Probabilistic Population Codes for Bayesian Decision Making

Jeffrey M. Beck, Wei Ji Ma, Roozbeh Kiani, Tim Hanks, Anne K. Churchland, Jamie Roitman, Michael N. Shadlen, Peter E. Latham, Alexandre Pouget

Research output: Contribution to journalArticle

Abstract

When making a decision, one must first accumulate evidence, often over time, and then select the appropriate action. Here, we present a neural model of decision making that can perform both evidence accumulation and action selection optimally. More specifically, we show that, given a Poisson-like distribution of spike counts, biological neural networks can accumulate evidence without loss of information through linear integration of neural activity and can select the most likely action through attractor dynamics. This holds for arbitrary correlations, any tuning curves, continuous and discrete variables, and sensory evidence whose reliability varies over time. Our model predicts that the neurons in the lateral intraparietal cortex involved in evidence accumulation encode, on every trial, a probability distribution which predicts the animal's performance. We present experimental evidence consistent with this prediction and discuss other predictions applicable to more general settings.

Original languageEnglish (US)
Pages (from-to)1142-1152
Number of pages11
JournalNeuron
Volume60
Issue number6
DOIs
StatePublished - Dec 26 2008

Fingerprint

Decision Making
Poisson Distribution
Population
Neurons

Keywords

  • SYSNEURO

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

Beck, J. M., Ma, W. J., Kiani, R., Hanks, T., Churchland, A. K., Roitman, J., ... Pouget, A. (2008). Probabilistic Population Codes for Bayesian Decision Making. Neuron, 60(6), 1142-1152. https://doi.org/10.1016/j.neuron.2008.09.021

Probabilistic Population Codes for Bayesian Decision Making. / Beck, Jeffrey M.; Ma, Wei Ji; Kiani, Roozbeh; Hanks, Tim; Churchland, Anne K.; Roitman, Jamie; Shadlen, Michael N.; Latham, Peter E.; Pouget, Alexandre.

In: Neuron, Vol. 60, No. 6, 26.12.2008, p. 1142-1152.

Research output: Contribution to journalArticle

Beck, JM, Ma, WJ, Kiani, R, Hanks, T, Churchland, AK, Roitman, J, Shadlen, MN, Latham, PE & Pouget, A 2008, 'Probabilistic Population Codes for Bayesian Decision Making', Neuron, vol. 60, no. 6, pp. 1142-1152. https://doi.org/10.1016/j.neuron.2008.09.021
Beck, Jeffrey M. ; Ma, Wei Ji ; Kiani, Roozbeh ; Hanks, Tim ; Churchland, Anne K. ; Roitman, Jamie ; Shadlen, Michael N. ; Latham, Peter E. ; Pouget, Alexandre. / Probabilistic Population Codes for Bayesian Decision Making. In: Neuron. 2008 ; Vol. 60, No. 6. pp. 1142-1152.
@article{58af0979bd374848b18be10eb4dbd552,
title = "Probabilistic Population Codes for Bayesian Decision Making",
abstract = "When making a decision, one must first accumulate evidence, often over time, and then select the appropriate action. Here, we present a neural model of decision making that can perform both evidence accumulation and action selection optimally. More specifically, we show that, given a Poisson-like distribution of spike counts, biological neural networks can accumulate evidence without loss of information through linear integration of neural activity and can select the most likely action through attractor dynamics. This holds for arbitrary correlations, any tuning curves, continuous and discrete variables, and sensory evidence whose reliability varies over time. Our model predicts that the neurons in the lateral intraparietal cortex involved in evidence accumulation encode, on every trial, a probability distribution which predicts the animal's performance. We present experimental evidence consistent with this prediction and discuss other predictions applicable to more general settings.",
keywords = "SYSNEURO",
author = "Beck, {Jeffrey M.} and Ma, {Wei Ji} and Roozbeh Kiani and Tim Hanks and Churchland, {Anne K.} and Jamie Roitman and Shadlen, {Michael N.} and Latham, {Peter E.} and Alexandre Pouget",
year = "2008",
month = "12",
day = "26",
doi = "10.1016/j.neuron.2008.09.021",
language = "English (US)",
volume = "60",
pages = "1142--1152",
journal = "Neuron",
issn = "0896-6273",
publisher = "Cell Press",
number = "6",

}

TY - JOUR

T1 - Probabilistic Population Codes for Bayesian Decision Making

AU - Beck, Jeffrey M.

AU - Ma, Wei Ji

AU - Kiani, Roozbeh

AU - Hanks, Tim

AU - Churchland, Anne K.

AU - Roitman, Jamie

AU - Shadlen, Michael N.

AU - Latham, Peter E.

AU - Pouget, Alexandre

PY - 2008/12/26

Y1 - 2008/12/26

N2 - When making a decision, one must first accumulate evidence, often over time, and then select the appropriate action. Here, we present a neural model of decision making that can perform both evidence accumulation and action selection optimally. More specifically, we show that, given a Poisson-like distribution of spike counts, biological neural networks can accumulate evidence without loss of information through linear integration of neural activity and can select the most likely action through attractor dynamics. This holds for arbitrary correlations, any tuning curves, continuous and discrete variables, and sensory evidence whose reliability varies over time. Our model predicts that the neurons in the lateral intraparietal cortex involved in evidence accumulation encode, on every trial, a probability distribution which predicts the animal's performance. We present experimental evidence consistent with this prediction and discuss other predictions applicable to more general settings.

AB - When making a decision, one must first accumulate evidence, often over time, and then select the appropriate action. Here, we present a neural model of decision making that can perform both evidence accumulation and action selection optimally. More specifically, we show that, given a Poisson-like distribution of spike counts, biological neural networks can accumulate evidence without loss of information through linear integration of neural activity and can select the most likely action through attractor dynamics. This holds for arbitrary correlations, any tuning curves, continuous and discrete variables, and sensory evidence whose reliability varies over time. Our model predicts that the neurons in the lateral intraparietal cortex involved in evidence accumulation encode, on every trial, a probability distribution which predicts the animal's performance. We present experimental evidence consistent with this prediction and discuss other predictions applicable to more general settings.

KW - SYSNEURO

UR - http://www.scopus.com/inward/record.url?scp=57649192772&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=57649192772&partnerID=8YFLogxK

U2 - 10.1016/j.neuron.2008.09.021

DO - 10.1016/j.neuron.2008.09.021

M3 - Article

VL - 60

SP - 1142

EP - 1152

JO - Neuron

JF - Neuron

SN - 0896-6273

IS - 6

ER -