired knowledge when being confronted with completely new statements.Furthermore, such variations might protect from the possibility that, at Time 2, participants remember the statements and what they answered at Time 1.However, in the current studies, it is unlikely that because of the high number of statements and the long delay between Time 1 and Time 2 (i.e., about 3 months), participants would have remembered the exact content and their answers at Time 1.
Although the current samples did not consist of clinical psychologists, the tested participants were students from a law faculty.It is unlikely that they will become practitioners in clinical settings, but arguably they will work in legal settings as for example lawyers or judges.This is important as the topic of repressed memory also thrives in legal settings.For example, Benton et al. (2006) showed that 50% (n = 25) of the judges in their sample believed in the concept of repressed memory.Why this is dangerous is because in some countries such as France, the period (i.e., statute of limitations) to prosecute sexual crimes has been extended because legal scholars stipulated that traumatic memories can be repressed for many years (Dodier & Tomas, 2019).Therefore, given that the concepts of dissociative amnesia and repressed memory are (re)emerging in the courtroom (see Otgaar et al., 2019), it is imperative for future legal professionals to be aware of such controversial ideas and their possible consequences.Of course, prospective research should focus on replicating the current results with other groups such as clinicians or judges.
Finally, although there is a developing research area on misinformation belief and its corrections (e.g., Ecker et al., 2022), research attempting to correct beliefs in repressed memory and dissociative amnesia is currently limited.However, because of the clear applied relevance of doing such work in this area (e.g., lowering chance for false recovered memories in therapy), future research could focus on the causes of believing in repressed memory and dissociative amnesia and which mechanisms underlie the current observed belief changes.
For example, research could examine whether the belief in repressed memory is linked to "the illusory truth effect" (e.g., Henderson et al., 2022) wherein people start to believe in repressed memories because of repeated (false) exposure of its existence (e.g., due to movies, newspapers, etc.).Furthermore, research could focus on whether the observed belief changes of the current studies are partly caused by participants believing in the content of the lectures (e.g., Rich et al., 2017).Finally, the concept of repressed memory is perhaps sometimes more difficult to debunk than plain fake news, as this latter one is more likely to be easily identified as false.Hence, future research enterprises might examine whether this also implies that stronger corrective measures are necessary to "completely" eradicate the belief in repressed memory.
To recap, scholars have sometimes argued that research on the belief in repressed memory is limited as it has exclusively focused on a narrow set of statements (Brewin et al., 2020).In the current studies, we have shown that high agreement ratings exist when a wide variety of statements are used assessing various aspects of repressed memory and dissociative amnesia.More importantly, we have revealed that beliefs in repressed memory and dissociative amnesia are not stable and can be changed.This is encouraging because it might pave the way for other effective methods to debias people's beliefs in repressed memory and dissociative amnesia.Our work further suggests that when expert witnesses are involved in cases on repressed memory, they could educate the court on the controversial aspects of repressed memory and dissociative amnesia.

FUNDING STATEMENT
The studies are funded by grants awarded to the first author: C1: C14/19/013 and FWO: G0D3621N.

APPENDIX A
Specific themes that were taught to students in the different studies.
Participants were students from the Faculty of Law and Criminology, KU Leuven following a Master's degree course on Legal Psychology 6 .The course was taught by the first author.In this study, 555 were registered and 108 students (mean age = 21.32,SD = .96,range: 20-25; 85 female) completed both surveys at Time 1 and 2. The attrition rate might be caused because of the similar reason as in Study 1 (i.e., coronavirus pandemic resulting in a main focus on online learning).
(e.g., "Therapy can help people with dissociative amnesia to recover their buried memories"), follow up questions (e.g., Therapists should look for any type of psychological stressor in their patients' life) were asked to examine more closely people's ideas behind the topic of repressed memory and dissociative amnesia.Furthermore, in Study 3, we also provided participants with a case vignette in which a person recovers a memory during therapy which she did not have before therapy.We inserted this case vignette to examine whether people's beliefs might also transfer to when people are confronted with a possible case of recovered memories.The use of case vignettes has been applied before in research on beliefs in repressed memory.For example,Houben et al. (2021) provided clinicians using Eye Movement Desensitization and Reprocessing (EMDR) therapy with a case vignette in which a patient recovered a memory during the course of treatment which the patient did not have before therapy.Houben et al. found that 75% (n = 9) of the EMDR practitioners indicated that it was (very) likely that the recovered memory referred to an authentic experience.What we do not know is whether people become more critical to label therapy-induced recovered memories as true memories when they receive education on the science of memory.
cant effects.Overall, our results replicated what we detected in Studies 1 and 2. However, what Study 3 added were the following two findings.First, fewer participants received the follow-up questions at Time 2 than at Time 1 showing that agreement ratings overall dropped when survey was completed the second time.Second, participants became critical toward the content of the case vignette by showing lower agreement ratings at Time 2 than at Time 1.
Such education could potentially lead to a more critical attitude among legal professionals concerning these concepts.Eventually, such corrective methods might help in lowering the chance of false accusations and wrongful convictions based on repressed memory and dissociative memory loss.
-related amnesia, dissociative amnesia, and repressed memory 8. False confession and interrogation 9. Children's disclosure of trauma and child interviewing 10.Psychopathy 11.Expert witness work and cognitive biases 1250 OTGAAR ET AL.
Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acp.4005by University Of Portsmouth, Wiley Online Library on [20/03/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Means and standard deviations of belief ratings at Time 1 and 2 (Study 1) Thus, based on Studies 1 and 2, controversial beliefs on repressed memory and dissociative amnesia can be changed.In Study 3, we wanted to take this research one step further.Although we once more aimed to replicate the findings from the previous studies, we wereT A B L E 1 T A B L E 2 Paired samples t-test statistics of belief ratings at Time 1 and 2 (Study 1) first author.The attrition rate could have occurred because of the similar reason as in Studies 1 and 2 (i.e., corona pandemic resulting in a main focus on online learning).Of the 279 students, 160 completed the first survey.Their mean age (M age = 19.01,SD = 2.25) did not statistically differ from students completed both surveys (t(216) = .22,p = .82,Cohen's d = .04).
Means and standard deviations of belief ratings at Time 1 and 2 (Study 2) words; see https://osf.io/swn28/) about a young woman undergoing therapy where she recovered a memory concerning sexual abuse that she did not have before therapy.After reading the case vignette, participants were asked T A B L E 3 Paired samples t-test statistics of belief ratings at Time 1 and 2 (Study 2) Means and standard deviations of belief ratings at Time 1 and 2 (Study 3) Paired samples t-test statistics of belief ratings at Time 1 and 2 (Study 3) T A B L E 4Note: Statements in italics refer to the follow-up questions, *first n refers to the number of subjects answering this question at Time 1, the second n refers to the number of participants answering it at Time 2, #these last statements were new statements on memory or emotion malleability.T A B L E 6Note: Italic indicatices follow-up statements.
10.1 | Limitations and future researchAlthough the current results are promising, there are some imperative empirical avenues ahead.One issue is whether a control group would be necessary in future research.It will be difficult to randomly assign some students to a control group university course, and it is anyway experiment, implying that the current results provide a good estimate of what happens in the population.Furthermore, follow up studies (Studies 2 and 3) used larger sample sizes and replicated the effects detected in Study 1.