The role of decision confidence in advice-taking and trust formation.

Abstract

In a world where ideas flow freely across multiple platforms, people must often rely on others' advice and opinions without an objective standard to judge whether this information is accurate. The present study explores the hypothesis that an individual's internal decision confidence can be used as a signal to learn the accuracy of others' advice, even in the absence of feedback. According to this "agreement-inconfidence" hypothesis, people can learn about an advisor's accuracy across multiple interactions according to whether the advice offered agrees with their own initial opinions, weighted by the confidence with which these initial opinions are held. We test this hypothesis using a judge-advisor system paradigm to precisely manipulate the profiles of virtual advisors in a perceptual decision-making task. We find that when advisors' and participants' judgments are independent, people can correctly learn advisors' features, like their accuracy and calibration, whether or not objective feedback is available. However, when their judgments (and thus errors) are correlated-as is the case in many real social contexts-predictable distortions in trust can be observed between feedback and feedback-free scenarios. Using agent-based simulations, we explore implications of these individual-level heuristics for networklevel patterns of trust and belief formation.

Metacognitive Computations for Information Search: Confidence in Control

“…In the real world, information is often not solely provided by faceless sources, but by other agents with their own intentions. Over and above just being noisy (and indeed nosey), such social sources might have their own biases and interests of which successful agents need to be aware when evaluating whether they should invest in hearing their opinion and using them to inform themselves (Hütter & Ache, 2016;Pescetelli & Yeung, 2020;van der Plas, David, & Fleming, 2019). This is a particular pressing issue when faced with mis-and dis-information (Lazer et al, 2018;Pennycook & Rand, 2020).…”

Susceptibility of agency judgments to social influence

“…However, conventional power calculations are difficult to perform for these models because of the multiple sources of variance that must be taken into account (Westfall et al, 2014). In the absence of existing data with regard to our research goal, sample size was determined a priori based on previous studies on social information use (Olsen et al, 2019;Pescetelli & Yeung, 2020), and constrained by participants availability. Given these elements, we aimed for a target sample size of 40 participants.…”

“…First, modern theoretical frameworks of SoA emphasize that a sense of agency results from the optimal integration of information from multiple sources, including (i) internal information related to the preparation and control of action, and (ii) external information related to the context of the action (e.g., social information) (Gallagher, 2012;Synofzik et al, 2013). Moreover, a significant number of studies has shown that human subjects show some sensitivity to social information, and occasionally modify their decisions or behaviours to match those of others (Jacquet et al, 2018(Jacquet et al, , 2019Olsen et al, 2019;Pescetelli et al, 2021;Pescetelli & Yeung, 2020). In principle, the use of social information allows individuals to benefit from solutions that have already been tried out by their peers, and is driven by the motivation to achieve an accurate representation of the world (but see Morin et al, 2021 for a recent discussion).…”

Hybrid social learning in human-algorithm cultural transmission

“…Similarly, when human biases are known, algorithms can intentionally be designed to exhibit complementary biases to their human counterpart to enhance collective performance [14]. Although several heuristics that humans employ are adaptive under assumptions of cognitive constraints and bounded rationality [15][16][17], they can be suboptimal under restricted circumstances-e.g., in digital environments for which they are not adapted to [18][19][20]. In this study we focus on a specific human bias, namely the tendency for myopic behavior when facing a sequential decision [21][22][23].…”