Flexing dual-systems models: How variable cognitive control in children informs our understanding of risk-taking across development.

Journal Article (Review;Journal Article)

Prevailing models of the development of decision-making propose that peak risk-taking occurs in adolescence due to a neural imbalance between two processes: gradual, linearly developing cognitive control and rapid, non-linearly developing reward-processing. Though many studies have found neural evidence supporting this dual-systems imbalance model, its behavioral predictions have been surprisingly difficult to document. Most laboratory studies have not found adolescents to exhibit greater risk-taking than children, and public health data show everyday risk-taking to peak in late adolescence/early adulthood. Moreover, when adolescents are provided detailed information about decision options and consequences, they evince similar behavior to adults. Such findings point to a critical feature of the development of decision-making that is missed by imbalance models. Specifically, the engagement of cognitive control is context dependent, such that cognitive control and therefore advantageous decision-making increases when available information is high and decreases when available information is low. Furthermore, the context dependence of cognitive control varies across development, such that increased information availability benefits children more than adolescents, who benefit more than adults. This review advances a flexible dual-systems model that is only imbalanced under certain conditions; explains disparities between neural, behavioral, and public health findings; and provides testable hypotheses for future research.

Full Text

Duke Authors

Cited Authors

  • Li, R

Published Date

  • October 2017

Published In

Volume / Issue

  • 27 /

Start / End Page

  • 91 - 98

PubMed ID

  • 28869843

Pubmed Central ID

  • PMC6987948

Electronic International Standard Serial Number (EISSN)

  • 1878-9307

International Standard Serial Number (ISSN)

  • 1878-9293

Digital Object Identifier (DOI)

  • 10.1016/j.dcn.2017.08.007


  • eng