Abstract
We tested whether someone's ability to tell a good story, in terms of the Reality Monitoring (RM) tool, affects the way s/he judges the stories told by others. Forty participants (undergraduate students) wrote down two statements – one about activities they did 30 minutes ago, and the other about a past event. Subsequently, they rated the quality of a target statement written by someone else. We found that the tendency to provide a not so detailed or a very detailed statement was stable across the two statements the participants wrote. Furthermore, this tendency affected how they judged the target statements: The richer a participant's statements were compared to the target statement, the more critical the participant was in judging the target statement. These findings imply that RM is subject to biases which are related to individual differences. We discuss the implications of these findings for applying the RM lie detection tool in the field.
Acknowledgements
This research was supported by the Israel Science Foundation grant (grant No. 59/11). We thank Hofit Vizman-Babay, Mor Cohen-Peled and Oshrit Turgeman for their assistance in this research.
Notes
1. Originally we had 41 participants. One participant was eliminated as her data were exceptional. She provided 97 perceptual details in her statements compared to an average of 22.63 details and was a statistical outlier in all the analyses.
2. The participants rated the criteria on scales rather than counting the frequency of occurrence mainly because scale rating is closer to what people typically do when they assess richness in detail while attempting to detect lies.
3. To illustrate this suggestion we ran two discriminant analyses for distinguishing between false and truthful statements (the target statement included false and truthful descriptions of activities). The discriminant function for the frequency coding of perceptual and contextual details correctly classified 82.5% of the statements, χ2(3) = 27.48, Wilk's Lambda = 0.47, p < 0.001, while the discriminant function for the participants' scale ratings of the same criteria was not significant. These findings support our suggestion to prefer frequency counts over scale ratings. However, one should note that each statement was rated by a different participant, while the frequency coding was carried out by the same person. This difference could have affected the discriminant functions.