How sequential changes in reward magnitude modulate cognitive flexibility: Evidence from voluntary task switching.

Abstract

There is much evidence that the prospect of reward modulates cognitive control in terms of more stable behavior. Increases in expected reward magnitude, however, have been suggested to increase flexible behavior as evidenced by reduced switch costs. In a series of experiments, the authors provide evidence that this increased cognitive flexibility following increases in reward magnitude also promotes deliberate task switching. A modified task switching paradigm with forced- and free-choice trials and varying reward prospects was used. In Experiments 1-3 the prospect of a reward increase as compared to unchanged high reward increased voluntary switching rate (VSR). Experiment 4 showed that the prospect of a reward decrease did not alter VSR as compared to unchanged low reward. Experiment 5 used a standard voluntary task switching procedure and confirmed VSR effects found in Experiments 1-4. These findings are strong evidence for a mechanism that biases the cognitive system either toward stability or flexibility depending on changing reward expectation. Results are discussed within the framework of the adaptive gain theory.

Striatal-frontal network activation during voluntary task selection under conditions of monetary reward

“…Typically, in such paradigms, a cue indicates which of the two tasks a participant should perform on any given trial. In fact, several behavioral task-switching studies have now demonstrated that under certain reward structures flexibility may be increased (Fröber & Dreisbach, 2016b;Kleinsorge & Rinkenauer, 2012;Shen & Chun, 2011). In all these studies, switch costs were reduced when reward value increased from the previous trial (e.g., low reward trial followed by a high reward trial) compared with when reward value remained the same or decreased.…”

“…A series of recent behavioral studies examined how task choice on VTS paradigms might be affected by changes in reward magnitude in trials immediately preceding the choice. (Fröber & Dreisbach, 2016b). Four of these studies mixed cued and voluntary task switching, with an additional experiment that used a standard VTS paradigm with only voluntary choice trials.…”

Striatal-frontal network activation during voluntary task selection under conditions of monetary reward

“…However, when reward is only presented infrequently, and is contingent on performance, conflict adaptation is enhanced following gains (Braem, Verguts, Roggeman, & Notebaert, 2012;Stürmer, Nigbur, Schacht, & Sommer, 2011)] Sequential changes in reward magnitude, on the other hand, appears to be more closely linked to flexibility. Increased magnitude of potential reward from the previous to the current trial, relative to decreased or consistent reward prospects, has been associated with reduced switch costs and increased proportion of voluntary task switches (Fröber & Dreisbach, 2016b;Kleinsorge & Rinkenauer, 2012;Shen & Chun, 2011). In each of these studies, participants could receive reward for producing responses faster than some criterion (e.g., top percentile of RTs in the previous block).…”

“…Typically, in such paradigms, a cue indicates which of the two tasks a participant should perform on any given trial. In fact, several behavioral task-switching studies have now demonstrated that under certain reward structures, flexibility may be increased (Fröber & Dreisbach, 2016b;Kleinsorge & Rinkenauer, 2012;Shen & Chun, 2011). In all these studies, switch costs were reduced when reward value increased from the previous trial (e.g., low reward trial followed by a high reward trial) compared to when reward value remained the same or decreased.…”

“…While the study by Fröber & Dreisbach (2016b) demonstrated that task choice on a single trial is influenced by reward on the preceding trial, a recent study by Braem (2017) demonstrated that repeated rewards can bias subsequent task choices for the remainder of the task. In this study, all participants performed a block of trials in which they were explicitly cued as to the task to perform, and hence were directed when to switch tasks, followed by a block of trials in which they themselves voluntarily decided when to switch tasks.…”