What is it about?

This study addresses a long-standing debate in recognition memory research: is recognition continuous or discrete? Instead of choosing one side, we propose a hybrid approach that integrates elements from both perspectives. We compared three models of recognition memory: a continuous model, where recognition depends on the strength of familiarity; a discrete model, based on detecting or not detecting an item, with guessing when detection fails; and a hybrid model, in which familiarity is followed by a secondary search for information when uncertainty arises.

Featured Image

Why is it important?

The hybrid model can distinguish between short, high-confidence responses that arise from a familiarity decision and long, high-confidence responses that result from a slower, secondary search for information, a pattern of data that other models lack a mechanism to explain. Moreover, our modeling approach extends traditional frameworks by incorporating response times, confidence, and accuracy measures that are rarely evaluated simultaneously but are essential for detecting specific data patterns and allow us to compare the three models. Overall, this study provides a theoretical framework for explaining variability across individuals in recognition memory strategies.

Perspectives

In a previous simulation study with the same co-authors (Gutkin et al., 2024), we explored how adding continuous variables could improve model comparisons, achieving a very high rate of correct detection of the data-generating model. However, when we applied this approach to real data, we found that some participants were better described by a continuous model, while others fit more closely to a discrete one. This pattern, which also appears repeatedly in the literature, led us to a new idea: there must be a model capable of integrating both mechanisms. Building on the traditional Atkinson–Juola (AJ) model (Atkinson & Juola, 1973, 1974; Juola et al., 1971), we extended and generalized it to achieve this goal. We see this work as an initial step toward a deeper understanding of individual differences in the mechanisms underlying recognition memory.

Anahí Gutkin
Universidad Francisco de Vitoria

Read the Original

This page is a summary of: Do all subjects fit the same recognition memory model? Comparisons of continuous, discrete, and hybrid models using extended multinomial processing trees., Journal of Experimental Psychology Learning Memory and Cognition, January 2026, American Psychological Association (APA),
DOI: 10.1037/xlm0001574.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page