Doubly Bayesian Analysis of Confidence in Perceptual Decision-Making

Aitchison, Laurence; Bang, Dan; Bahrami, Bahador; Latham, Peter E.

doi:10.1371/journal.pcbi.1004519

Cited by 129 publications

(197 citation statements)

References 35 publications

(40 reference statements)

Supporting

Mentioning

181

Contrasting

Order By: Relevance

“…Confidence correlate of model-based and model-free strategies According to Folke et al (2016)Folke et al (2017, confidence can be strongly predicted by the difference of action values; the same is reported in perceptual decisionmaking where the difference in evidence is correlated with confidence Aitchison et al (2015). We exploited the same concept and checked if model-based and model-free methods differ in estimating the difference of Q-values for first-step choices.…”

Section: Figurementioning

confidence: 92%

Higher Meta-cognitive Ability Predicts Less Reliance on Over Confident Habitual Learning System

Ershadmanesh

Miandari

Vahabie

et al. 2019

Preprint

View full text Add to dashboard Cite

Many studies on human and animals have provided evidence for the contribution of goal-directed and habitual valuation systems in learning and decision-making. These two systems can be modeled using model-based (MB) and model-free (MF) algorithms in Reinforcement Learning (RL) framework.Here, we study the link between the contribution of these two learning systems to behavior and meta-cognitive capabilities. Using computational modeling we showed that in a highly variable environment, where both learning strategies have chance level performances, model-free learning predicts higher confidence in decisions compared to model-based strategy. Our experimental results showed that the subjects' meta-cognitive ability is negatively correlated with the contribution of model-free system to their behavior while having no correlation with the contribution of model-based system. Over-confidence of the model-free system justifies this counter-intuitive result. This is a new explanation for individual difference in learning style. * Corresponding author (saraershad@ipm.ir) † These two authors contributed equally

show abstract

Section: Figurementioning

confidence: 92%

Higher Meta-cognitive Ability Predicts Less Reliance on Over Confident Habitual Learning System

Ershadmanesh

Miandari

Vahabie

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…This definition has a long history in psychophysics 7,12,13 and has been recently used in several studies 14–20 . This is also what many authors call confidence 21–27 , even if they don’t always formally define it as such.…”

Section: Confidence: Definition and Computationsmentioning

confidence: 99%

Confidence and certainty: distinct probabilistic quantities for different goals

2016

View full text Add to dashboard Cite

When facing uncertainty, adaptive behavioral strategies demand that the brain performs probabilistic computations. In this probabilistic framework, the notion of certainty and confidence would appear to be closely related, so much so that it is tempting to conclude that these two concepts are one and the same. We argue that there are computational reasons to distinguish between these two concepts. Specifically, we propose that confidence should be defined as the probability that a decision or a proposition, overt or covert, is correct given the evidence, a critical quantity in complex sequential decisions. We suggest that the term certainty should be reserved to refer to the encoding of all other probability distributions over sensory and cognitive variables. We also discuss strategies for studying the neural codes for confidence and certainty and argue that clear definitions of neural codes are essential to understanding the relative contributions of various cortical areas to decision making.

show abstract

“…Seemingly suboptimal behaviors have also been observed in post-decisional perceptual judgments other than confidence (Stocker and Simoncelli, 2008;Luu and Stocker, 2018), leading these authors to hypothesize that these suboptimalities may stem from limitations on computational (i.e., neural) resources or a drive towards self-consistent behavior. One alternative theory of confidence, therefore, proposes that subjective confidence relies primarily on the magnitude of evidence supporting an observer's decision, while ignoring or downplaying evidence supporting alternative, unchosen decisions (Zylberberg et al, 2012;Aitchison et al, 2015;Koizumi et al, 2015;Maniscalco et al, 2016; drive decisions and less normalized accumulator neurons drive confidence may be sufficient to account for some of the most counterintuitive empirical findings on confidence in perceptual decision-making.…”

Section: Significancementioning

confidence: 99%

Tuned normalization in perceptual decision-making circuits can explain seemingly suboptimal confidence behavior

Maniscalco

Odegaard

Grimaldi

et al. 2019

Preprint

View full text Add to dashboard Cite

of the model; M.A.K.P. and B.M. formally specified the model and conducted model simulations; P.G., S.H.C., and M.A.B. collected the monkey data; M.A.K.P., B.O., P.G. and S.H.C. analyzed the monkey data; M.A.K.P., B.M., and H.L. wrote the manuscript. AbstractCurrent dominant views hold that perceptual confidence reflects the probability that a decision is correct. Although these views have enjoyed some empirical support, recent behavioral results indicate that confidence and the probability of being correct can be dissociated. An alternative hypothesis suggests that confidence instead reflects the magnitude of evidence in favor of a decision while being relatively insensitive to the evidence opposing the decision. We considered how this alternative hypothesis might be biologically instantiated by developing a simple leaky competing accumulator neural network model incorporating a known property of sensory neurons: tuned normalization. The key idea of the model is that each accumulator neuron's normalization 'tuning' dictates its contribution to perceptual decisions versus confidence judgments. We demonstrate that this biologically plausible model can account for several counterintuitive findings reported in the literature, where confidence and decision accuracy were shown to dissociate --and that the differential contribution a neuron makes to decisions versus confidence judgments based on its normalization tuning is vital to capturing some of these effects. One critical prediction of the model is that systematic variability in normalization tuning exists not only in sensory cortices but also in the decision-making circuitry. We tested and validated this prediction in macaque superior colliculus (SC; a region implicated in decisionmaking). The confirmation of this novel prediction provides direct support for our model. These findings suggest that the brain has developed and implements this alternative, heuristic theory of perceptual confidence computation by capitalizing on the diversity of neural resources available. Samaha et al., 2016Samaha et al., , 2017. In other words, to compute confidence the system uses a suboptimal heuristic that overly relies on decision-congruent evidence magnitude rather than optimal computations. Indeed, a recent study reported evidence for these decision-congruent evidence confidence computations using human intracranial electrocorticography (Peters et al., 2017b).However, to date no biologically plausible mechanism has been proposed that might explain these dissociations between confidence and performance, or the decision-congruent confidence computations on which they seem to depend. We therefore developed a simple leaky competing accumulator network model (Usher and McClelland, 2001) to test a new hypothesis of how these computations might be implemented. This model extends previous work to incorporate a known property of perceptual circuitry: tuned normalization Ruff et al., 2016;Verhoef and Maunsell, 2017), meaning each neuron is characterized by the specific degree to which i...

show abstract

Doubly Bayesian Analysis of Confidence in Perceptual Decision-Making

Cited by 129 publications

References 35 publications

Higher Meta-cognitive Ability Predicts Less Reliance on Over Confident Habitual Learning System

Higher Meta-cognitive Ability Predicts Less Reliance on Over Confident Habitual Learning System

Confidence and certainty: distinct probabilistic quantities for different goals

Tuned normalization in perceptual decision-making circuits can explain seemingly suboptimal confidence behavior

Contact Info

Product

Resources

About