Component processes underlying voluntary task selection Institutional Repository Document

Most theories describing the cognitive processes underlying task switching allow for contributions of active task-set reconfiguration and task set inertia. Manipulations of the Cue-to-Stimulus-Interval (CSI) are generally thought to influence task set reconfiguration, while Response-to-Stimulus Interval (RSI) manipulations are generally thought to influence task set inertia (i.e., proactive interference from the previous task-set). However, these theories do not adequately account for the processes underlying voluntary task selection, because a participant can theoretically prepare for an upcoming trial at any point. To this end we used drift diffusion models to examine the contributions of reconfiguration and task set inertia in 216 undergraduate students who performed either cued or voluntary task switching paradigms. In both task versions, longer CSIs allowed for better preparation on all trial types. For the voluntary condition, but not the explicit condition, longer RSIs also reduced the effect of switching on preparation when CSIs were short. Further, when given enough time to prepare, participants in the voluntary version prepared more efficiently for switches than repeats. Together, these results indicate the use of a more proactive strategy when participants chose to switch in the voluntary version. In both paradigms, RSI manipulations produced the expected effect on switch costs; however, they consistently slowed repeat performance and generally did not affect performance on switch trials. The results suggest that drift diffusion models can quantify differences in strategy across voluntary and explicit task switching as well as measure contributions of inertia and preparation to voluntary task switching performance, including identifying preparation that occurs outside of the CSI in voluntary switching. The results also suggest that reductions in switch cost caused by reduced inertia might be more related to impeding repeat performance rather than facilitating switch performance. Future work should extend the current findings with manipulations of proactive vs. reactive strategies and other manipulations of inertia.