Recent accounts of feeling of knowing (FOK) judgements assume that they arise from an assessment of cue familiarity, whilst retrospective confidence judgements arise from an assessment of the retrieval process. An experiment was conducted to extend this laboratory work to the area of eyewitness memory, in order to examine whether subjects are able to make accurate feeling of knowing judgements and retrospective confidence judgements for eyewitness memory (EM), in contrast to general knowledge (GK). For confidence judgements there was a reliable within-subject assessment of confidence for both GK and EM, but reliable between-subjects confidence-accuracy correlations for general knowledge only. For FOK a different pattern emerged, with no evidence of FOK accuracy for eyewitness memory at all. The theoretical implications of this pattern are discussed.
There is a growing interest in the nature of metacognition, and a realization that an understanding of the processes of metacognition and metamemory is crucial for a full understanding of memory (for a recent review, see Metcalfe and Shimamura, 1994). Much of this work has been based in the laboratory, with the aim of developing theoretical accounts of metamemory and its links to memory processes. Little attempt has been made to apply the recent developments in metamemory theory to applied issues, such as the metacognitive judgements of eyewitnesses. The work that has been conducted on eyewitness metacognition has largely been conducted in a separate framework from the cognitive laboratory (e.g. see Luus and Wells, 1994) and has focused on applied issues such as the validity of confidence judgements as predictors of performance (e.g. Wells and Murray, 1984) and the conditions under which the confidence-accuracy relation is optimal (Bothwell, Deffenbacher and Brigham, 1987;Cutler and Penrod, 1989; Deffenbacher, 1980). This paper represents an attempt to bridge the gap beween recent cognitive theory and the applied issue of eyewitness feeling of knowing and confidence judgements.
The authors would like to thank two anonymous reviewers for their helpful comments on a previous draft of this paper.