Intersections of behavior analysis with cognitive models of contingency detection

Area of Science:

• Cognitive Psychology
• Developmental Psychology
• Infant Learning

Background:

• Bower and Watson proposed two models for infant contingency detection: logical hypothesis-testing and conditional probability.
• These models aim to explain cognitive processes linked to operant learning in infants.

Purpose of the Study:

• To evaluate the empirical support for Bower's and Watson's models of infant contingency detection.
• To identify the limitations of each proposed model.
• To suggest a more parsimonious theoretical approach by integrating overlapping concepts.

Main Methods:

• Critical analysis of existing literature on infant contingency detection models.
• Discussion of the theoretical underpinnings and empirical evidence (or lack thereof) for each model.
• Comparative analysis to identify areas of convergence and divergence between the models.

Main Results:

• Empirical evidence supporting both the logical hypothesis-testing and conditional probability models is currently sparse.
• Each model, while potentially relevant to operant learning, faces significant limitations in explaining infant behavior.
• Significant conceptual overlap exists between the two models, suggesting potential for integration.

Conclusions:

• The current empirical basis for Bower's and Watson's models of infant contingency detection is weak.
• A more parsimonious and potentially more accurate understanding can be achieved by focusing on the shared elements of these models.
• Future research should aim to empirically test integrated models of infant contingency detection.