Dissociating neural variability related to stimulus quality and response times in perceptual decision-making

Stefan Bode, Daniel Bennett, David K. Sewell, Bryan Paton, Gary F. Egan, Philip L. Smith, Carsten Murawski

Research output: Contribution to journalArticleResearchpeer-review

1 Citation (Scopus)

Abstract

According to sequential sampling models, perceptual decision-making is based on accumulation of noisy evidence towards a decision threshold. The speed with which a decision is reached is determined by both the quality of incoming sensory information and random trial-by-trial variability in the encoded stimulus representations. To investigate those decision dynamics at the neural level, participants made perceptual decisions while functional magnetic resonance imaging (fMRI) was conducted. On each trial, participants judged whether an image presented under conditions of high, medium, or low visual noise showed a piano or a chair. Higher stimulus quality (lower visual noise) was associated with increased activation in bilateral medial occipito-temporal cortex and ventral striatum. Lower stimulus quality was related to stronger activation in posterior parietal cortex (PPC) and dorsolateral prefrontal cortex (DLPFC). When stimulus quality was fixed, faster response times were associated with a positive parametric modulation of activation in medial prefrontal and orbitofrontal cortex, while slower response times were again related to more activation in PPC, DLPFC and insula. Our results suggest that distinct neural networks were sensitive to the quality of stimulus information, and to trial-to-trial variability in the encoded stimulus representations, but that reaching a decision was a consequence of their joint activity.

Original languageEnglish
Pages (from-to)190-200
Number of pages11
JournalNeuropsychologia
Volume111
DOIs
Publication statusPublished - 1 Mar 2018

Keywords

  • Decision difficulty
  • Evidence accumulation
  • Functional magnetic resonance imaging
  • Perceptual decision-making
  • Sequential sampling models

Cite this