An exemplar-retrieval model of speeded same--different judgments.

Area of Science:

• Cognitive Psychology
• Computational Neuroscience
• Perception

Background:

• The exemplar-based random-walk (EBRW) model previously explained speeded classification.
• Understanding same-different judgments is crucial for cognitive models.

Purpose of the Study:

• To extend the EBRW model to account for speeded same-different judgments.
• To investigate the role of exemplar memory in same-different decisions for integral-dimension stimuli.

Main Methods:

• Extension of the EBRW model to incorporate same-different judgment processes.
• Modeling the retrieval of exemplar pairs from memory based on similarity.
• Simulating a random-walk decision process driven by retrieved exemplars.

Main Results:

• The extended EBRW model accurately predicts that same responses are faster for stimuli in isolated regions of similarity space compared to dense regions.
• The model correctly accounts for the influence of same-identity versus same-category instructions on judgment speed.
• The model demonstrates sensitivity to individual observers' past experiences with specific stimulus pairs.

Conclusions:

• The exemplar-based random-walk framework provides a robust account of speeded same-different judgments.
• Memory for specific exemplars and their relationships plays a key role in perceptual decision-making.
• The model's success highlights the importance of integrating memory retrieval and decision processes in cognitive modeling.