Electrophysiological indices of anterior cingulate cortex function reveal changing levels of cognitive effort and reward valuation that sustain task performance. (4th February 2019)

• Record Type: Journal Article
• Authors: Umemoto, Akina; Inzlicht, Michael; Holroyd, Clay B.
• Is Part Of: Neuropsychologia. Volume 123(2019)
• Page Start: 67
• Page End: 76
• Publication Date: 2019-02-04
• Subjects: Sustained effort -- Reward valuation -- Cognitive control -- Anterior cingulate cortex -- Electrophysiology; Neuropsychology -- Periodicals; Neurology -- Periodicals; Psychophysiology -- Periodicals; Neuropsychologie -- Périodiques; Neuropsychology; Periodicals; Electronic journals; 616.8
• DOI: 10.1016/j.neuropsychologia.2018.06.010
• Languages: English
• ISSNs: 0028-3932

Abstract

Abstract: Successful execution of goal-directed behaviors often requires the deployment of cognitive control, which is thought to require cognitive effort. Recent theories have proposed that anterior cingulate cortex (ACC) regulates control levels by weighing the reward-related benefits of control against its effort-related costs. However, given that the sensations of cognitive effort and reward valuation are available only to introspection, this hypothesis is difficult to investigate empirically. We have proposed that two electrophysiological indices of ACC function, frontal midline theta and the reward positivity (RewP), provide objective measures of these functions. To explore this issue, we recorded the electroencephalogram (EEG) from participants engaged in an extended, cognitively-demanding task. Participants performed a time estimation task for 2 h in which they received reward and error feedback according to their task performance. We observed that the amplitude of the RewP, a feedback-locked component of the event related brain potential associated with reward processing, decreased with time-on-task. Conversely, frontal midline theta power, which consists of 4–8 Hz EEG oscillations associated with cognitive effort, increased with time-on-task. We also explored how these phenomena changed over time by conducting within-participant multi-level modeling analyses. Our results suggest that extended execution of a cognitively-demanding task is characterized by an early phase in which high control levels foster rapid improvements in task performance, and a later phase in which high control levels were necessary to maintain stable task performance, perhaps counteracting waning reward valuation. Highlights: Participants performed a time estimation task for 2 h. Reward valuation (RewP) and task engagement (P300) declined with time on task. Frontal midline theta power, reflecting cognitive effort, increased with time one task. Within-person variations among behavior and electrophysiological measures were tested. High control levels fostered learning and maintained stable task performance.