Overdistribution illusions: Categorical judgments produce them, confidence ratings reduce them

Abstract

Overdistribution is a form of memory distortion in which an event is remembered as belonging to too many episodic states, states that are logically or empirically incompatible with each other. We investigated a response formatting method of suppressing 2 basic types of overdistribution, disjunction and conjunction illusions, which parallel some classic illusions in the judgment and decision making literature. In this method, subjects respond to memory probes by rating their confidence that test cues belong to specific episodic states (e.g., presented on List 1, presented on List 2), rather than by making the usual categorical judgments about those states. The central prediction, which was derived from the task calibration principle of fuzzy-trace theory, was that confidence ratings should reduce overdistribution by diminishing subjects' reliance on noncompensatory gist memories. The data of 3 experiments agreed with that prediction. In Experiment 1, there were reliable disjunction illusions with categorical judgments but not with confidence ratings. In Experiment 2, both response formats produced reliable disjunction illusions, but those for confidence ratings were much smaller than those for categorical judgments. In Experiment 3, there were reliable conjunction illusions with categorical judgments but not with confidence ratings. Apropos of recent controversies over confidence-accuracy correlations in memory, such correlations were positive for hits, negative for correct rejections, and the 2 types of correlations were of equal magnitude. (PsycINFO Database Record

Figure 1

Disjunction illusions in 264 sets of conjoint recognition data. O = old cues (targets) and NS = new-similar distractor cues. O? = the judgment that a cue is old, NS? = the judgment that a cue is new-similar, and O-NS? = the judgment that a cue is either old or new-similar. Panel A plots the mean probabilities of making each one of these judgments for O and NS cues. Panel B plots the mean overdistribution probability p(O) + p(NS) − p(O-NS) for both O and NS cues.

Figure 2

Relations between categorical judgments and confidence ratings, pooled over the conditions of Experiments 1–3. Panel A = the separation between response probabilities for correct (black bars) versus incorrect white bars) source probes that is produced by categorical judgments and confidence ratings. Panel B = the best-fitting regression line with correct source probes, for categorical judgments versus confidence ratings. Panel C = the best-fitting regression line with incorrect source probes, for categorical judgments versus confidence ratings.