Cognitive Interview for Eliciting Information from a Witness - Literature review Example

240540 Effectiveness of the Cognitive Interview as a Method of Eliciting Information from a Witness: A Review of Literature Geiselman and Fisher developed the basic technique of cognitive interview (CI) and newer dimensions are being added to it continuously. The CI is based on social dynamics, context building, and imagery. It uses the open-ended questions, unlike the close-ended questions of structured or standard interviewing (SI) seek information from a witness. The technique gained recognition when the US department of justice turned to CI after the large number of DNA exoneration cases. An attempt is made in the present article to find effectiveness of this technique with witnesses of different age groups. The article also explores whether the increased recall of event comes with increased errors and confabulations also and how misattribution and memory conformity may affect the validity of the process The witness of a crime event is not able to recall the incident, as there is nothing like a photograph in human mind. A cooperative witness also finds it difficult since there is limitations of memory and perception. These problems are aggravated when information in not sought in proper way (Taylor 161). She refers to Geiselman and Fisher generally for their well-researched technique of CI. The principle of latter's research is that the memory has many fragments and that the retrieval system should have similar elements. Brown and Geiselman (14) had further elaborated the theory of Geiselman and Fisher by adding recall of events in different orders and perspectives. The correct and more memory recall, as mentioned above, is facilitated by social rapport between interviewer and the witness. The witness is given a feeling that everything he/she says is important. Taylor (161-162) shows agreement with these pioneers in that after the rapport with the witness , the interviewer should ask open ended questions and let the witness elaborate crime related information. Periods of long silences be allowed as the witness's mind goes back to crime scene and bring out finer details. CI researchers caution interviewer to value non verbal evidences such as drawing etc.. Taylor (162) proceeds to techniques of memory and cognition after the social dynamics. She finds that both the CI pioneers have emphasized on context restoration. The witness simulates the context complete with environment, lighting, feelings, emotions, physiological condition and perception into the mind of witness through their open-ended approach. The context building reportedly yields as much recall as the whole CI It is to be noted here that the social angle is not only the initial phase of interrogation ,it remains there always and if circumstances arouse or distress a witness , he/she/ should be allowed time out till feels safe and is in control (Memon and Hingham par. 3). A discussion of the work of CI researcher is presented in the light of these elements of the technique. The methodology in the experiments to assess efficiency of CI uses students, children and other person by showing them a video clip or magic show and then using CI to find out the amount and accuracy of recall and other observations. One useful experiment on comparative study of two major techniques, CI and SI, of witness examination was certainly an appropriate dimension to look for. Hence Memon et al ( 1997 b:642-651) conducted their study with 64 adolescents of between the age of 16 and 19 years. Their findings revealed that, contrary to expectation, there was no significant difference between CI and SI condition for errors and confabulations which were (Fisher PLSD= 7.19, p=.07 and (p>.05) for CI and SI respectively. The accuracy rates were also same (84%) for both CI and SI. The similar results of SI and CI here are influenced by identical components of both the interviews including the rapport building and facilitating communication. The CI differed with SI in having unique components of the technique, the context rebuilding, imagery and 'tell everything'. Most of the experiments on CI have been conducted using college students as sample. Do the education or economic levels affect recall This thought led Stein and Memon, (598-599) change their methodology and draw a sample of 64 Brazilian women from low-socioeconomic group that had upto elementary school education. The women were shown a video. The recall was same as in case of college students. The CI and SI were comparable in accuracy rates viz. 76.5% and 74.8% respectively. However, the correct information and total recall was more in CI. Wells et al were worried that though CI techniques are immensely useful for recall of crime event these generally fail identification of actual perpetrator from the lineups or the person errors are very high (586-587). So they used a methodology to compare sequential lineup from the conventional simultaneous lineup. In the former, the perpetrator and fillers are brought one at a time while in the latter, the perpetrator is lined up along with the fillers. Wells et al (581) were keen to improve person identification after referring to a guide prepared by US department of justice incorporating psychological science into judicial system. The need for such collaboration has arisen after the DNA exoneration cases whereby the innocent persons were falsely convicted on the testimony of eyewitnesses. Gabberts et al proceeded with another thought that post event interaction may lead to memory conformity. In their experiment two separate groups witnessed the event in slightly different perspective yet they were told that they are seeing the same video (3). Two groups were interviewed individually and then allowed to discuss the event with the co-witness of other group. The individual recall did not include unseen item and was accurate, while co-witness recall included an unseen item in 70% or more cases. Overall 79% have changed their version in the co-witness recall (1-11). Do these different aspects dealt by various researchers in their methodologies provide significant conclusions The answer certainly encourages future researchers as these findings reveal. Wells et al (582-584) and Stein and Memon (599-603) show agreement with the technique of Fisher and Geiselman which generates 35-75% more recall of event when the components of social dynamics, memory cognition and communication were used correctly. The latter team found that CI increases reporting of forensically relevant information such as whether the person was pointing a gun towards the victim. The errors were more person errors than of location or object (599-603). Wells et al., on the other hand, had a logic that in sequential lineup the witness would use more absolute criterion than the relative one that dominated the simultaneous lineup. It was really so! Mistaken identity was 17 % in sequential lineups compared to 43% in the simultaneous ones (585-587). The experimental results encouraged them to look for other points of lineup, which actually have been the worries of a number of other researchers also (594), such as using double blinding during line up and avoiding the instruction that the actual perpetrator may not be in the lineup (599-603). These examples certainly stress the need for more research attempts in different set ups to ascertain reliability and validity of conclusions. Moreover these will provide new observations also which might have been overlooked earlier. Memon et al ( 1997 b:642-651) however given an altogether different angle by showing no significant advantages of cognitive over standard technique. While Gabberts et al (1-11) cautions that too long a gap and post event discussion may not actually give reasonably true recall of the event. It is yet too early, however to overlook the cognitive advantages those work in association with concentration. The excessive importance to social rapport is to be analyzed critically as discussed later in this article How do children particularly very small one respond to CI Can we apply our observations with the adults to them as such It is important issues since child abuse accuses get off taking advantage of a child's vulnerability. What cautions are necessary while finding facts from a preschooler Orbach et al (734) referred Steinberg and her colleagues stating that open ended question reveal more information from children rather than direct and leading questions. To promote a relaxed and supportive environment, first of all, interviewers are asked to ensure that the room is free of distractions such as other people, noise, toys, and incoming phone calls. Ceci et al did an experiment to know whether misattribution happens with young children They added a new false incident in every new session with children. If children assented to new fictional event after hearing it only once then it was comfort with the experimenter and procedure. Ceci et al found that children needed to hear a fictional incident many times before assenting to it, so it was misattribution. Their observations indicated that 34% of 3-4 years age group and 25% of 5-6 years age group children assented to fictional event in first session while in the last (12th) session this percentage rose to 45 and 40 respectively. They also observed that positive and neutral false events had 88% more chance to be assented than negative and threatening events in these preschoolers. As far as actual events are concerned 91-100% recalled these correctly during the past 12 months (308-311). Memon et al. ( 1997 a :pars 11-12) drew a sample of 109 children in the age group 8-9 year . They were shown a magic show to assess the efficiency of context rebuilding methodology, a very crucial component of CI . For percentage correct recall there was a significant effect (F(1, 59) = 6.17, p < .05) with the CI producing more correct recall. Similarly, the absolute numbers of errors was also significantly higher in the CI condition (F (1, 59) = 4.17, p < .05). After two days of show, it was found that, on average, 88% of the children's recall was correct with 8% errors and 4% confabulated details which increased as sessions for recall were widely gapped in time . The two terms differ in meaning. An error is reporting blue jacket as black while confabulation is reporting something, which was not there (pars 26-28). Further the CI increased percentage correct information for action details (F(1, 59) = 11.77, p Read More