Sometimes-competing retrieval (SOCR): a formalization of the comparator hypothesis

Area of Science:

• Cognitive Psychology
• Behavioral Neuroscience
• Animal Learning

Background:

• Cue competition is a key phenomenon in associative learning, with ongoing debate regarding its nature: learning deficit vs. performance deficit.
• The comparator hypothesis suggests performance, not learning, is subject to competition, based on changes in reinforcement.
• Recent learning-focused models and cue facilitation phenomena complicate existing theories.

Purpose of the Study:

• To formalize and extend the comparator hypothesis.
• To sharpen the differentiation between the comparator hypothesis and recent learning-focused models of cue competition.
• To address phenomena of cue facilitation within a formal model of conditioning.

Main Methods:

• Theoretical modeling and formalization of the comparator hypothesis.
• Extension of the comparator hypothesis to account for a broader range of conditioning phenomena.
• Comparative analysis of the extended comparator hypothesis against existing learning-focused models.

Main Results:

• The formalized comparator hypothesis provides a clearer distinction from learning-focused accounts.
• The extended model offers a framework for understanding both cue competition and facilitation.
• The comparator hypothesis, as extended, maintains its focus on performance as a function of reinforcement change.

Conclusions:

• The comparator hypothesis offers a robust framework for understanding associative learning phenomena.
• Distinguishing between learning and performance deficits remains crucial in associative learning research.
• The formalized model provides a basis for future empirical investigations into cue competition and facilitation.