The Generality of Dynamic Adjustments in Decision Processes across Trials and Tasks.

Abstract

There is a growing interest in understanding how aspects of binary decision-making change dynamically at the trial level. For example, in lexical decision, there is a well-established interaction between current and previous trial characteristics (e.g., lexicality and stimulus degradation) that suggests that participants adjust their decision processes based on the relative match in signal strength between the current and previous trial. The present study assessed the generality of this finding by examining the presence of such cross-trial adjustments in two new tasks, syntactic classification, and memory scanning. Stimulus degradation is manipulated in both tasks. Results indicate that response latencies are facilitated when salient aspects of the stimulus repeat across trials. These findings are interpreted within the context of a flexible processor that utilizes information from the prior target to prime the relevant processing pathway on the current trial. Candidate models that potentially can accommodate the pattern are briefly discussed.