Related Experiment Videos
The regularities of recognition memory
M Glanzer1, J K Adams, G J Iverson
1Department of Psychology, New York University, New York 10003.
Psychological Review
|July 1, 1993
Summary
This study introduces attention/likelihood theory to explain recognition memory. It demonstrates how likelihood ratios account for key memory regularities, suggesting a revision of existing theories.
Area of Science:
- Cognitive Psychology
- Human Memory
- Decision Making
Background:
- Recognition memory research has identified several empirical regularities.
- Existing theories struggle to comprehensively explain these phenomena.
- Signal detection theory provides a framework but requires extensions for recognition memory.
Purpose of the Study:
- To introduce and demonstrate attention/likelihood theory for recognition memory.
- To explain three key regularities: the mirror effect, ROC slope order, and distribution symmetry.
- To show how likelihood ratios are central to recognition decisions.
Main Methods:
- Derivation of regularities from attention/likelihood theory.
- Fitting the theory to data from two experimental series.
- Analysis of yes-no tests with confidence ratings.
- Analysis of forced-choice experiments.
Main Results:
- Attention/likelihood theory successfully accounts for the mirror effect.
- The theory explains the observed order of receiver operating characteristic (ROC) slopes.
- Symmetry in the movement of underlying distributions is explained by the theory.
- Likelihood ratios are shown to be central to recognition decision processes.
Conclusions:
- Attention/likelihood theory offers a unified explanation for key recognition memory regularities.
- The theory extends the implications of signal detection theory for recognition memory.
- The findings necessitate a revision of most current recognition memory theories.