URJHS Volume 7

Dissociation and the Creation of False Memories

Jenny Carstens and Reid Webster, Ph.D.*

Thompson Rivers University
Kamloops, B. C., Canada

Abstract

This study was intended to further the understanding of the process of false memory formation by combining a visual method with dissociation as a personality attribute. Thompson Rivers University students were administered the Dissociative Experiences Scale (DES) to measure their experiences of dissociation. Pictures of stereotypical scenes were presented to them, after which they participated in a recollection task. The students’ recollection was differentiated between “knowing” and “remembering” to determine the type of their memory. It was hypothesized that students’ DES scores would be positively related to the number of lures they recognized, more specifically, as “remembered,” Results indicated some support for the hypothesis. Thus, individuals with higher levels of dissociative experiences were somewhat more likely to create false memories.

Dissociation and the Creation of False Memories

Several studies have been conducted to show that false memories can be created by researchers (Baxter, 1990; Ceci & Bruck, 1993; Loftus, 1979). Typically these studies consisted of a story or a picture that was presented to the participants, with deceiving suggestions of details and a later period of recall or recollection. The details added by the researcher were often being recalled by the subjects as actually having been part of the events. Also, leading questions often resulted in the participants’ claim of having observed specific details that were not part of the original event (Marin, Holmes, Guth & Kovac, 1979). The following is an example of suggestion in a question that was used by Marin et al. (1979): “Was the package the man carried small?” in contrast to the non-suggestive question, “Was the man carrying a package?”

Actions that were witnessed by the participants could be altered by suggestion as well. In a study by Schooler, Gerhard and Loftus (1986), participants viewed a slideshow depicting a traffic accident. In one group, the slides included a yield sign and in the other group the sign was absent. The presence of a yield sign was suggested in a questionnaire. Many participants in both groups reported seeing the sign. According to Lindsay and Read (1994), suggestibility has been shown to have an effect on participants of all age groups, although the effect is greater with younger participants.

Past research has shown that individual differences had an influence on the likelihood of the creation of false memories. Subjects in various studies have undergone personality tests and have been tested on their tendency to falsify memory. Personality tests included, among others, the Beck Depression Inventory (BDI), the Questionnaire on Mental Imagery (QMI), the Verbalizer-Visualizer Questionnaire (VVQ), the Dissociative Experiences Scale (DES), the Creative Imagination Scale (CIS), and the Subjective Memory Questionnaire (SMQ) (Roberts, 2002; Winograd, Peluso & Glover, 1998). According to Roberts (2002), subjects with more vivid imagery and those who are stronger visualizers had higher rates of false memory creation. Wilkinson and Hyman (1998) also found that individuals who used more imagery were more likely to create false memories. They more often falsely recognized critical lures of the Deese-Roediger-McDermott (DRM) paradigm than individuals that used less imagery. This paradigm consists of lists of words that are all related to a non-presented word (the “critical lure”). Examples given by Winograd et al. (1998) were the words bed, rest, awake, tired, and dream (all associated with the critical lure “sleep”), which would then often be falsely recalled. Thus, the study showed that more imaginative individuals were more likely to create false memories.

Therapists that engage in guided imagery often ask their patients to imagine childhood events. This technique is thought to help trigger the patient’s repressed memories but in fact might lead to the creation of false memories such as childhood abuse. It has been shown by a number of authors that memories are subject to suggestion. A study conducted by Garry, Manning, Loftus, and Sherman (1996) demonstrated that guided imagery is often linked to the creation of false memories. Subjects are then more likely to believe the event occurred in real life (as cited in Shaffer & Oakley, 2005).

Disorders that involve experiences of dissociation are considered one of the most controversial at present. This skepticism is even reflected in the DSM-IV; specifically the American Psychiatric Association (2000) points out the possibility of inaccurate reports of childhood abuse, which is thought to be a contributing factor to the disorder. The reason for this is the likelihood of distorted memory as well as the patient’s high tendency to be suggestive. The proneness of dissociation is thought to be the degree of suggestibility (American Psychiatric Association, 2000).

Hyman and Billings (1998) investigated personality characteristics that could predict a subject’s willingness to accept suggested events as their own childhood memories. They found that the higher the scores on the DES or the CIS, the more the subjects reported false memories as their own. The former is associated with dissociation and the latter with hypnotizability. Their findings indicated that the DES was a predictor of false autobiographical memories.

A study conducted by Winograd et al. (1998) examined the relationship between false memory creation and several personality traits using the DRM paradigm. They found that a higher score on the DES was related to a higher tendency to report the critical lure and also a lower tendency to report “knowing” but “remembering” that it had been presented. The difference between these two is that “remembering” “refers to those items for which they have a vivid memory, a subjective feeling of having seen the item during the study episode, and a conscious recollection of it occurring” (Rajaram, 1993, p. 90), whereas “knowing” occurs in the absence of conscious recollective experience (Rajaram, 1993). An example given by Rajaram (1993) of “remembering” was the ability of describing a visit to a national park with the recollection of details and mentally reliving the event. In contrast, an example of “knowing” was described as meeting someone on the street that one had met before at a party. Although one might not consciously recollect meeting the person, the recognition of him or her was considered a “knowing” response. The concept of remembering and knowing is important for this study because it demonstrates the difference of actually creating a new memory with new details (remembering) in contrast to simply knowing it has been presented without the memory of its presentation.

Platt, Lacey, Iobst, and Finkelman (1998) also conducted a study to test the relationship between the DES and the creation of false memories using the DRM paradigm and autobiographical memories. But unlike Hyman and Billings (1998) and Winograd et al. (1998), they failed to find a significant relationship between the DES and any of the two forms of false memories (DRM paradigm or autobiographical). Platt et al. noted that their experiment differed in some aspects from the ones conducted by the others and that this might be the reason for the lack of a relationship between dissociation as a personality trait and false memory creation. In contrast to Hyman and Billings, the study of Platt et al. involved participants’ memories about actual events and not memories about events that were created by the researchers. Whereas Hyman and Billings used suggestion to alter memories, Platt et al. studied the alteration of memories that occurred over time. They argued that participants with high DES scores were more susceptible to suggestion, as the study of Hyman and Billings suggested, “while not showing reduced accuracy in everyday memory” (p. 86). The different results of Platt et al. and Winograd et al. could have occurred because of different materials used by the researchers. Roediger and McDermott (1995) included two experiments in their study: the first one consisted of 6 lists with 12 words each, and the second one consisted of 16 lists with 15 words each. With some variations, subjects were asked to recall the items on the list after they were presented. They were also asked to engage in a recollection task after the last list of words had been presented. The results showed that subjects recalled critical lures significantly more often than unrelated words in both experiments. Also, the second experiment was associated with higher rates of false recall and false recognition of the critical lures than the first experiment. In the study of Winograd et al., Roediger and McDermott’s Experiment 2 was used, whereas Platt et al. used Experiment 1. Thus, the difference in rates of false recall between the two experiments of Roediger and McDermott might have been the reason for the different results found between Platt et al. and Winograd et al.

Miller and Gazzaniga (1998) used a visual test as a measure of false memory creation. They presented the subjects with 12 pictures of stereotypical scenes in which objects that were strongly related to the scene (“critical lures”) were removed. At a later recognition period, “critical lures,” non-related objects, and depicted objects were presented, and the subjects were asked to indicate which objects had been shown at the prior time. When they identified an object as having been previously presented, they were further asked to distinguish between “remember” and “know” responses. The results showed that the lures were more often recognized as having been in the picture than non-related words, with a high proportion of “remember” responses. The visual test was shown to be as successful as the DRM paradigm in false memory creation, although it might not have involved identical memory processes. They noted that further research was needed to examine possible differences between the memories created by the two methods.

Because the DES has been part of experiments involved in autobiographical memories and with the DRM as a word list, the research was extended by combining the DES with a visual measure of the creation of false memories. Therefore, in this research the identical procedure as in Miller and Gazzaniga (1998) was used.

It was hypothesized that the higher a person’s score on the Dissociative Experiences Scale (DES), the higher their recollection of critical lures would be, which would be false. Because of the important component of suggestibility in concepts, dissociation and false memory creation, a strong positive correlation was predicted. Miller and Gazzaniga’s (1998) findings of subject’s higher tendency to “remember” critical lures also led to the hypothesis that the higher the subject’s DES score would be, the more often they would recognize lures as “remembered.”

Method

Participants

One hundred and fourteen first and second year psychology students attending the Thompson Rivers University, Kamloops, in the winter semester of 2007 participated in this study. From the 109 participants that specified their gender, 33% were male (N = 36) and 67% female (N = 73). Only 96 participants indicated their age, which ranged from 18 to 46 years (M = 20.48, SD = 3.94). They were informed of the nature of the study, that their participation was voluntary and unpaid, and that they could discontinue their participation at any time they wished.

Materials

All the material was used with permission. The Dissociative Experiences Scale (DES) is a 28-item self-report measure of dissociation that has been used in numerous published studies (Carlson & Armstrong, 1994). Its scale ranges from 0 to 100 in intervals of ten and the participants are asked to circle the number that most appropriately describes their experiences. The DES takes approximately ten minutes to complete. The score is calculated by averaging the answers of the 28 questions. The higher the score, the more a subject experiences dissociation. A cut-off score of 30 or above is suggested to be indicative of a dissociative disorder (Carlson & Putnam, 1993). Subjects were asked not to report experiences that occurred under the influence of drugs or alcohol. The DES has been shown to be a reliable and valid measurement for dissociation. It is internally consistent and shows high test-retest reliability. Frischolz et al. (1990) reported a test-retest reliability of r = .96 with a test-retest interval of four weeks. Pitblado and Sanders (1991) reported a test-retest reliability of r = .79 with an interval of six to eight weeks. They also reported an internal reliability of r = .93 (as cited in Carlson & Putnam, 1993). According to Carlson and Putnam (1993) the DES has high construct validity. People who are expected to score high have high scores (for example people with Dissociative Identity Disorder) and those that are expected to score low actually do so. High scores have been shown in conditions of dissociative disorders (Carlson & Armstrong, 1994).

As mentioned earlier, the identical procedure as in Miller and Gazzaniga (1998) was used, which included eighteen images from The Saturday Evening Post that showed stereotypical scenes from the time period of 1909 to 1967. From each picture four objects that were strongly related with the scene were identified. Two groups of pictures were created: the first group was created by removing two of these four related objects (set A) and the second group by removing the other two (set B). The removal of these objects was accomplished with Adobe Photoshop. For example, a classroom scene had the chalkboard and apples on the teacher’s desk removed in set A, and chalkboard erasers and the teacher’s chair were removed in set B (see Figure 1). To counterbalance the order of pictures, three groups were created (1, 2, and 3). Each group consisted of a varying number of pictures from set A and set B, resulting in 12 pictures in total that were shown to the subjects and six pictures that were not shown to them. The following recognition test included the identified and presented objects (24 “old” items), the identified but missing objects (24 “lure” items), and the four identified objects of the six pictures that were not presented to the subjects (24 “new” items), resulting in 72 items in total.

Figure 1. Example of one picture in set A and set B.

   

Procedure

After informing the subjects of their rights as participants and of the nature of the study, they were presented with one of the three groups of pictures (1, 2, or 3). Their task was to remember as many things as possible without taking any notes. The pictures were presented on a computer screen for ten seconds each, with a five-second break during which they saw a countdown from five to one on their computer screen.

The participants were instructed to complete the DES after the pictures had been presented. This was followed by a recognition test during which the 72 items noted in the materials section were played on a tape to the participants. Prior to this, they were told that some of the items on the tape were contained in the pictures and some were not. They were asked to indicate if they recognized the items by writing down “old” on a sheet of paper if they did, or “new” if they did not. If they indicated the recognition of an item they were asked to further indicate if they “remembered” or “knew” the item had been presented previously. Instructions regarding the difference and judgement of remembering or knowing an item, as outlined in Rajaram (1993), were given to them before they began.

Results

Participant’s DES scores ranged from 2.86 to 50.36 (M = 17.08; SD = 9.04). The mean number of lures recognized was 13.80 (SD = 4.39), the mean number of lures “remembered” was 6.78 (SD = 3.82), and the mean number of lures “known” was 7.02 (SD = 3.47). In Figure 2, the mean number of responses (distinguished between “remembered” and “known”) within the different item categories can be seen. It shows that, overall, a greater number of “lures” than “new” items were recognized. Also, the number of “lures” recognized was only a little smaller than the number of presented items recognized, which showed that the visual manipulation served as a mean to create false memories.

Figure 2. Mean number of items recalled, dependent on the nature of the item.

Pearson r was used to examine the relationship between participant’s DES scores and the number of items they recognized. This correlational analysis indicated that participants with higher DES scores recognized significantly more lures as “remembered” (r (112) = .20; p = .03) (see Figure 3). There, a slight positive relationship is shown between the two; specifically, the higher the DES score, the more lures the participants recognized as “remembered.” No significant relationships were found between DES scores and the number of lures recognized in general (r (112) = .14; p = .15) or number of lures recognized as “known” (r (112) = -.05; p = .62). Similarly, no significant relationships were found between DES scores and number of “old” items recognized, either as “remembered” or as “known” (r’s (112) = -.03; -.02; -.01, n.s., respectively), or between DES scores and the number of “new” items recognized, either as “remembered” or as “known” (r’s (112) = .09; .13; .02, n.s., respectively).

Figure 3. Relationship between DES scores and the number of lures recognized as “remembered”.

To further analyze the data, a median split was conducted to divide the subjects into two groups. The mean of the high DES group was 23.50 (SD = 8.12), whereas the mean of the low DES group was 10.43 (SD = 3.33). Table 1 displays the DES scores, number of lures recognized, and the number of lures recognized as “remembered” between the high and low DES group. It shows that the high DES group recognized slightly more lures and lures as “remembered.” A t-test showed that the subjects in the high DES group recognized significantly more lures than the subjects in the low DES group (t (112) = 2.11; p= .04). Another significant difference was found between the two groups and the number of lures recognized as “remembered.” Subjects in the high DES group recognized a significantly greater number of lures as “remembered” as compared to the subjects in the low DES group (t (112) = 2.39; p = .02). Table 2 displays the values of the t-test analysis of the two groups. One can see that the difference in recognized items between the high and low DES group was only small, but nonetheless significant for lures and lures “remembered.” No significant difference in the number of lures recognized as “known” was found between the high and low DES groups

(t (112) = .05; p = .96). Likewise, the high and low DES groups did not differ significantly in the number of “old” items recognized, either as “remembered” or as “known” (t’s (112) = .34, .23, .10, n.s., respectively). There were also no significant differences between the high DES and low DES group in the number of “new” items recognized, either as “remembered” or as “known” (t’s (112) = .85, .10, .1.09, n.s., respectively).

Table 1

Differences in the DES mean scores, the number of lures recognized, as well as the number of lures recognized as “remembered” between the high and low DES groups

	High DES group	Low DES group
DES mean	23.50	10.43
Mean number of lures recognized	14.64	12.93
Mean number of lures “remember” recognized	7.60	5.93

Table 2

Results of the t-test analysis of the high and the low DES groups

	t - value	df	p - value	Mean difference
Lures	2.11	112	.04	1.71
Lures ”remember”	2.39	112	.02	1.67
Lures “know”	.05	112	.96	.03

Discussion

This study was intended to extend our understanding of individual differences in memory creation. As hypothesized, there was a significant positive relationship between the DES scores and the number of lures recognized as “remembered;” specifically, the higher the subject’s DES score, the more lures he or she recognized as “remembered.” However, no significant relationship was found between the DES scores and the number of falsely recognized lures altogether (with “remember” and “know” responses combined), which does not support the hypothesis. Therefore, the hypothesis has only been partly supported. The fact that even the non-significant results were in the predicted direction (that is, the relationship between DES scores and lures recognized was stronger than between DES scores and “old” items recognized) was encouraging. The relationship between DES scores and the number of “new” items recognized was also greater than the relationship between DES scores and the number of “old” items recognized. Because the recollection of new items indicated the creation of false memories, this showed a trend in the expected direction. One reason for these results might be the restricted range of the data. From the 114 participants only eleven had a score of thirty or above, which was the suggested cutoff score for individuals with dissociative experiences. Perhaps in a population that reported higher DES scores, the results of this study would have shown a greater correlation between DES scores and the creation of false memories.

The results of the t-test analysis supported the hypothesis. The number of lures recognized, specifically as “remembered,” was greater in the high DES group than in the low DES group.

Once the data were analyzed, they showed that the clinical significance might not be as great as suspected. The correlation between DES scores and lures recognized as “remembered” indicated only a weak relationship. The situation is similar for the significant results of the t-test, which showed that subjects in the high DES group recognized more lures, as well as lures “remembered” than the subjects in the low DES group. These differences between the high and the low DES group, however, were only small and may therefore not be meaningful.

In the literature reviewed, some studies found significant relationships between DES scores and the creation of memories (e.g. Hyman & Billings, 1998; Winograd et al., 1998) and others did not (e.g. Platt et al. 1998). In this study, even the non-significant results tended to show that dissociation might be related to the creation of false memories, whereas even the statistically significant results implied little clinical significance.

In this study the pictures were combined with an auditory recollection task, where the items were named on a tape that was played to the participants. One possibility to extend this study is to combine the visual means to create false memories with a visual recollection task, for example by showing the different items to the subjects, instead of naming them. In that case the study would involve the same sensory information, but it might show more conclusive results. A possible study could include recognition conditions of a visual method only, an audio method only, and a combination of the two, to see how this manipulation would influence the participants’ false recognition.

Although there are several studies that investigated the relationship of certain personality characteristics and the creation of false memories (e.g. Platt et al. 1998, Roberts, 2002; Winograd et al. 1998), it is still unknown what specifically contributes to the likelihood of these creations. Some studies suggest that individuals who are strong visualizers and those who have a more vivid imagery are more likely to create false memories (e.g., Hyman & Billings, 1998; Roberts, 2002). Winograd et al. (1998) suggested that the reason some individuals were more likely to falsely recall items may have been that they were less capable at “source monitoring,” meaning that these people had trouble distinguishing between internally and externally created events. This study showed that more research is needed to determine how exactly dissociation, other factors, and possibly an interplay of various numbers of attributes are related to the likelihood to create false memories. It would be beneficial to be able to identify individuals that are especially prone to these creations. For example it might be of value in the legal system, where witnesses sometimes determine the fate of other individuals on the basis of their memory recollection. It might also aid in the debate concerned with the accuracy of memories reported by individuals with dissociative disorders. The question is whether individuals that report a higher level of dissociative experiences have falsely created the memory of childhood abuse or that dissociation is a consequence of such abuse. The present results indicated that the memory of individuals who reported a high level of dissociative experiences might be less reliable and more prone to suggestion than the memory of individuals with less dissociative experiences.

References

American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (4 th ed.). Washington D.C.:

Baxter, J. S. (1990). The suggestibility of child witnesses: A review. Applied Cognitive Psychology, 4, 393-407.

Carlson, E. B., & Armstrong, J. (1994). The diagnosis and assessment of dissociative disorders. In S. J. Lynn & J. W. Rhue, Dissociation: Clinical and theoretical perspectives (pp. 159-174). New York, US: Guilford Press.

Carlson, E. B., & Putnam, F. W. (1993). An update on the dissociative experiences scale. Dissociation, 6, 16-25.

Ceci, S. J., & Bruck, M. B. (1993). Suggestibility of the child witness: A historical review and synthesis. Psychological Bulletin, 113, 403-439.

Hyman, I. E., & Billings, F. J. (1998). Individual differences and the creation of false childhood memories. Memory, 6, 1-20.

Lindsay, D. S., & Read, J. D. (1994). Psychotherapy and memories of childhood sexual abuse: A cognitive perspective. Applied Cognitive Psychology, 8, 281-338.

Loftus, E. F. (1979). The malleability of human memory. American Scientist, 67, 312-320.

Marin, B. V., Holmes, D. L., Guth, M., & Kovac, P. (1979). The potential of children as eyewitnesses. Law and Human Behaviour, 3, 295-306.

Miller, M. B., & Gazzaniga, M. S. (1998). Creating false memories for visual scenes. Neuropsychologia, 36, 513-520.

Platt, R. D., Lacey, S. C., Iobst, A. D., & Finkelman, D. (1998). Absorption, dissociation, and fantasy-proneness as predictors of memory distortion in autobiographical and laboratory-generated memories. Applied Cognitive Psychology, 12, 77-89.

Rajaram, S. (1993). Remembering and knowing: Two means of access to the personal past. Memory & Cognition, 21, 89-102.

Roberts, P. (2002). Vulnerability to false memory: The effects of stress, imagery, trait anxiety, and depression. Current Psychology, 21, 240-252.

Roediger, H. L., & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology, 21, 803-814.

Schooler, J. W., Gerhard D., & Loftus, E. F. (1986). Qualities of the unreal. Journal of Experimental Psychology, 12, 171-181.

Shaffer, M., & Oakley, J. S. (2005). Some epistemological concerns about dissociative identity disorder and diagnostic practices in psychology. Philosophical Psychology, 18, 1-29.

Winograd, E., Peluso, J. P., & Glover, T. A. (1998). Individual differences in susceptibility to memory illusions. Applied Cognitive Psychology, 12, 5-27.

Wilkinson, C., & Hyman, I. E. (1998). Individual differences related to two types of memory errors: Word lists may not generalize to autobiographical memory. Applied Cognitive Psychology, 12, 29-46.