Journal of Behavioral Decision Making最新文献

Subjective inferences of probability are critical for decisions involving uncertainty. Recent studies have suggested that extreme outcomes bias beliefs about the value of uncertain options toward the best/worst outcome possible when learning the odds through experience, leading to increased preferences for uncertain options over equivaluable sure bets when there is the prospect of gain and, conversely, aversion to uncertain options when there is the prospect of loss. However, prior studies regarding the influence of extreme outcomes on decisions involving uncertainty have only done so using 50/50 gambles, and it was unclear whether extreme outcomes biased probability perception more broadly. Across three pre-registered experiments, we found that when people made decisions between equivaluable certain and uncertain options, they particularly preferred uncertain options at low probabilities (20%) when there was the prospect of gain and avoided them when there was the prospect of loss, with these preferences being reduced or even reversed at medium (50%) and high (80%) probabilities. We also found that uncertainty preferences were influenced by outcome extremity and the relative certainty associated with safe reference options. We conclude that extreme outcomes accentuate the overweighting of low probabilities and the underweighting of high probabilities in experience-based choice.

Emotions integral to a task are often adaptive, particularly in situations where outcomes and probabilities are not known. However, decisions are also influenced by emotions that arise from situations unrelated to the task. This is especially the case with negative emotions like fear and anger, which also tend to be accompanied by ruminative thinking that might divert decision-makers' attention from the task at hand. In two preregistered experiments, we show how self-distancing regulates the influence of incidental anger (vs. fear) on decision-making under uncertainty. Participants recalled and reflected on a fear-related or anger-related event from either a self-immersed or self-distanced perspective. Next, they completed a task that is commonly used to measure affective decision-making under uncertainty, the Iowa Gambling Task. The results in both experiments indicated that self-immersed angry (vs. fearful) decision-makers were significantly slower to avoid the risky, disadvantageous decks. These findings demonstrate how the ways in which we process negative emotional events shape their carryover effects in decision-making under uncertainty and point to self-distancing as a potential tool to control incidental emotional influences.

There is ongoing debate over the usefulness of and need for attention checks in online experiments. This paper investigates the value of these tests in decisions-from-experience (i.e., multi-trial repeated choice) tasks. In five studies (Ntotal = 1519), we comprehensively compared the behavior of attentive and inattentive participants (i.e., those who passed or failed a simple attention check) among online participants; and also compared those results to the results of lab studies reported elsewhere. We found meaningful differences between the behavior of attentive and inattentive participants even at the first trial. Overall, attentive participants were more likely to notice less-obvious average values of the different alternatives, while inattentive participants exhibited higher sensitivity to typical outcomes. The findings show that even one simple attention test is sufficient to differentiate between attentive and inattentive participants in repetitive tasks. Importantly, our results fully replicated three previously run lab studies among attentive participants, but not inattentive ones. This finding highlights the importance of using attention tests to avoid spurious conclusions.

Experts are usually valued for their knowledge. However, do they possess metaknowledge, that is, knowing how much they know as well as the limits of that knowledge? The current research examined expert metaknowledge by comparing experts' and nonexperts' confidence when they made correct versus incorrect choices as well as the difference in-between (e.g., Murphy's Resolution and Yate's Separation). Across three fields of expertise (climate science, psychological statistics, and investment), we found that experts tended to display better metaknowledge than nonexperts but still showed systematic and important imperfections. They were less overconfident than nonexperts in general and expressed more confidence in their correct answers. However, they tend to exhibit low Murphy's Resolution similar to nonexperts and gave endorsed wrong answers with equal to higher confidence than did their nonexpert peers. Thus, it appears that expertise is associated with knowing with more certainty what one knows but conceals awareness of what one does not know.

Advice taking and related research is dominated by deterministic weighting indices, specifically ratio-of-differences-based formulas for investigating informational influence. Their arithmetic is intuitively simple, but they pose several measurement problems and restrict research to a particular paradigmatic approach. As a solution, we propose to specify how strongly peoples' judgments are influenced by externally provided evidence by fitting corresponding mixed-effects regression models. Our approach explicitly distinguishes between endogenous components, such as updated beliefs, and exogenous components, such as independent initial judgments and advice. Crucially, mixed-effects regression coefficients of various exogenous sources of information also reflect individual weighting but are based on a conceptually consistent representation of the endogenous judgment process. The formal derivation of the proposed weighting measures is accompanied by a detailed elaboration on their most important technical and statistical subtleties. We use this modeling approach to revisit empirical findings from several paradigms investigating algorithm aversion, sequential collaboration, and advice taking. In summary, we replicate and extend the original finding of algorithm appreciation and initially demonstrate a lack of evidence for both systematic order effects in sequential collaboration as well as differential weighting of multiple pieces of advice. In addition to opening new avenues for innovative research, appropriate modeling of information sampling and utilization has the potential to increase the reproducibility and replicability of behavioral science. Furthermore, the proposed method is relevant beyond advice taking, as mixed-effects regression weights can also inform research on related cognitive phenomena such as multidimensional belief updating, anchoring effects, hindsight bias, or attitude change.

Gambles that involve a large but unlikely gain coupled with a small but likely loss—like a lottery ticket—are known as positively skewed. There is evidence that people tend to prefer these positively skewed choices, leading to what is called a positive-skew bias. In this study, we attempt to better understand under what conditions people are more drawn towards positively skewed, relative to symmetric, gambles. Based on the animal literature, there is reason to believe that preference for skewed gambles is dependent on the strength of the skew, with a greater preference for more strongly skewed options. In two online studies (Study 1: N = 209; Study 2: N = 210), healthy participants across the lifespan (ages 22–85) made a series of choices between a positively skewed risky gamble and either a certain outcome (Study 1) or risky symmetric gamble (Study 2). Logistic regression analyses revealed that people were more likely to choose moderately and strongly skewed gambles over certain outcomes, with the exception of when there were large potential losses (Study 1). However, a stronger skewness did not increase preference for positively skewed gambles over symmetric gambles, findings that also may depend on the valence of the expected outcome (Study 2). Taken together, these results suggest that there may be a greater preference for more strongly positively skewed gambles, but it (1) is dependent on what other gamble is presented and (2) is most prevalent for positive expected values. Additionally, contrary to previous findings, we did not find strong evidence of an age-related increase in positive skew bias in either study. However, exploratory analyses revealed that decision making strategy and cognitive abilities may play a role.

Selection decisions are often affected by irrelevant variables such as gender or race. People can discount this irrelevant information by adjusting their predictions accordingly, yet they fail to do so intuitively. In five online studies (N = 1077), participants were asked to make selection decisions in which the selection test was affected by irrelevant attributes. We examined whether in such decisions people are willing to be advised by algorithms, human advisors or prefer to decide without advice. We found that people fail to adjust for irrelevant information by themselves, and those who received advice from an algorithm or human advisor made better decisions. Interestingly, although most participants stated they prefer advice from human advisors, they tend to rely equally on algorithms in actual selection tasks. The sole exception is when they are forced to choose between an algorithm and a human advisor. In that case, they pick human advisors. We conclude that while algorithms may not be people's preferred source of advice in selection decisions, they are equally useful and can be implemented.