A role for proactive control in rapid instructed task learning

Michael W. Cole, Lauren M. Patrick, Nachshon Meiran, Todd S. Braver

Research output: Contribution to journalArticlepeer-review

8 Scopus citations

Abstract

Humans are often remarkably fast at learning novel tasks from instructions. Such rapid instructed task learning (RITL) likely depends upon the formation of new associations between long-term memory representations, which must then be actively maintained to enable successful task implementation. Consequently, we hypothesized that RITL relies more heavily on a proactive mode of cognitive control, in which goal-relevant information is actively maintained in preparation for anticipated high control demands. We tested this hypothesis using a recently developed cognitive paradigm consisting of 60 novel tasks involving RITL and 4 practiced tasks, with identical task rules and stimuli used across both task types. A robust behavioral cost was found in novel relative to practiced task performance, which was present even when the two were randomly inter-mixed, such that task-switching effects were equated. Novelty costs were most prominent under time-limited preparation conditions. In self-paced conditions, increased preparation time was found for novel trials, and was selectively associated with enhanced performance, suggesting greater proactive control for novel tasks. These results suggest a key role for proactive cognitive control in the ability to rapidly learn novel tasks from instructions.

Original languageEnglish
Pages (from-to)20-30
Number of pages11
JournalActa Psychologica
Volume184
DOIs
StatePublished - 1 Mar 2018

Keywords

  • Cognitive control
  • Proactive control
  • Rapid instructed task learning

Fingerprint

Dive into the research topics of 'A role for proactive control in rapid instructed task learning'. Together they form a unique fingerprint.

Cite this