Abstract
<p>The mechanisms by which default nudges influence complex decision-making under cognitive load remain poorly understood. Although defaults can shape choice behaviour, behavioural measures alone cannot reveal how they affect covert cognitive processes. We tested whether default nudges reduce task-evoked pupil responses or instead alter their temporal profile during complex decision-making. An online study (N = 89) first behaviourally validated five levels of digit-span working-memory load for use in the subsequent laboratory experiment. In the laboratory study (N = 56), participants completed complex decision task embedded within the validated working-memory paradigm while pupil diameter was recorded. Cognitive load produced the expected parametric increase in encoding-locked pupil responses, confirming the sensitivity of the manipulation to working-memory demands. Behaviourally, default nudges did not significantly influence decision accuracy, default-consistent responding, decision response time, recall accuracy, or recall response time. Pupil responses revealed a different pattern. During the decision interval, participants exposed to a nudge showed more pronounced decision-locked pupil responses than those without, particularly when the nudge was accepted. During recall, this pattern reversed, with the No Nudge group exhibiting larger recall-locked pupil responses than the Nudge group. These phase-dependent effects indicate that default presentation altered the temporal profile of pupil-linked arousal across the task rather than producing a uniform reduction in task-evoked pupil responses. The findings suggest that nudges, and nudge acceptance in particular, is not associated with reduced pupil responses but may involve active evaluation of a preselected response (nudge) before commitment, challenging strong effort-saving accounts of default nudges during complex decision-making.</p>