Confidence matching in group decision-making

Bang, Dan; Aitchison, Laurence; Moran, Rani; Castañón, Santiago Herce; Rafiee, Banafsheh; Mahmoodi, Ali; Lau, Jennifer Y. F.; Latham, Peter E.; Bahrami, Bahador; Summerfield, Christopher

doi:10.1038/s41562-017-0117

Cited by 111 publications

(148 citation statements)

References 33 publications

Supporting

Mentioning

130

Contrasting

Order By: Relevance

“…However, after committing to a choice, it may be more efficient to combine these representations into a statistic that summarises the probability that a choice is correct (1). This summary statistic, decision confidence, can be used to guide multi-stage decisions (50), help optimise learning from feedback (51,52) and be shared with others to make group decisions (4,5). Our study supports a distinction between choice-independent and choice-dependent estimates of certainty (1), and indicate a dissociation in the neural encoding of these quantities in perceptual decision-making.…”

Section: Cc-by-nc-nd 40 International License Not Peer-reviewed) Is supporting

confidence: 67%

“…S9). More broadly, there is evidence that rlPFC manages task sets and rules (39)(40)(41)(42)(43), functions which presumably are involved in maintaining a consistent mapping from internal to explicit confidence estimates, or updating a mapping in response to a particular communicative context (5). Future studies directly manipulating confidence mappings are required to test this hypothesis.…”

Section: Cc-by-nc-nd 40 International License Not Peer-reviewed) Is mentioning

confidence: 99%

“…For example, after having made a difficult choice, an animal might correctly estimate that its decision is unlikely to be correct and thus avoid wasting time waiting for a reward that may not arrive (3). Humans may communicate such estimates of decision confidence to make more accurate decisions together than alone (4,5). Despite widespread agreement that decision confidence is a useful quantity for adaptive control of behaviour, neurobiological support for a distinct encoding of decision confidence is lacking.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Distinct encoding of decision confidence in human medial prefrontal cortex

Bang

Fleming

2018

Preprint

Self Cite

View full text Add to dashboard Cite

Our confidence in a choice and the reliability of evidence pertaining to that choice appear to be inseparable quantities. However, recent computational frameworks suggest that the brain should encode distinct estimates of these quantities for adaptive behavioural control. Consider a line judge in a tennis match deciding whether a ball is out. When the ball is moving quickly, the sensory signal is unreliable. Nevertheless, if the ball lands far from the line, a decision may be easy and confidence high. Decision confidence should reflect the probability that a choice is correct and must therefore, unlike other estimates of certainty such as evidence reliability, be flexibly conditioned on a choice. Here we devised a psychophysical approach to decouple confidence in a sensory choice from the reliability of the sensory evidence upon which a choice is based. Using human fMRI, we observed that an area in medial prefrontal cortex (perigenual cingulate cortex, pgACC) tracked probability correct, the normative definition of decision confidence, whereas neural areas previously thought to track decision confidence, such as ventral striatum and posterior parietal cortex, instead tracked sensory reliability. We show that pgACC activity does not simply co-vary with objective performance but is also linked to withinsubject and between-subject variation in a subjective sense of decision confidence. Our study is consistent with the recent proposal that the brain separately maintains choice-dependent and choice-independent estimates of certainty, and sheds light on why dysfunctional confidence estimates often emerge following prefrontal lesions and/or degeneration.

show abstract

Section: Cc-by-nc-nd 40 International License Not Peer-reviewed) Is supporting

confidence: 67%

Section: Cc-by-nc-nd 40 International License Not Peer-reviewed) Is mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Distinct encoding of decision confidence in human medial prefrontal cortex

Bang

Fleming

2018

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Relatedly, Pescetelli and colleagues found that a dyad's joint performance is positively correlated with how high its individual members’ accuracy correlated with their confidence ratings (i.e., how well their confidence judgments were calibrated relative to their actual performance). Finally, a recent study further investigated the role of confidence ratings, revealing that dyads aim to align their relationships between confidence ratings and performance (i.e., “confidence matching”). That is, members of a dyad aim to express their certainty (or lack thereof) to each other in a similar manner.…”

Section: Joint Perceptual Decision‐making Tasksmentioning

confidence: 99%

Group benefits in joint perceptual tasks—a review

Wahn

Kingstone

König

2018

Annals of the New York Academy of Sciences

View full text Add to dashboard Cite

In daily life, humans often perform perceptual tasks together to reach a shared goal. In these situations, individuals may collaborate (e.g., by distributing task demands) to perform the task better than when the task is performed alone (i.e., attain a group benefit). In this review, we identify the factors influencing if, and to what extent, a group benefit is attained and provide a framework of measures to assess group benefits in perceptual tasks. In particular, we integrate findings from two frequently investigated joint perceptual tasks: visuospatial tasks and decision-making tasks. For both task types, we find that an exchange of information between coactors is critical to improve joint performance. Yet, the type of exchanged information and how coactors collaborate differs between tasks. In visuospatial tasks, coactors exchange information about the performed actions to distribute task demands. In perceptual decision-making tasks, coactors exchange their confidence on their individual perceptual judgments to negotiate a joint decision. We argue that these differences can be explained by the task structure: coactors distribute task demands if a joint task allows for a spatial division and stimuli can be accurately processed by one individual. Otherwise, they perform the task individually and then integrate their individual judgments.

show abstract

“…Given that most important decisions are uncertain, one method of information aggregation which is particularly important is the sharing and calibration of confidence between group members (15). Confidence sharing has been shown to improve the accuracy of judgments made by dyads (16,17); to introduce sub-optimalities from pro-social biases into judgment (18); to influence collective learning (19); and to introduce systematic errors in group judgment (18,20). This research shows that confidence sharing is a complex issue; at times facilitating and at times inhibiting group decisions.…”

mentioning

confidence: 99%

Informational Openness Enhances Decentralized Decision-Making: A Cognitive Agent Based Study

Skewes¹,

Håkonsson²,

Bilde³

et al. 2017

Preprint

View full text Add to dashboard Cite

Collaborative decision making is central to the organization of society. Juries deliberate cases, voters elect government officials, open innovation networks converge on innovative solutions. It is common to think of such groups as decision making entities. But this language is imprecise. Real decision processes do not occur within any group or organization as an abstract entity. Collaborative decision making happens within and between autonomous individuals. This emphasizes the importance of the relationships between individual and social decision-making processes to social organization. Despite a rich body of literature on collaborative decision making, we know little about how individuals decide to commit to decision making groups in the first place; and how, once joined, they communicate their distributed information for optimal group performance. Using a synthesis of cognitive models and agent-based simulation, we investigate the influence of information flow on commitment and performance in open collaborative decision making groups. The main results of the simulations are that 1) commitment is enhanced when groups are designed so agents have realistic knowledge about the forgone gains and losses associated with choosing to join a decision making group; and 2) that collaborative decision making groups gain most when communication between members conveys more information about individual's preferences. We thus demonstrate that collaborative decision making is done best when it is done by groups that are informationally open.

show abstract

Confidence matching in group decision-making

Cited by 111 publications

References 33 publications

Distinct encoding of decision confidence in human medial prefrontal cortex

Distinct encoding of decision confidence in human medial prefrontal cortex

Group benefits in joint perceptual tasks—a review

Informational Openness Enhances Decentralized Decision-Making: A Cognitive Agent Based Study

Contact Info

Product

Resources

About