Judgments of associative memory

Area of Science:

• Cognitive Psychology
• Memory Research
• Computational Neuroscience

Background:

• Understanding associative memory is crucial for explaining human cognition.
• Previous research has explored various factors influencing memory judgments.
• The relationship between subjective memory ratings and objective associative strengths requires further elucidation.

Purpose of the Study:

• To investigate the functional relationship between subjective judgments of associative memory (JAM) and objective measures of associative strength.
• To characterize the parameters of the JAM function, specifically its intercept and slope.
• To explore factors influencing the JAM function and test a computational model's ability to replicate observed phenomena.

Main Methods:

• Participants provided ratings for cue-response word pairs.
• Normed probabilities of word associations (p(response|cue)) were obtained from free association norms.
• Statistical analyses examined the linear relationship between ratings and probabilities.
• A computer simulation model (MINERVA 2) was used to model JAM functions.

Main Results:

• A robust linear relationship was found between JAM ratings and associative probabilities.
• The JAM function consistently exhibited high intercepts (approx. 50/100) and shallow slopes (<0.5).
• The JAM function's characteristics were invariant across different materials and rating methods, and unaffected by expectancies or semantic similarity.
• The MINERVA 2 model successfully replicated the linear JAM function and explained more complex associative memory phenomena.

Conclusions:

• The shallow slope of the JAM function indicates a limited discrimination among varying associative strengths.
• The high intercept suggests an independent post-mnemonic bias contributing to the overestimation of associations.
• These findings provide insights into the cognitive mechanisms underlying associative memory judgments and highlight the role of bias in memory recall.