Using a Wald model to study time perception in intervals with varying proportions of cognitively demanding and idle phases

Human time perception is strongly influenced by cognitive load. When individuals estimate durations while simultaneously performing a cognitively demanding non-timing task, they typically underestimate time; conversely, when there is no concurrent task–or when it requires low effort–durations tend to be overestimated. However, no prior study has investigated how subjective time is shaped when a single interval contains both cognitively demanding and relatively idle (low-demand) periods in varying proportions. To address this question, we introduced a novel interval-reproduction task in which participants first chose between two lotteries (decision-making phase) and then waited until a deadline (idle phase). Participants (n = 21) then reproduced the total duration spanning both phases. Decision difficulty and deadline duration were manipulated to vary the fraction of the interval spent in each phase. We modelled reproductions as an evidence accumulation process using a Hierarchical Bayesian Wald model. To our knowledge, this is the first study to apply an evidence accumulation model to investigate time reproductions of intervals under varying levels of cognitive load. We found a trade-off between susceptibility to time modulation by changes in the relative durations of the loaded and idle phases and the strength of the central tendency bias.