Citation

Säfström D, Domellöf E. Neuroimage 2018; 172: 608-619.

Affiliation

Department of Psychology, Umeå University, 901 87 Umeå, Sweden; Umeå Center for Functional Brain Imaging (UFBI), Umeå University, 901 87 Umeå, Sweden. Electronic address: erik.domellof@umu.se.

Copyright

(Copyright © 2018, Elsevier Publishing)

DOI

10.1016/j.neuroimage.2018.02.014

PMID

29428579

Abstract

Most everyday manual tasks, like grabbing a cup of coffee to drink, are comprised of a sequence of action phases. Efficient phase transitions, or linking, are achieved using a predictive control policy where motor commands for the next phase are specified and released in anticipation of sensory confirmation of goal completion of the current phase. If there is a discrepancy between predicted and actual sensory feedback about goal completion, corrective actions are employed to complete the current action phase before proceeding to the next. However, we lack understanding about brain activations supporting such predictive linking and corrective actions in manual tasks. In this study, during 3-T MRI-scanning, sixteen participants (5 males, 11 females; mean age 27.3 years, range 23-37) performed a sequential manual task, with or without the possibility for predictive linking. We found that predictive linking of action phases was associated with increased activation in a network that included right-sided fronto-parietal areas related to visuospatial attention, eye movements and motor planning, left-sided parietal areas related to implicit timing and shifts of motor attention, occipital regions bilaterally reflecting visual processing related to the attended next target, and finally, the anterior midcingulate cortex involved in continuous performance monitoring. Corrective actions were associated with increased activation in the left dorsolateral prefrontal cortex involved in reestablishing executive control over previously automatized behavior.

Copyright © 2018. Published by Elsevier Inc.

Language: en

Keywords

Error detection; Motor prediction; Sensorimotor control; Sequential actions; fMRI