Collaborative recall reduces the effect of a misleading post event narrative.
Leading questions (e.g. Loftus, 1975; Loftus & Palmer, 1974) and narrative accounts (e.g. Loftus, Burns & Miller, 1978) can change episodic memories and thereby affect the responses witnesses give immediately and during subsequent sessions. These effects have been replicated numerous times and many factors can contribute to them. Gudjonsson has explored suggestibility differences based on numerous individual differences and the techniques of interviewers. For example, suggestibility is negatively correlated with intelligence and memory capacity, and positively correlated with field dependence (Singh & Gudjonsson, 1992). Witnesses suffering from alcohol withdrawal (Gudjonsson, Hannesdottir, Petursson & Bjornsson, 2002) or sleep deprivation (Blagrove, 1996) are more likely to accept suggestions. Interviewers who are firm rather than friendly are more likely to elicit the effect in witnesses (Baxter, Boon & Marley, 2006). Furthermore, repeated questioning about the event can lead to greater misinformation effects (Roediger, Jacoby, & McDermott, 1996). Though younger children are more susceptible to misinformation than older children (Ceci & Bruck, 1993) and older adults show stronger effects than younger adults (Karpel, Hoyer, & Toglia, 1993), people of all ages are affected by post-event misinformation (Loftus, 2005).
Co-witnesses of a crime or traffic accident may also be sources of misinformation when they spontaneously discuss what they have seen. Gabbert, Memon, Allan, and Wright (2004) found that misinformation effects are stronger with a face to face co-witness discussion than with a post-event narrative. Likewise, Paterson and Kemp (2006) found misleading post event information presented through a co-witness discussion produced misinformation errors more consistently than did leading questions, a newspaper report on the witnessed event, or a video showing a discussion between two other co-witnesses.
In a procedure typical of those used in the co-witness literature, Gabbert, Memon, and Allan (2003) had pairs of participants view videos of the same incident, but from different perspectives. One witness was able to see the girl in the video steal a small amount of money, but the action was not visible from the other witness's perspective. Participants were asked to discuss the witnessed event before answering questions independently. Over a third of the participants who had discussed the event with another witness included in their statements information that had been learned through the discussion. Furthermore, 60% of these participants indicated that the girl was guilty even though they had not actually seen her take the money. In a similar study, 60% of children reported a detail that could not have been observed in the version of the video that they viewed (Candel, Memon, & Al-Harazi, 2007).
Social factors of the co-witness interaction can affect misinformation effects. Information from post event discussions was more likely to be incorporated into reports of the event if the co-witness was a friend or romantic partner (Hope, Ost, Gabbert, Healey, & Lenton, 2008) and misinformation effects can be greater when misinformation is presented in one-on-one discussion than in a group discussion (Dalton & Daneman, 2006).
Co-witnesses might also participate in collaborative recall. In studies mentioned previously, participants discussed the witnessed event but then answered questions or freely recalled the event independently. In other studies, participants have been questioned simultaneously. Performance of collaborators is often compared to that of individuals or of nominal groups, which are pairs of participants who recall independently but have their non-redundant recollections pooled. In general, collaborative recall tends to reduce performance compared to nominal groups, but still be better than individual recall (e.g. Basden, Basden, Bryner & Thomas, 1997; Weldon & Bellinger, 1997). However, collaborative recall can reduce the frequency of false memories as well. Ross, Spencer, Blatz and Restorick (2008) found that collaborative recall reduced both self-generated and other-generated false memories. Takahashi (2007) found collaborative recall reduced false memories in the Deese-Roediger-McDermott (DRM) task (Roediger & McDermott, 1995), and Yaron-Antar and Nachson (2006) found fewer correct and incorrect details were reported in collaborative recall than nominal groups and individual recall.
The literature on group brainstorming shows similar patterns. Group members are less productive than nominal groups but better than individuals due to reduced motivation and increased self-attention (see Mullen, Johnson & Salas, 1991 for meta-analysis). When working in a group, task responsibility can be diffused, so individuals may feel less motivated to contribute. Participants may also be more concerned with meeting social expectations and therefore may withhold risky ideas. When engaging in collaborative recall, witnesses might also withhold details that they are less confident about to avoid the embarrassment of mentioning something other witnesses might scoff at.
In summary, the many sources of potential misinformation, different ways of measuring eyewitness accuracy, and different situational and social factors make it difficult to predict whether a witness will give accurate, credible testimony. Leading questions alone can affect testimony (e.g. Loftus, 1975; Loftus & Palmer, 1974), but co-witness interactions appear to be more influential (Paterson & Kemp, 2006). Co-witness interactions prior to questioning can increase the amount of details reported, though witnesses report details they did not observe themselves (e.g. Paterson & Kemp). Structured collaborative recall of co-witnesses can reduce self-generated false memories relative to individual recall and nominal groups (Ross et al., 2008). Furthermore, collaborative recall has been found to increase subsequent individual recall for word lists (Basden, Basden, & Henry, 2000) and emotional events (Yaron-Antar & Nachson, 2006). However, whether collaborative recall can diminish the effects of narrative accounts remains to be seen.
In the present study, participants viewed a brief video of a car accident. After completing a filler task, participants were given a narrative account of the accident, which was either accurate or contained misinformation. Participants then answered questions individually or with a partner. We expected the participants who received the accurate narrative to have greater recall accuracy than those that received the misinformation (e.g. Loftus, Burns & Miller, 1978). We also expected accuracy to be greater for collaborative groups than for individuals (e.g. Basden, Basden, Bryner, & Thomas, 1997; Weldon & Bellinger, 1997). Finally, we expected collaboration to reduce the misinformation effect because Ross et al. (2008) found a similar effect on self-generated misinformation.
One hundred seventy-five students ages 18 to 50 (M = 20.33, SD = 1.81) volunteered for this study. All participants were recruited from undergraduate classes at St. Vincent College; a small liberal arts, Catholic school in PA. Participants were randomly assigned to both a description and a group condition. There were 109 women and 66 men from an array of classes at the college. Almost all of the students were offered extra credit for their participation in this study.
Materials and Procedure
After granting their informed consent, participants watched a 15 sec video clip depicting a car accident (Moo-seung & Yon-hawa, 2006). The video clip showed a woman talking to herself in her rearview mirror, not paying attention to the road. She rear-ended a stationary vehicle driven by a man at a stop light. Typically 15-20 students viewed the clip in a standard 42-seat classroom. The room was filled to near capacity for only 2 of 10 sessions. The video was projected from a computer onto a wall-mounted screen located at the front of the classroom. After viewing the video clip, participants worked independently on a word search (Livewire, 2007) as a filler activity for 2 min. Without the participants' knowledge, we randomly assigned them to complete the recall test independently or to collaborate with another participant. Participants were separated by condition and taken to different rooms for the remainder of the session. Once separated, the participants in the collaboration condition were told that they would be working with a partner and partners were seated together. Every participant completed a demographic questionnaire and answered questions about their previous familiarity with the clip. Participants in the collaboration condition were also asked questions about their familiarity with their partner. Each participant in the individual condition and each pair in the collaborative condition received one packet. The packet contained one narrative description of the video clip (see appendix A), which was either accurate or inaccurate. The inaccurate description described the car accident that was viewed in the video clip, but contained some details that were changed. After reading the narrative participants were asked to answer questions about the video based upon their memory of the event. Participants in the collaboration condition were told to discuss the questions about the video clip and answer collaboratively. The questionnaire required participants to give specific details from the video clip. It contained 12 questions, 7 of which addressed critical details that were changed in the inaccurate description (see appendix B). These critical questions are considered to be crucial in identifying the cause of and responsibility for the accident. After returning the packet to the experimenters, the participants received a written debriefing and were dismissed.
A 2(description: inaccurate or accurate) x 2(recall type: individual or collaborative) univariate independent subjects analysis of variance (ANOVA) was performed on the proportion of total questions correct. Data analysis revealed significant main effects of description and recall type and also a significant interaction between the two variables. Participants who received accurate descriptions performed better (M = 0.86, SD = 0.11) than participants who received inaccurate descriptions (M = 0.68, SD = 0.19), F(3, 112) = 19.73, p < .001, [[eta].sup.2] = .97. Collaborative pairs (M = 0.82, SD = 0.14) performed better than individuals (M = 0.71, SD = 0.20), F(1, 112) = 15.16, p < .001, [[eta].sup.2] = .12.
As shown in Figure 1, there was also a significant interaction demonstrating that the misinformation effect was greater for individuals than for collaborative pairs, F(2, 112) = 22.25, p < .001, [[eta].sup.2] = 0.28.
A 2(description: inaccurate or accurate) x 2(recall type: individual or collaborative) univariate independent subjects ANOVA was also performed on the proportion of critical questions correct, those questions that were discussed in the description. Like the previous analysis, this ANOVA revealed significant main effects of description and recall type. Participants who received accurate descriptions (M = 0.91, SD = 0.14) correctly answered more of the questions addressed in the description than participants who received the inaccurate description (M = 0.65, SD = 0.29), F(1, 112) = 42.85, p < .001, [[eta].sup.2] = .28. Collaborative pairs (M = 0.85, SD = 0.21) answered more of the critical questions correctly than did individuals (M = 0.70, SD = 0.29), F(1, 112) = 14.24, p < .001, [[eta].sup.2] = .11. The interaction between description and recall type for the critical questions failed to reach significance, F(1, 112) = 3.05, p = .084. (See Figure 2).
A Mann-Whitney Test was performed to determine whether individuals or collaborative pairs accepted at least one piece of inaccurate information more often. Of those presented with the inaccurate description, 25 of 30 individuals, and 20 of 30 collaborative pairs accepted at least one piece of misinformation. This difference in misinformation acceptance between collaborative pairs and individuals did not differ, U = 375, p = 0.14.
[FIGURE 1 OMITTED]
Our results supported several of our hypotheses. As predicted, those participants who read accurate descriptions had higher recall accuracy for all of the questions asked and for the subset of critical questions. Collaborative pairs also had more accurate recall than individuals. However, collaborative recall did not reduce the effects of the misleading narrative nor did collaborative groups have better recall than individuals for the critical questions.
[FIGURE 2 OMITTED]
That we found the misinformation effect, greater accuracy for those who read the accurate description, is not surprising. When answering questions, participants may have relied on their memory for the video or for the narrative. In some cases the participants may have misattributed the source of the information such that they came to believe they had seen a particular detail in the video (Zaragoza & Lane, 1994), or they may have accepted information from the narrative (Loftus, 1991). The performance of participants who read the accurate description would not have been hindered by source misattribution or information acceptance, but the performance of participants who read the inaccurate description could have been negatively affected by either of these effects. Moreover, those that read accurate descriptions may have had their initial memory reinforced by the narrative.
As expected, collaborative pairs had higher overall recall accuracy and higher recall accuracy for the critical questions than did participants who worked individually. The pairs had two sets of memories to draw upon, so when one participant could not provide an answer the other one often could. When collaborating, pairs may elect to defer to the person deemed to have better memory for the information sought by the question (Weldon & Bellinger, 1997; Yaron-Antar & Nachson, 2006) or is most confident in their memory (Wright, Self, & Justice, 2000).
The focus of the present study was on whether collaborative recall could weaken the misinformation effect. Our results showed a significant interaction between description and group size on overall recall accuracy. Collaborative pairs showed a smaller misinformation effect than individuals. It appears that collaborative recall may allow witnesses to dismiss misinformation generated by an inaccurate narrative. Ross et al. (2008) found that when misinformation was brought up during discussion and the partner expressed reservation about the item, the incorrect information was usually excluded from the pair's final answer. However, it should be noted that the misinformation effect was smaller for collaborative pairs than for individuals only when all questions were evaluated. When just the critical questions were examined the interaction failed to reach significance. This result could have been due to the smaller range of scores associated with fewer questions.
Of the participants who received inaccurate descriptions, two-thirds of both individuals and collaborative pairs accepted at least one piece of misinformation. Even though participants may have accepted only one piece of misinformation, that detail could contain crucial information of the witnessed event. For instance, one of our critical questions asked what, if any, damage was done to the cars. Some participants who answered the question incorrectly still provided detailed, but inaccurate answers. Testimony as to whether damage was actually caused by the accident could be the difference between negligence on the part of the accused and insurance fraud on the part of the accuser. Also, inaccurate testimony about the physical appearance of a suspect can lead to the prosecution of an innocent person.
The proportion of witnesses who reported misinformation is a matter of concern given the brief delays between the witnessed event, narrative, and collaborative recall test. Discrepancy detection is easiest when time between the event and the misinformation is brief because the memory for the original event is still strong (Loftus, 2005). Likewise, if the delay between the narrative and testing is brief then the memory for the narrative is still strong, thereby increasing the chance of discrepancy detection. In the present study both delays were brief. Participants completed only a 2 min. filler task between the video and the narrative, and they completed the questions immediately after reading the narrative and turning the page. With such brief delays, it is surprising that so many participants demonstrated the misinformation effect.
After witnessing a crime or accident there may be opportunities for witnesses to interact and share information. Late arriving bystanders or members of the media may ask witnesses to recall the event before law enforcement or legal representatives have the opportunity to interview them. Our study simulates the case where two witnesses might first hear the event retold by a third witness before being asked to provide their own, collaborative account. Our results suggest that collaborative recall may decrease the effects of misinformation provided by an inaccurate narrative, which could lead to a more accurate composite account of the event. However, others have shown that subsequent testimonials may contain information that was not initially witnessed (e.g. Candel et al., 2007; Gabbert et al., 2003). So although the pair's responses may be more accurate because of the collaboration, their later, individual responses may contain more fabrications.
This research can be extended in several ways to further build upon the existing literature. Using free recall in addition to the specific questions would allow for the performance of individuals and collaborative pairs to be compared with nominal groups. Yet to be determined is whether the effect of collaborative recall on the misleading narrative effect will persist under later, individual recall. By adding both free recall and a second recall session more comparisons to the results of Basden et al. (2000), Weldon and Bellinger (1997), and Yaron-Antar and Nachson (2006) could be made.
Descriptive Accounts of the Accident Accurate
At approximately 12:30 in the afternoon on Friday, a young woman named Katherine Ching hit the backend of a silver sedan stopped at an intersection. The sedan was driven by Ben Yip. Witnesses say they saw Ching looking at herself in the rearview mirror immediately before the accident occurred. There was no damage to Yip's car but, the front bumper of Ching's red compact car was damaged. There are no other details to report at this time.
At approximately 12:30 in the afternoon on Friday, a young woman named Katherine Ching hit the back end of a black sedan in motion. The sedan was driven by Ben Yip. Witnesses say they saw Ching talking on her cell phone immediately before the accident occurred. Yip's sedan and Ching's blue compact car were not damaged. There are no other details to report at this time.
Post Event Questions
1. * What color was the car responsible for the accident?
2. * Were any cars damaged? If yes, describe the car and the damage done to the car.
3. *What was the driver of the car responsible for the accident doing before the accident occurred?
4. *Was the car that was hit in motion or stationary?
5. What color was the license plate of the car that was hit?
6. Was the woman's hair long or short?
7. How many people were in the car that caused the accident?
8. * Were there any pedestrians hit?
9. What color was the woman's steering wheel?
10. Was the man wearing a watch?
* indicates a critical question. Item number 2 was divided into three items for scoring: (a) were any cars damaged, (b) describe the car that was damaged, (c) describe the damage done to the car.
Basden, B., Basden, D., Bryner, S., & Thomas, R. (1997). A comparison of group and individual remembering: Does collaboration disrupt retrieval strategies? Journal of Experimental Psychology: Learning, Memory, and Cognition, 23(5), 1176-1191.
Basden, B., Basden, D., & Henry, S. (2000). Costs and benefits of collaborative remembering. Applied Cognitive Psychology, 14(6), 497-507.
Baxter, J., Boon, J., & Marley, C. (2006). Interrogative pressure and responses to minimally leading questions. Personality and Individual Differences, 40(1), 87-98.
Blagrove, M. (1996). Effects of length of sleep deprivation on interrogative suggestibility. Journal of Experimental Psychology: Applied, 2(1), 48-59.
Candel, I., Memon, A., & Al-Harazi, F. (2007). Peer discussion affects children's memory reports. Applied Cognitive Psychology, 21(9), 1191-1199.
Ceci, S., & Bruck, M. (1993). Suggestibility of the child witness: A historical review and synthesis. Psychological Bulletin, 113(3), 403-439.
Dalton, A., & Daneman, M. (2006). Social suggestibility to central and peripheral misinformation. Memory, 14(4), 486-501.
Gabbert, F., Memon, A., & Allan, K. (2003). Memory conformity: Can eyewitnesses influence each other's memories for an event? Applied Cognitive Psychology, 17, 533-543.
Gabbert, F., Memon, A., Allan, K., & Wright, D. (2004). Say it to my face: Examining the effects of socially encountered misinformation. Legal and Criminological Psychology, 9(2), 215-227.
Gudjonsson, G., Hannesdottir, K., Petursson, H., & Bjornsson, G. (2002). The effects of alcohol withdrawal on mental state, interrogative suggestibility and compliance: An experimental study. Journal of Forensic Psychiatry, 13(1), 53-67.
Hope, L., Ost, J., Gabbert, F., Healey, S., & Lenton, E. (2008). 'With a little help from my friends...': The role of co-witness relationship in susceptibility to misinformation. Acta Psychologica, 127(2), 476-484.
Karpel, M., Hoyer, W., & Toglia, M. (2001, March). Accuracy and qualities of real and suggested memories: Nonspecific age differences. Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 56(2), 103-110.
Livewire Puzzles. (2007). Free Printable Wordsearches (n.d.) Animals 1. Retrieved October 20, 2007 from, http://www.puzzles.ca/wordsearch.html.
Loftus, E.F. (1975). Leading questions and the eyewitness report. Cognitive Psychology, 7, 560-572.
Loftus, E.F. (1991) Made in memory: Distortions of recollection after misleading information. In G. Bower (Ed.) Psychology of Learning and Motivation, 27, 187-215. NY: Academic Press.
Loftus, E.F. (2005). Planting misinformation in the human mind: A 30-year investigation of the malleability of memory. Learning & Memory, 12, 361-366.
Loftus, E.F., Burns, H.J., & Miller, D. G. (1978). Semantic integration of verbal information into a visual memory. Journal of Experimental Psychology: Human Learning and Memory, 4(1), 19-31.
Loftus, E.F. & Palmer, J.C. (1974). Reconstruction of automobile destruction: An example of the interaction between language and memory. Journal of Verbal Learning and Verbal Behavior, 13, 585-589.
Moo-seung, P. (Producer), Yong-hawa, K. (Writer/Director). (2006). 200 pound beauty [Motion picture]. Korea: KM Culture Co.
Mullen, B., Johnson, C., & Salas, E. (1991). Productivity loss in brainstorming groups: A meta-analytic integration. Basic and Applied Social Psychology, 12(1), 3-23.
Paterson, H.M., & Kemp, R.I. (2006). Comparing methods of encountering post-event information: The power of co-witness suggestion. Applied Cognitive Psychology, 20, 1083-1099.
Roediger, H., Jacoby, J., & McDermott, K. (1996). Misinformation effects in recall: Creating false memories through repeated retrieval. Journal of Memory and Language, 35(2), 300-318.
Roediger, H., & McDermott, K. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(4), 803-814.
Ross, M., Spencer, S., Blatz, C., & Restorick, E. (2008). Collaboration reduces the frequency of false memories in older and younger adults. Psychology and Aging, 23(1), 85-92.
Singh, K., & Gudjonsson, G. (1992). Interrogative suggestibility among adolescent boys and its relationship with intelligence, memory, and cognitive set. Journal of Adolescence, 15(2), 155-161.
Takahashi, M. (2007). Does collaborative remembering reduce false memories? The British Psychological Society, 98, 1-13.
Weldon, M., & Bellinger, K. (1997). Collective memory: Collaborative and individual processes in remembering. Journal of Experimental Psychology: Learning, Memory, and Cognition, 23(5), 1160-1175.
Wells, G., Malpass, R., Lindsay, R., Fisher, R., Turtle, J., & Fulero, S. (2000). From the lab to the police station: A successful application of eyewitness research. American Psychologist, 55(6), 581-598.
Wright, D.B., Self, G., & Justice, C. (2000). Memory conformity: Exploring misinformation effects when presented by another person. British Journal of Psychology, 91, 189-202.
Yaron-Antar, A., & Nachson, I. (2006, January). Collaborative remembering of emotional events: The case of Rabin's assassination. Memory, 14(1), 46-56.
Zaragoza, M., & Lane, S. (1994). Source misattributions and the suggestibility of eyewitness memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20(4), 934-945.
Tara E. Karns, Sara J. Irvin, Samantha L. Suranic, and Mark G. Rivardo
Saint Vincent College
Author info: Correspondence should be sent to: Dr. Mark Rivardo, Psychology Dept., St. Vincent College, 300 Fraser Purchase Road, Latrobe, PA 15650.
|Printer friendly Cite/link Email Feedback|
|Author:||Karns, Tara E.; Irvin, Sara J.; Suranic, Samantha L.; Rivardo, Mark G.|
|Publication:||North American Journal of Psychology|
|Date:||Mar 1, 2009|
|Previous Article:||Behavioral problems of adolescents from returned Portuguese immigrant families.|
|Next Article:||An interview with Judith Beck about cognitive therapy.|