Human online adaptation to changes in prior probability

Elyse H. Norton, Luigi Acerbi, Wei Ji Ma, Michael Landy

Research output: Contribution to journalArticle

Abstract

Optimal sensory decision-making requires the combination of uncertain sensory signals with prior expectations. The effect of prior probability is often described as a shift in the decision criterion. Can observers track sudden changes in probability? To answer this question, we used a change-point detection paradigm that is frequently used to examine behavior in changing environments. In a pair of orientation-categorization tasks, we investigated the effects of changing probabilities on decision-making. In both tasks, category probability was updated using a sample-and-hold procedure: probability was held constant for a period of time before jumping to another probability state that was randomly selected from a predetermined set of probability states. We developed an ideal Bayesian change-point detection model in which the observer marginalizes over both the current run length (i.e., time since last change) and the current category probability. We compared this model to various alternative models that correspond to different strategies—from approximately Bayesian to simple heuristics—that the observers may have adopted to update their beliefs about probabilities. While a number of models provided decent fits to the data, model comparison favored a model in which probability is estimated following an exponential averaging model with a bias towards equal priors, consistent with a conservative bias, and a flexible variant of the Bayesian change-point detection model with incorrect beliefs. We interpret the former as a simpler, more biologically plausible explanation suggesting that the mechanism underlying change of decision criterion is a combination of on-line estimation of prior probability and a stable, long-term equal-probability prior, thus operating at two very different timescales.

Original languageEnglish (US)
Article numbere1006681
JournalPLoS computational biology
Volume15
Issue number7
DOIs
StatePublished - Jul 1 2019

Fingerprint

Prior Probability
Change-point Detection
Observer
Decision Making
Model
Model Averaging
Human
decision making
Model Comparison
Run Length
Decision making
Categorization
Period of time
Data Model
jumping
Time Scales
Update
Paradigm
Data structures
Alternatives

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Modeling and Simulation
  • Ecology
  • Molecular Biology
  • Genetics
  • Cellular and Molecular Neuroscience
  • Computational Theory and Mathematics

Cite this

Human online adaptation to changes in prior probability. / Norton, Elyse H.; Acerbi, Luigi; Ma, Wei Ji; Landy, Michael.

In: PLoS computational biology, Vol. 15, No. 7, e1006681, 01.07.2019.

Research output: Contribution to journalArticle

@article{3e73420d20b0465e8cc1ae2940c83bbb,
title = "Human online adaptation to changes in prior probability",
abstract = "Optimal sensory decision-making requires the combination of uncertain sensory signals with prior expectations. The effect of prior probability is often described as a shift in the decision criterion. Can observers track sudden changes in probability? To answer this question, we used a change-point detection paradigm that is frequently used to examine behavior in changing environments. In a pair of orientation-categorization tasks, we investigated the effects of changing probabilities on decision-making. In both tasks, category probability was updated using a sample-and-hold procedure: probability was held constant for a period of time before jumping to another probability state that was randomly selected from a predetermined set of probability states. We developed an ideal Bayesian change-point detection model in which the observer marginalizes over both the current run length (i.e., time since last change) and the current category probability. We compared this model to various alternative models that correspond to different strategies—from approximately Bayesian to simple heuristics—that the observers may have adopted to update their beliefs about probabilities. While a number of models provided decent fits to the data, model comparison favored a model in which probability is estimated following an exponential averaging model with a bias towards equal priors, consistent with a conservative bias, and a flexible variant of the Bayesian change-point detection model with incorrect beliefs. We interpret the former as a simpler, more biologically plausible explanation suggesting that the mechanism underlying change of decision criterion is a combination of on-line estimation of prior probability and a stable, long-term equal-probability prior, thus operating at two very different timescales.",
author = "Norton, {Elyse H.} and Luigi Acerbi and Ma, {Wei Ji} and Michael Landy",
year = "2019",
month = "7",
day = "1",
doi = "10.1371/journal.pcbi.1006681",
language = "English (US)",
volume = "15",
journal = "PLoS Computational Biology",
issn = "1553-734X",
publisher = "Public Library of Science",
number = "7",

}

TY - JOUR

T1 - Human online adaptation to changes in prior probability

AU - Norton, Elyse H.

AU - Acerbi, Luigi

AU - Ma, Wei Ji

AU - Landy, Michael

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Optimal sensory decision-making requires the combination of uncertain sensory signals with prior expectations. The effect of prior probability is often described as a shift in the decision criterion. Can observers track sudden changes in probability? To answer this question, we used a change-point detection paradigm that is frequently used to examine behavior in changing environments. In a pair of orientation-categorization tasks, we investigated the effects of changing probabilities on decision-making. In both tasks, category probability was updated using a sample-and-hold procedure: probability was held constant for a period of time before jumping to another probability state that was randomly selected from a predetermined set of probability states. We developed an ideal Bayesian change-point detection model in which the observer marginalizes over both the current run length (i.e., time since last change) and the current category probability. We compared this model to various alternative models that correspond to different strategies—from approximately Bayesian to simple heuristics—that the observers may have adopted to update their beliefs about probabilities. While a number of models provided decent fits to the data, model comparison favored a model in which probability is estimated following an exponential averaging model with a bias towards equal priors, consistent with a conservative bias, and a flexible variant of the Bayesian change-point detection model with incorrect beliefs. We interpret the former as a simpler, more biologically plausible explanation suggesting that the mechanism underlying change of decision criterion is a combination of on-line estimation of prior probability and a stable, long-term equal-probability prior, thus operating at two very different timescales.

AB - Optimal sensory decision-making requires the combination of uncertain sensory signals with prior expectations. The effect of prior probability is often described as a shift in the decision criterion. Can observers track sudden changes in probability? To answer this question, we used a change-point detection paradigm that is frequently used to examine behavior in changing environments. In a pair of orientation-categorization tasks, we investigated the effects of changing probabilities on decision-making. In both tasks, category probability was updated using a sample-and-hold procedure: probability was held constant for a period of time before jumping to another probability state that was randomly selected from a predetermined set of probability states. We developed an ideal Bayesian change-point detection model in which the observer marginalizes over both the current run length (i.e., time since last change) and the current category probability. We compared this model to various alternative models that correspond to different strategies—from approximately Bayesian to simple heuristics—that the observers may have adopted to update their beliefs about probabilities. While a number of models provided decent fits to the data, model comparison favored a model in which probability is estimated following an exponential averaging model with a bias towards equal priors, consistent with a conservative bias, and a flexible variant of the Bayesian change-point detection model with incorrect beliefs. We interpret the former as a simpler, more biologically plausible explanation suggesting that the mechanism underlying change of decision criterion is a combination of on-line estimation of prior probability and a stable, long-term equal-probability prior, thus operating at two very different timescales.

UR - http://www.scopus.com/inward/record.url?scp=85070116720&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070116720&partnerID=8YFLogxK

U2 - 10.1371/journal.pcbi.1006681

DO - 10.1371/journal.pcbi.1006681

M3 - Article

C2 - 31283765

AN - SCOPUS:85070116720

VL - 15

JO - PLoS Computational Biology

JF - PLoS Computational Biology

SN - 1553-734X

IS - 7

M1 - e1006681

ER -