Fitting model-based psychometric functions to simultaneity and temporal-order judgment data: MATLAB and R routines

Area of Science:

• Cognitive psychology
• Computational neuroscience
• Psychophysics

Background:

• Temporal-order perception is studied using temporal-order judgment (TOJ) and synchrony judgment (SJ) tasks.
• Current methods fit arbitrary functions, yielding uninterpretable parameters for sensitivity and subjective simultaneity.
• A need exists for model-based analyses with interpretable parameters in temporal perception research.

Purpose of the Study:

• To develop and describe routines for fitting model-based psychometric functions to TOJ and SJ data.
• To provide interpretable parameters reflecting underlying perceptual processes.
• To offer flexible routines for analyzing single or multiple task data, including bootstrap statistics.

Main Methods:

• Development of routines in MATLAB and R implementing an independent-channels model.
• Model assumes arrival latencies with exponential distributions and a trichotomous decision space.
• Routines are designed to fit data from SJ2, SJ3, and TOJ tasks, individually or jointly, with options for bootstrap analysis.

Main Results:

• The developed routines fit model-based functions with interpretable parameters for temporal-order and simultaneity judgments.
• Routines accommodate fitting data from single tasks (SJ2, SJ3, TOJ) or combinations thereof.
• Provision of bootstrap p-values, confidence intervals, and performance measures from fitted functions.

Conclusions:

• The new routines provide a powerful tool for analyzing temporal-order and synchrony judgment data.
• Interpretable parameters offer deeper insights into the mechanisms of temporal perception.
• Available R and MATLAB code facilitates broader adoption and application in the field.