Evaluating models of remember-know judgments: complexity, mimicry, and discriminability

Area of Science:

• Cognitive Psychology
• Neuroscience
• Memory Research

Background:

• Remember-know judgments in recognition memory are debated.
• Competing models (one-dimensional, STREAK, dual-process) exist for these judgments.

Purpose of the Study:

• To evaluate competing models of remember-know judgments.
• To assess model fit and complexity in recognition memory.
• To determine the most suitable model for explaining memory data.

Main Methods:

• Simulated data using three popular remember-know paradigms for each competing model.
• Fit simulated data to each of the competing models.
• Compared model complexity and goodness-of-fit.

Main Results:

• The one-dimensional model, despite being less complex, provided the best fit to the data.
• Model complexity can influence perceived fit, potentially mimicking other models.
• The trinary task (remember/know/new judgments) is less effective for distinguishing between models.

Conclusions:

• The one-dimensional model is preferred due to its superior fit and relative simplicity.
• Researchers should be cautious when using certain paradigms, like the trinary task, for model discrimination.