Retrieving information from memory: spreading-activation theories versus compound-cue theories

Area of Science:

• Cognitive Psychology
• Memory Research
• Computational Neuroscience

Background:

• Compound-cue theories and spreading-activation models are prominent in explaining memory retrieval.
• McNamara (1992a) raised criticisms regarding compound-cue theories' ability to explain mediated priming and sequential effects.
• Previous research relied on free association probabilities to evaluate these memory models.

Purpose of the Study:

• To re-evaluate McNamara's criticisms of compound-cue theories.
• To investigate the predictive power of free-association production probabilities on priming effects.
• To determine the compatibility of compound-cue models with sequential effects in memory.

Main Methods:

• Analysis of free-association production probabilities in relation to priming effects.
• Examination of sequential effects using modified compound-cue models with differential item weighting.
• Comparison of empirical data with predictions from compound-cue and spreading-activation models.

Main Results:

• Free-association production probabilities were found to be poor predictors of priming effects.
• Sequential effects in response time and accuracy are explainable by compound-cue models when target items receive higher weighting.
• Both compound-cue and spreading-activation models align with existing data.

Conclusions:

• McNamara's criticisms based on free association are unsubstantiated.
• Compound-cue theories provide a parsimonious and empirically supported explanation for mediated priming and sequential effects.
• Compound-cue theory, despite its constraints, has demonstrated greater empirical rigor.