Geiselman (1996) has raised several issues in his commentary, and suggests that the future is not as bleak for the Cognitive Interview (CI) as we might suggest. In the following reply we have attempted to respond to Geiselman's concerns, particularly about the number of errors, number of questions, interview length, and the use of the CI with children. While we agree with Geiselman that the CI has a great deal of potential as an investigative tool, we emphasise the very definite value of examining the limitations of the CI in the pursuit of an even better technique.
2. One of the critical issues in the recovered memory debate is whether one can elicit accurate recall of long buried memories and whether false memories are possible. In light of this, Memon (1995) questioned the use of the CI in a recovered memories context. We also are careful in our target article, to caution against the use of CI in this context until several issues have been addressed. However, the role of the CI in the therapeutic domain has been alluded to by others and it was specifically mentioned in the British Psychological Society Report on recovered memories (published in January, 1995 and available from the British Psychological Society (BPS)). The reason for raising the topic of recovered memories in the present context was primarily to caution against its use in this domain but also to attract a wider discussion on investigative interviews.
3. Geiselman has also reflected on our emphasis on the absolute number of errors rather than on the error rate. On this point we would like to remind the readers that the decision of whether to consider the absolute or the relative number of errors must be related to the context in which the CI is being used. In line with this we have asked numerous practitioners (including police officers) and academics what they think about the risks of an increase in errors with the CI (given that it yields so much information that could be potentially useful in an investigation). Unfortunately, there does not seem to be any agreement on this issue. Geiselman suggests that one could minimise the absolute number of errors by curtailing the interview. This would indeed achieve a lower number of errors but would, of course, result in less correct information as well. In contrast, one view from practitioners is that the consequences of a high number of errors depends on the nature of these errors. If this is the case then it is doubtful whether the CI can be viewed in a favourable light because the errors we are getting in our lab are mainly concerned with the descriptions of persons (see also Hernandex-Fernaud & Alonso-Quecuty, 1995). Of course there may be several reasons for this and these are fully discussed in the target article. While this is clearly an important mistake in a forensic setting, the point that the target article was trying to impress is that if we can isolate these errors and modify the interview so that the number of "person" errors are minimised that would be a worthwhile task. Kebbell and Wagstaff (1996) provide a discussion of a possible means of achieving this, however, we refer the reader to our reply on these points (Memon & Stevenage, 1996b). In the meantime, a recent meta analysis (Koehnken et al., 1994) provides us with the suggestion that errors are more prevalent both in child witness interviews and when the enhanced version of the CI is used. Consequently, under these two conditions, it would appear to be a sensible and cautious practice to obtain as much corroborating evidence as possible, especially with respect to person information.
4. Geiselman questions the importance of factors such as motivation (of interviewer and interviewee) and demand characteristics. We feel that these are important factors which should not be dismissed lightly. With this in mind we will spend a bit of time examining the discussion from Geiselman et al. (1985). This study compared the cognitive and hypnosis interviews. Geiselman et al.'s reasoning was that if demand characteristics were a plausible explanation for our increase in errors, then the hypnotised subjects should show a high number of errors. However, this was not the case; both techniques were found to generate increases in correct details with no increase in errors. This is at odds with a recent meta-analysis of 24 lab studies which reported that where there was an increase in errors associated with hypnosis interviews there was also an increase in confidence in recall. This may plausibly be the result of the motivational factors or demand characteristics of the hypnosis context (Steblay & Bothwell, 1994). Geiselman et al. (1985) does, however, consider several other explanations for these results including the fact that in the study, investigators got the impression that all subjects were equally motivated. While this may be the case in Geiselman et al.'s study, we should be cautious in generalising this impression to other studies, particularly where there is some reward associated with conducting a successful interview and/or being a good witness.
5. Geiselman moves on to consider the choice of controls in Memon's work. This is clearly a contentious issue and one on which opinions differ. We have already engaged in a lengthy discussion in the target article as to the advantages of using a structured interview (incorporating some elements of the enhanced CI - namely social components) as compared to a standard interview. We will not reiterate those points here but will agree that when it comes to impressing upon police officers that they need to improve their interviewing skills, data using the standard interview is relevant. However, the question we were more interested in addressing in our target article concerns the relative effectiveness of the components of the CI together with the reasons for any differences in effectiveness.
6. It is interesting that Geiselman cites Boon and Noon's (1994) paper as support for the effectiveness of individual elements of the CI. Boon and Noon used the "change perspective" instruction and while this technique does appear to produce additional information, both Memon et al. (1993) and Clifford and George (1996) have encountered several problems in using it (see Memon & Koehnken, 1992, for a full discussion). Moreover, police officers report that they do not like to use this technique as they may be accused of asking the witness to speculate. In addition, we find the "reverse order" instruction does not elicit any more information than an additional retrieval attempt (see paras. 4 and 17 of the target article).
7. Geiselman also expresses concern regarding the number of questions being asked by the interviewer. On this point we need to remember that the enhanced CI (ECI) is different from the original procedure. The ECI interviewers do indeed ask more questions. However, this does not mean that the interviewer is doing all of the talking. The majority of the talking is still done by the witnesses. The increase in number of questions seems to arise from the probing that takes place with context reinstatement and imagery. If the earlier studies of the CI did not use the procedure as described in the Fisher and Geiselman (1992) text, then it may well be that the question number is reduced in these studies.
8. Actually the link between question number and recall is not a straightforward one. There is evidence from a study conducted in the medical domain that the effect of detailed questioning may depend on interviewer style and context. Morriss (1992), in a study of interviewing skills to detect psychiatric problems, found that when patients were interviewed by doctors who used a patient led style of interview (similar to the social elements of the ECI) more information was elicited in response to detailed questions than when patients were interviewed by doctors who did not adopt a patient centred interview style. This concurs with our suggestion that the instruction to "transfer control" could be the key to the success of the ECI.
9. Geiselman also raises the very important issue of differences in interview length across standard, structured and cognitive interviews. In the literature, interview length does vary tremendously depending on the nature of the event, interviewer training, techniques used by interviewers and so forth. Most studies have found that the CI interviews are longer; we have no problem with this. If more information is being generated then we would expect the interviews to be longer. Indeed, this may be a function of establishing a greater degree of rapport between the interviewer and the interviewee. The Memon et al. (1993) study with the 6 minute interview was a special case, however. In this study, the event was very brief (we looked at children's memory for an eye test) and the children had difficulty using the CI mnemonics (see also Memon et al., 1996). In fact, we found that rather than getting the children to talk, use of the original CI instructions in this study actually inhibited them.
10. Geiselman closes with a reflection of the usefulness of the CI with children. The aim of the ongoing research with preschoolers is to see how well the CI compares with other procedures for interviewing children of this age group. With Memon's previous data in mind, we agree with Geiselman's hope that current research will direct the modification of the CI for use with children. As indicated by Geiselman in his commentary, the work of Karen Saywitz and colleagues (e.g., Saywitz & Geiselman, in press) offers us a good model for future work in this area.
11. As indicated at the start of our target article the CI holds great promise in the field of investigative interviewing. There is no doubt that it has the potential to produce more information and more detailed witness reports. We therefore agree with Geiselman that the CI has considerable value in the forensic field. Our paper does not question the general usefulness of the technique but instead, our aim was to raise what appear to us to be interesting theoretical, methodological and practical questions which may modify the use of the CI. We have also attempted to draw attention to the limitations of the CI in the hope that further research would result in an even more effective procedure.
Boon, J., & Noon, E. (1994). Changing perspectives in cognitive interviewing. Psychology, Crime and Law, 1, 59-69.
Clifford, B., & George, R. (1996). A field evaluation of training in three methods of witness/victim investigative interviewing. Psychology, Crime and Law 2, 231-248.
Fisher, R.P. & Geiselman, R.E. (1992). Memory enhancing techniques for investigative interviewing: The Cognitive Interview. Springfield III: Charles C. Thomas.
Geiselman, R.E. (1996). On the Use and Efficacy of the Cognitive Interview. PSYCOLOQUY 7(6) witness-memory.2.geiselman.
Geiselman, R.E., Fisher, R.P., MacKinnon, D.P. & Holland, H.L. (1985). Eyewitness memory enhancement in the police interview: Cognitive retrieval mnemonics versus hypnosis. Journal of Applied Psychology, 70, 401-412.
Hernandex-Fernaud, E. & Alonso-Quecuty, M. (1995). Cognitive Interview confronting true and false statements. Paper presented at the V European Conference on Psychology and Law, Sept, Budapest.
Kebbell, M.R. & Wagstaff, G.F. (1996). Enhancing the Practicality of the Cognitive Interview in Forensic Situations: Commentary on Memon and Stevenage. PSYCOLOQUY 7(6) witness-memory.3.kebbell.
Koehnken, G., Thurer, C. & Zorberbier, D. (1994). The cognitive interview: Are interviewers memories enhanced too? Applied Cognitive Psychology, 8, 13-24.
Memon, A., (1995). Response to the British Psychological Society report on recovered memories. Expert Evidence, 3(4), 155-158.
Memon, A., Cronin, O., Eaves, R. & Bull, R. (1996) An empirical test of the mnemonic components of the Cognitive Interview. In G.M. Davies, S. Lloyd-Bostock, M. McMurran & C. Wilson (Eds) Psychology, Law & Criminal Justice. (pp. 135-145). Berlin: De Gruyter. http://www.soton.ac.uk/~am/mceb1996.html
Memon, A., Cronin, O., Eaves, R. & Bull, R. (1993). The cognitive interview and child witnesses. In G.M. Stephenson, & N.K. Clark, (Eds.). Children, Evidence, and Procedure. Issues in Criminological and Legal Psychology. No. 20. British Psychological Society, Leicester, UK.
Memon, A. & Koehnken, G. (1992). Helping the witness to remember more: The Cognitive Interview. Expert Evidence: the International Digest of Human Behaviour, Science and Law 1, 39-48.
Memon, A. & Stevenage, S.V. (1996). Interviewing Witnesses: What Works and What Doesn't? PSYCOLOQUY 7(6) witness-memory.1.memon.
Memon, A. & Stevenage, S.V. (1996b). The Importance of Time and Training for Cognitive Interviewers: A Reply to Kebbell and Wagstaff. PSYCOLOQUY 7(6) witness-memory.5.memon.
Morriss, R. (1992). Interviewing skills and detection of psychiatric problems. International Review of Psychiatry, 4, 287-92.
Saywitz, K.J. & Geiselman, R.E. (in press). Interviewing the child witness: Maximising completeness and minimising error. In S. Lynn (Ed.), Truth and Memory. New York: Guilford.
Steblay, N.M. & Bothwell, R. (1994). Evidence for hypnotically refreshed testimony. Law and Human Behaviour, 18(6), 635-651.