The Mechanism of Dissociation in Reward-Based Dual-task Processing: An ERP Study

Abstract

The process of integrating working memory with attentional resource is referred to as branching. Branching is observed in everyday life, which enables people to hold a primary goal in mind while exploring and processing a secondary goal. The hemispheric effect of branching has been noted in several dual-task studies using high spatial resolution brain imaging techniques (e. g. functional magnetic resonance imaging). However, the brain mechanisms underlying the motivational system’s action on the pursuit of concurrent goals and the associated time course of such an action is less understood. The dissociation mechanism of dual-task processing based on the reward expectation is also not known. The present study attempted to address these issues using high-density event-related potentials (ERPs) and standard tasks. We recorded ERPs in 16 healthy right-handed participants. Each block began with a black fixation cross displayed on a gray background for 1000 ms, followed by a digital item for 500 ms, and a blank interval (ISI) ranged from 1500 to 2000 ms. The test stimuli were digits pseudo-randomly chosen from the series "135791" and successively presented on a black screen within a square frame. Subjects performed backward digit-matching tasks and pressed "F" and "J" buttons for match and non-match responses. They began each block of digits by indicating whether the first digit was "1" and proceeded by indicating whether two successively presented digits were also in immediate succession in the series "135791". We referred to this task as the primary task. Triangle cues appeared at random times which instructed participants to start a secondary backward digit-matching task by either abandoning the primary task (single-task condition) or delaying its execution (dual-task condition). When the contextual cues disappeared, participants were required to abandon the secondary task, and started the primary task over again (single-task condition) or reverted back to the primary task and finished the execution (dual-task condition). In both conditions, the digits were accompanied by incentive cues indicating a reward associated with the ongoing task. The reward could be small or large and was earned only when the task was performed with no errors. At the end of the block, a visual feedback was presented indicating the monetary reward obtained from this block. One thousand ms after the offset of the feedback, the next block of digits started. Behavioral results confirmed that under both conditions, the primary and secondary rewards drove primary and secondary task performance, respectively. Furthermore, the reaction time in the secondary task was significantly longer in the dual task compared to the single task condition. ERP waveform analysis revealed that the single-task responses elicited a larger amplitude (N2) than did the dual-task responses in 290~330 ms and the dual-task responses elicited a larger amplitude (P3) than did the single-task responses in 350~800 ms. Moreover, we observed that the right hemispheric dominance drove the single-task performance according to the secondary reward during 500~700 ms in the dual-task condition. The right hemispheric dominance encoded the primary task reward driving primary task while the left and right hemispheric dominance jointly encoded the secondary task reward driving secondary task during 500~800ms. The overall findings suggest that the amount of mental resources consumed in the dual-task condition was much more than that in the single-task condition. This study provided electrophysiological evidence of dynamic hemispheric dominance of the primary task reward effect based on the character of primary task in dual-task processing.