In their commentary, Higham & Roberts argue that the increased errors that emerge when using the cognitive interview (CI) may be reduced by taking account of whether the witness "remembers" a detail or merely "knows" it. In this reply we discuss evidence from Bekerian and Dennett (1994) which addresses this issue directly. Their results would suggest that, contrary to Higham & Roberts' suggestion, the errors in recall are not characterised by a "know" status.
2. Bekerian and Dennett reasoned as follows: In standard memory conditions "know" responses dominate when information is incorrect, but the CI promotes enhanced memory. Hence it is possible that incorrect information when recalled under the CI will be erroneously regarded as being remembered and be accompanied by contextual detail. In other words the errors reflect "remember" responses. This makes sense since one of the major CI techniques is to reinstate context. Bekerian and Dennett conducted a somewhat complex study to test their hypotheses. In the first phase, participants viewed a crime video and performed a rating task and gave a free narrative account of the video. The next phase took place 24 hours later. After random assignment to CI and standard interview conditions they were told they should forget what they had said previously and remember all they could. The CI group were given context reinstatement and "report in detail" instructions. All participants were probed for details. Phase 3 took place 24 hours later. Participants read statements they had made in their recall and the question was whether they remembered circumstances under which the information occurred or whether they knew the information. For each statement a "remember"/"know" judgment was requested. The statement was the exact description given by the participant in Phase 2 (verbatim). The experimenter made it clear that she did not know if the information in the statement was correct or incorrect. (We refer readers to the paper itself for the detailed description of the criteria used to select statements).
3. The results of this study are theoretically very important. Taking the interview (phase 2) data first, the CI manipulation was found to produce more correct information and more errors. The analysis of "remember"/"know" responses was complex. Essentially, it was found that the CI did not produce a greater percentage of remember responses to errors. However the pattern of errors was different across the standard and cognitive conditions. The CI group showed more remember responses to errors than would be expected by chance. The authors draw the following conclusion from their work: "Such data offer some preliminary findings to suggest that techniques like the CI are still in need of investigation in the laboratory in order to provide a more complete understanding of their use".
Bekerian, D.A. & Dennett, J.L. (1994). The fate of errors produced under the Cognitive Interview. Paper presented at the Practical Aspects of Memory Conference, Maryland, July.
Higham, P.A. & Roberts, W.T. (1996). Analyzing States of Consciousness During Retrieval as a Way to Improve the Cognitive Interview. PSYCOLOQUY 7(6) witness-memory.4.higham.
Memon, A. & Stevenage, S. (1996). Interviewing Witnesses: What Works and What Doesn't? PSYCOLOQUY 7(6) witness-memory.1.memon.