Measuring psychometric functions with the diffusion model.

Area of Science:

• Cognitive Psychology
• Perceptual Science
• Decision Making Models

Background:

• Traditional psychometric functions often rely solely on accuracy, which can reach ceiling effects, limiting evaluation of high-performance conditions.
• The diffusion decision model (DDM) offers a method to assess performance beyond accuracy by incorporating response times.
• Comparing the DDM's drift rate with signal detection theory's measures provides insights into decision processes.

Purpose of the Study:

• To evaluate the DDM's applicability across diverse perceptual discrimination tasks.
• To compare the DDM's drift rate with signal detection theory measures.
• To investigate the DDM's ability to extend the range of evaluated difficulty levels.

Main Methods:

• The diffusion decision model (DDM) was applied to numerosity, number, brightness, motion, speed, and length discrimination tasks.
• Psychometric functions were constructed by plotting drift rate against task difficulty.
• Drift rates were analyzed in relation to accuracy and response times across different difficulty levels.

Main Results:

• The DDM's drift rate closely matched signal detection theory measures when accuracy was not at ceiling.
• The DDM effectively distinguished between conditions even when accuracy was at ceiling, by utilizing response time data.
• Discrepancies were observed between the models at high accuracy levels, suggesting different interpretations of variability.

Conclusions:

• The diffusion decision model (DDM) provides a robust measure of perceptual performance, extending beyond accuracy limitations.
• The DDM offers a valuable alternative and complement to signal detection theory for understanding decision processes.
• Further research is needed to reconcile the differing interpretations of variability between the DDM and signal detection theory.