An adaptive drift-diffusion model of interval timing dynamics.

Area of Science:

• Cognitive psychology
• Computational neuroscience
• Animal behavior

Background:

• Animals exhibit remarkable interval timing abilities, adapting to changing temporal cues.
• Drift-diffusion models (DDMs) model decision-making but require extensions for interval timing.
• Existing time-adaptive DDMs (TDDMs) explain steady-state timing but struggle with non-stationary intervals.

Purpose of the Study:

• To extend time-adaptive drift-diffusion models (TDDMs) to account for interval timing dynamics under non-stationary conditions.
• To propose a novel TDDM variant where the response threshold is a linear function of the observed event rate.
• To evaluate the proposed model against existing timing models using empirical data.

Main Methods:

• Introduced a modified time-adaptive drift-diffusion model (TDDM) with a rate-dependent response threshold.
• Compared the model's performance against standard TDDMs and the multiple-time-scale (MTS) habituation model.
• Validated the models using three published datasets on interval timing in pigeons.

Main Results:

• The extended TDDM with a dynamic threshold demonstrated superior performance in modeling interval timing dynamics.
• The model successfully captured deviations from timescale invariance observed in non-stationary timing tasks.
• Results indicate that the response threshold is modulated by recent interval history.

Conclusions:

• The proposed TDDM extension provides a more accurate account of interval timing under changing temporal conditions.
• Response threshold adaptation is a critical mechanism for flexible interval timing.
• This work advances computational models of timing and decision-making.