Confidence estimation as a stochastic process in a neurodynamical system of decision making

Ziqiang Wei, Xiao-Jing Wang

Research output: Contribution to journalArticle

Abstract

Evaluation of confidence about one’s knowledge is key to the brain’s ability to monitor cognition. To investigate the neural mechanism of confidence assessment, we examined a biologically realistic spiking network model and found that it reproduced salient behavioral observations and single-neuron activity data from a monkey experiment designed to study confidence about a decision under uncertainty. Interestingly, the model predicts that changes of mind can occur in a mnemonic delay when confidence is low; the probability of changes of mind increases (decreases) with task difficulty in correct (error) trials. Furthermore, a so-called “hard-easy effect” observed in humans naturally emerges, i.e., behavior shows underconfidence (underestimation of correct rate) for easy or moderately difficult tasks and overconfidence (overestimation of correct rate) for very difficult tasks. Importantly, in the model, confidence is computed using a simple neural signal in individual trials, without explicit representation of probability functions. Therefore, even a concept of metacognition can be explained by sampling a stochastic neural activity pattern.

Original languageEnglish (US)
Pages (from-to)99-113
Number of pages15
JournalJournal of Neurophysiology
Volume114
Issue number1
DOIs
StatePublished - May 6 2015

Fingerprint

Stochastic Processes
Decision Making
Aptitude
Cognition
Uncertainty
Haplorhini
Neurons
Brain
Behavior Observation Techniques
Metacognition

Keywords

  • Decision confidence
  • Lateral intraparietal cortex
  • Line-attractor neural model

ASJC Scopus subject areas

  • Physiology
  • Neuroscience(all)

Cite this

Confidence estimation as a stochastic process in a neurodynamical system of decision making. / Wei, Ziqiang; Wang, Xiao-Jing.

In: Journal of Neurophysiology, Vol. 114, No. 1, 06.05.2015, p. 99-113.

Research output: Contribution to journalArticle

@article{822cfafb3035497993f28c7c87a5d4d1,
title = "Confidence estimation as a stochastic process in a neurodynamical system of decision making",
abstract = "Evaluation of confidence about one’s knowledge is key to the brain’s ability to monitor cognition. To investigate the neural mechanism of confidence assessment, we examined a biologically realistic spiking network model and found that it reproduced salient behavioral observations and single-neuron activity data from a monkey experiment designed to study confidence about a decision under uncertainty. Interestingly, the model predicts that changes of mind can occur in a mnemonic delay when confidence is low; the probability of changes of mind increases (decreases) with task difficulty in correct (error) trials. Furthermore, a so-called “hard-easy effect” observed in humans naturally emerges, i.e., behavior shows underconfidence (underestimation of correct rate) for easy or moderately difficult tasks and overconfidence (overestimation of correct rate) for very difficult tasks. Importantly, in the model, confidence is computed using a simple neural signal in individual trials, without explicit representation of probability functions. Therefore, even a concept of metacognition can be explained by sampling a stochastic neural activity pattern.",
keywords = "Decision confidence, Lateral intraparietal cortex, Line-attractor neural model",
author = "Ziqiang Wei and Xiao-Jing Wang",
year = "2015",
month = "5",
day = "6",
doi = "10.1152/jn.00793.2014",
language = "English (US)",
volume = "114",
pages = "99--113",
journal = "Journal of Neurophysiology",
issn = "0022-3077",
publisher = "American Physiological Society",
number = "1",

}

TY - JOUR

T1 - Confidence estimation as a stochastic process in a neurodynamical system of decision making

AU - Wei, Ziqiang

AU - Wang, Xiao-Jing

PY - 2015/5/6

Y1 - 2015/5/6

N2 - Evaluation of confidence about one’s knowledge is key to the brain’s ability to monitor cognition. To investigate the neural mechanism of confidence assessment, we examined a biologically realistic spiking network model and found that it reproduced salient behavioral observations and single-neuron activity data from a monkey experiment designed to study confidence about a decision under uncertainty. Interestingly, the model predicts that changes of mind can occur in a mnemonic delay when confidence is low; the probability of changes of mind increases (decreases) with task difficulty in correct (error) trials. Furthermore, a so-called “hard-easy effect” observed in humans naturally emerges, i.e., behavior shows underconfidence (underestimation of correct rate) for easy or moderately difficult tasks and overconfidence (overestimation of correct rate) for very difficult tasks. Importantly, in the model, confidence is computed using a simple neural signal in individual trials, without explicit representation of probability functions. Therefore, even a concept of metacognition can be explained by sampling a stochastic neural activity pattern.

AB - Evaluation of confidence about one’s knowledge is key to the brain’s ability to monitor cognition. To investigate the neural mechanism of confidence assessment, we examined a biologically realistic spiking network model and found that it reproduced salient behavioral observations and single-neuron activity data from a monkey experiment designed to study confidence about a decision under uncertainty. Interestingly, the model predicts that changes of mind can occur in a mnemonic delay when confidence is low; the probability of changes of mind increases (decreases) with task difficulty in correct (error) trials. Furthermore, a so-called “hard-easy effect” observed in humans naturally emerges, i.e., behavior shows underconfidence (underestimation of correct rate) for easy or moderately difficult tasks and overconfidence (overestimation of correct rate) for very difficult tasks. Importantly, in the model, confidence is computed using a simple neural signal in individual trials, without explicit representation of probability functions. Therefore, even a concept of metacognition can be explained by sampling a stochastic neural activity pattern.

KW - Decision confidence

KW - Lateral intraparietal cortex

KW - Line-attractor neural model

UR - http://www.scopus.com/inward/record.url?scp=84939794227&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84939794227&partnerID=8YFLogxK

U2 - 10.1152/jn.00793.2014

DO - 10.1152/jn.00793.2014

M3 - Article

VL - 114

SP - 99

EP - 113

JO - Journal of Neurophysiology

JF - Journal of Neurophysiology

SN - 0022-3077

IS - 1

ER -