Memory framing.

Abstract

Some prior studies of item recognition, source recognition, and judgments of learning have produced evidence of framing effects in episodic memory. A theoretical consequence of these phenomena is that emotional arousal is not a necessary condition for framing effects because, unlike the classic framing effects of social psychology and behavioral economics, different memory frames are not designed to stimulate different emotional reactions. However, a review of available evidence on framing effects in item recognition revealed that the data were inconsistent for old items and nonexistent for similar distractors. We attempted to secure definitive evidence of whether these types of items display framing effects by analyzing a corpus of 478 sets of data, in which old items and similar distractors are factorially crossed with old? and similar? recognition frames. Both types of items exhibited large framing effects: Recognition was far more accurate for similar items than for old items when probes asked if test items were old, whereas recognition was more accurate for old items than for similar items when probes asked if test items were similar. There was a frame-independence effect, too, such that accuracy in one memory frame was dissociated from accuracy in the other frame. The conjoint-recognition model predicted both the core memory framing effect and its direction (superior recognition of similar distractors in the old? frame but superior recognition of old items in the similar? frame). The model also explained frame independence as a by-product of between-frame differences in how three retrieval processes (true recollection, false recollection, and semantic familiarity) affect accuracy in the different frames. (PsycInfo Database Record (c) 2025 APA, all rights reserved).