Interactions of stimulus attributes, base rates, and feedback in recognition

Area of Science:

• Cognitive Psychology
• Human Memory Research

Background:

• Previous research on continuous old-new recognition has overlooked stimulus attributes, old-new base rates, and feedback effects.
• Understanding these factors is crucial for refining memory models.

Purpose of the Study:

• To investigate the impact of stimulus attributes, old-new base rates, and informative feedback on continuous old-new recognition.
• To test predictions derived from composite-memory and array-similarity models.

Main Methods:

• Continuous old-new recognition task administered under varying conditions of stimulus properties, old-new base rates, and feedback.
• Analysis of hits and false alarms across trials and their dependence on experimental factors.

Main Results:

• Both hits and false alarms increased over trials, with recognition measures strongly dependent on stimulus properties like interitem similarity.
• Hit and false-alarm rates were independent of old-new base rates without feedback.
• With feedback, false-alarm rates approached true old-stimulus base rates for some stimuli, but not for words.

Conclusions:

• Current composite-memory models generally align with the findings.
• An array-similarity model accurately predicted the observed recognition patterns, highlighting its utility in explaining categorization-based memory.
• Stimulus attributes and feedback play significant roles in recognition memory, moderating the influence of base rates.