Exemplar-based judgment or direct recall: On a problematic procedure for estimating parameters in exemplar models of...

Area of Science:

• Cognitive Psychology
• Decision Making
• Computational Modeling

Background:

• Exemplar models are standard in multiple-cue judgment research to understand participant responses.
• These models often assume a single process for judging learned exemplars, overlooking distinct cognitive mechanisms.

Purpose of the Study:

• To investigate the impact of conflating judgment and recall processes in exemplar models.
• To evaluate the effects on parameter recovery, model fit, and validity in multiple-cue judgment research.

Main Methods:

• A simulation study was conducted to assess parameter recovery and model fit.
• Reanalysis of five existing experimental datasets using current and a proposed latent-mixture model.

Main Results:

• Disregarding the distinction between judgment and recall biases parameter estimates and impairs model validity.
• Current modeling procedures negatively affect model fit and predictive performance.
• A latent-mixture extension of exemplar models offers a potential solution.

Conclusions:

• Current exemplar modeling practices in multiple-cue judgments require refinement to account for distinct cognitive processes.
• The proposed latent-mixture model provides a more accurate approach to analyzing judgment and recall.
• Accurate modeling is crucial for valid parameter estimation and understanding cognitive processes.