Managing our natural resources often involves learning about the dynamics of the resource while society simultaneously harvests from the resource. However, little is known about how society’s actions affect our ability to learn about a vulnerable resource. Here, we investigate how experience managing a shared resource impacts one’s ability to learn about the underlying dynamics of a resource. We hypothesised that participants' capacity to learn from experience would depend on the behaviour of other group members. 320 participants played a four-player resource management game with computer partners who acted sustainably, unsustainably, or conditionally cooperatively. Participants’ ability to learn the underlying resource dynamics was then assessed in a subsequent single-player resource management game, where the social dynamics were removed and the participants’ understanding of the resource could be evaluated in isolation. Compared to controls with no prior experience, performance on the single-player game improved after experience with sustainable and unsustainable partners, but not after experience with conditionally cooperative partners. However, experience with a single-player game (rather than a group game) outperformed all other experimental conditions; thus individual experience improved performance more than experience in any group dynamic tested. These findings suggest that the presence of others generally hinders learning in resource management contexts, though how much this learning is hindered depends on the behaviour of others.
From an early age, children grapple with questions of fairness and generosity, but do boys and girls approach these moral challenges in the same way? This study examines altruistic behavior in 4- and 5-year-old children using a modified version of the dictator game with two framing conditions: give (sharing from one's own endowment) and take (subtracting from another's endowment). These frames elicit distinct moral intuitions, offering a more nuanced perspective on early prosocial behavior. The gender of the recipient (boy vs. girl) was also manipulated to assess whether gender-based solidarity or competitive biases influenced sharing decisions. Results revealed that, on average, boys and girls responded differently to the framing of the task. Boys shared less when giving from their own endowment (37%) and more when taking from another's (61%), resulting in a notably large framing effect (d = 1.21). Girls, instead, shared more similar amounts across both contexts (45% in give scenarios vs. 51% in take scenarios; d = 0.31). While children varied in their individual responses, these differences suggest that boys were more sensitive to who originally "owned" the resource, whereas girls displayed a steadier inclination toward equality. The overall framing effect was substantial (d = 0.77) and appeared stronger among younger children, hinting at developmental variability even within this narrow age range. No effects of recipient gender were found, indicating that gender-based distinctions may not influence prosocial behavior at this age. Taken together, these findings suggest that preschoolers' sharing behavior is shaped by the moral framing of the action, revealing gender-specific sensitivities.
To validate novel measurement instruments of people's decision making, the behavioral sciences often assess how frequently people make specific risky choices in real life. Yet, it remains unclear to what degree the risky choices tapped by such frequency measures reflect those choices people actually (have to) make. To address this issue, we compared 100 risky choices representing laypeople's perspective with 63 risky choices representing researchers' perspective. In study 1, we leveraged natural language processing calibrated on human judgments to gauge the similarity between choices of both perspectives and found that they only had 18% of the choices in common. In study 2, we further examined the implications of this mismatch in terms of the psychological mechanisms that the various choices may tap into by asking 825 participants to rate the perceived relevance of seven classes of psychological mechanisms to their decision making in these choices. Bayesian mixed effects models revealed credible differences between the two perspectives in five out of seven classes of mechanisms: For choices representing laypeople's perspective, choice attributes, time factors, experience and knowledge, and goals and motivation were on average perceived as more relevant, and social factors were perceived as less relevant, relative to the choices representing researchers' perspective. These findings suggest that decision-making paradigms calibrated on frequency measures from the researchers' perspective may have limited generalizability to the broad range of risky choices people face in their lives, underscoring the need to better understand the complexity of real-life decision making
Experimental work reveals that participants who have access to all situational perspectives in moral scenarios (full perspective-taking accessibility) are more prosocial in their moral judgments than those who receive scenarios offering only one situational perspective (partial perspective-taking accessibility; Martin, Kusev, & van Schaik, 2021). Since previous studies have only focused on decisions made directly after scenario presentation, in the present work, we have explored how perspective-taking accessibility influences moral judgments under varying cognitive priming tasks (no prime, task-relevant prime, and task-irrelevant prime). We found that with full perspective-taking accessibility, participants were consistently utilitarian in their moral judgments, regardless of the cognitive priming task employed. However, with partial perspective-taking accessibility, participants were more utilitarian in their moral judgments after undertaking a task-irrelevant prime (an anagram task) compared to a task-relevant prime or no prime. In Experiments 1 and 2, we found that placing the anagram task after the moral scenario (late prime) induced participants' utilitarian moral judgments. Moreover, in Experiment 3, we explored whether placing the anagram task before the moral scenario (early prime) would have a similar effect on participants' moral judgments. With partial perspective-taking accessibility, regardless of the anagram task placement (early or late prime), participants were more utilitarian in their moral judgments compared to participants who were not primed with an anagram task. However, crucially, the results revealed no statistically significant difference between receiving an early and late prime; the properties of the anagram task itself (and not a distraction period between the scenario and judgment task) enhance participants' utilitarian behavior.
Recent studies on how and when people rely on information generated by human experts or algorithms have produced mixed results. When choosing between the two sources of information or rating their expected accuracy, people often show systematic preferences depending on the task domain, preferring algorithms for more objective and quantitative tasks and humans for more subjective and qualitative tasks. Results also indicate people prefer hybrid advice, which combines both human and algorithmic inputs, to either source on its own across various domains. However, when judges are provided with unsolicited advice and tasked with updating their own prior independent judgments in judge advisor system experiments, this pattern of results often vanishes or even reverses. We attempt to reconcile these differences in two solicited judge advisor system geopolitical forecasting experiments in which judges must first solicit advice from their preferred advisor before deciding if and how to revise their forecasts. We find that the pattern of selection decisions remains consistent with prior research: People tend to choose hybrid advisors when they are available over either humans or algorithms on their own. However, we find no differences in how people revise their beliefs based on the advice source selected, and we find no benefit to accuracy when hybrid advice is available versus when it is not. We also find clear effects of imposing a cost on soliciting advice. When advice was made costly, people were both more judicious about soliciting advice and saw greater accuracy gains when they did solicit it.
There has been intense interest in biases in legal decision making, such as order effects and evaluation biases (biases arising from making judgments, as opposed to just observing some information). We extend previous work in three ways. First, we employ a population sample including judges, prosecutors, and attorneys, as well as na & iuml;ve participants, to investigate the extent of biases for legal professionals. Second, we use realistic materials, summaries of real legal cases. Finally, we study two biases, order effects and the Evaluation Bias, the latter being a bias corresponding to more extreme evaluations if a previous, oppositely valenced piece of information had been evaluated versus just observed. Both biases were reliably observed across all groups of legal professionals and a group of lay participants; there was no evidence that different groups of participants displayed either of the two biases to a lesser extent. The presence of two basic decision biases in a study involving realistic legal stimuli and with legal professionals raises questions about the robustness of decision processes in the legal system.
In a 6-week randomized, triple-blind, placebo-controlled trial, we investigated the impact of a probiotic intervention on risky choices in healthy adults as well as a potential link between gut microbiota and risky decision-making. Our study explored whether the gut-brain interaction was mediated by gut bacteria or cardiac vagal activity, representing the vagal pathway in the gut-brain axis. Additionally, we examined whether these potential mediations would be moderated by interoceptive accuracy. To assess risky decision-making, the Iowa Gambling Task (IGT; Bechara et al., 1994) and the Balloon Analogue Risk Task (BART; Lejuez et al., 2002) were employed. Interoceptive accuracy was captured with the heartbeat perception task (Dunn et al., 2012; Schandry, 1981). Neither the probiotic intervention nor many of the tested bacteria and cardiac vagal activity predicted risky decision-making. Our data suggest a potential gut-cognition link in healthy adults predominantly via Faecalibacterium prausnitzii and Lactobacillaceae that merits further investigation. This connection was found for risky decision-making in the BART and IGT tasks and was moderated by interoceptive accuracy. Participants with high accuracy in perceiving internal bodily signals while also exhibiting higher numbers of specific bacteria (i.e., F. prausnitzii, Lactobacillaceae) were less inclined to make risky choices. While our results warrant further research concerning the role of Bacillota family of gut bacteria, other recent studies have brought neurotransmitters into play. Future research should consider these possible factors together, scrutinizing psychophysiological mechanisms in risky decision-making.
When decision makers receive multiple uncertain forecasts (as with the COVID-19 pandemic), they need displays that will help them to integrate the competing predictions. One display option is presenting the full suite of forecasts. A second is presenting a summary ensemble, reducing cognitive load at the price of obscuring disagreements. One compromise is presenting both, allowing decision makers to tailor their usage, with an even greater cognitive load. A second compromise is presenting one display but allow users to click through to see the other. In two experiments, using a suite of 10 CDC forecasts and an ensemble based on median values, we compared users' performance with displays presenting the (a) suite, then adding the ensemble; (b) the ensemble, then adding the suite; and (c) the suite and the ensemble combined. Using diverse, online convenience samples and hypothetic scenarios, we found that participants responded similarly to the combined displays, if they saw them initially together or only after seeing the suite or the ensemble. Participants responded differently if they saw just the suite or just the ensemble-as would happen if they did not click through from the single display. Outliers in the suite display appeared responsible for the difference. The result was robust whether the suite displays were ordered randomly, by confidence interval widths, or by median prediction. Comprehension was good with all displays. These results indicate the importance of ensuring that decision makers receive and understand both the suite and the ensemble perspective when there are multiple uncertain forecasts.
Prisoner's dilemma (PD) games exist in a variety of configurations. Although the normative solution for any one-shot PD game is defection, empirical research shows that games differ in the amount of cooperation they elicit from human players. Different researchers have developed a variety of indexes, which are functions of the specific payoffs in a given PD, aiming to measure the level of anticipated cooperation elicited by that game. The purpose of the first study in this article was to collect, organize, and compare the mathematical properties of these various indexes. Our results show that many superficially different indexes are the same and that in general six fundamental PD indexes exist. We propose that the six families of indexes correspond to psychological processes and preferences involving prosociality, greed, fear, and risk. In the second study, we provide a meta-analysis of the relationship between selected indexes and cooperation in PD games (k = 280). The amount of variance in cooperation explained by each index ranged from 0.00% to 7.57%. The first family of indexes (described as indexes of the incentive for cooperation in the first study) had the most consistently strong relationships with cooperation. In general, the findings of the meta-analysis aligned with predictions based on our own review and taxonomy of six index families proposed in Study 1.
Loss aversion-the tendency to avoid losses-has been one of the fundamental ideas in decision research across psychology and behavioral economics. Building upon the general notion of bad being stronger than good and on prospect theory's value function, the statement "losses loom larger than gains" became accepted as a fundamental principle of human behavior. However, the possibly important role of the magnitude of stakes in the psychological valuation process has not been formally considered, though it is a salient aspect in both theoretical and empirical terms. To fill that gap, we review various studies before and after prospect theory and find evidence that underscores the role of magnitude in loss aversion. Magnitude-dependent loss aversion is a nuanced proposition that explicitly foregrounds the role of stake size in the psychological valuation of gains versus losses.
We human beings are naturally inclined to characterize the morality of others. Whereas some actions lead us to immediate moral characterization of the agent, others can be more difficult to evaluate. Prior research suggests that when judging ambiguous actions, we rely upon our perception of the agent's motivation. Yet we are often not privy to the reasons behind agents' actions, making us search for cues of their moral motivation. Building on research on motivation attribution and person perception, we suggest here that self-sacrifice, the willingness to incur a personal cost, is a powerful cue that judges use to infer an agent's motivation. We hypothesize that people will be more likely to judge an agent positively when their action involves self-sacrifice, as it is perceived as a reflection of moral motivation. However, when the agent's motivation is clearly immoral, self-sacrifice will not affect moral judgment. Six vignette studies (N = 3,931) each manipulating the actor's self-sacrifice, supported this hypothesis across various domains, including views of political policies, assessment of risky military decisions, and monetary decisions.
Experimental research in decision-making often relies on tasks that provide participants with all the information they need to make their decisions. Here we consider the process by which decision-makers seek information about their choice alternatives when it is not immediately provided. Recent advances in computational theories have proposed that people will seek to process more information about options for which they are less certain about the value. Specifically, they will allocate more attention to options with lower certainty in order to increase their certainty by processing additional information. We tested this hypothesis with a behavioral and eye-tracking experiment in which participants observed pairs of random streams of numerical stimuli and were incentivized to report which stream was generated by the distribution with the greater mean. We induced uncertainty by manipulating the variance of the value distribution for each option. In addition, we randomly replaced some of the stimuli with meaningless letters. This decreased the accessibility of information about the options. The results show that people fixate more on options with lower accessibility and to a lesser extent options with greater value. Interestingly, this pattern changes across response time, with early fixations driven by variability and accessibility, and late fixations driven by value and accessibility. Moreover, people were more likely to choose options with greater accessibility, and they felt more confident about their choices when accessibility was greater. This research could help illuminate the process of information seeking to reduce uncertainty during choice deliberation.
In complex environments, the space of possible plans is vast. Generating a good plan therefore requires judicious selection of which parts of the plan space to mentally explore. Drawing on past studies of human exploration, we propose that mental exploration might invoke similar mechanisms. In particular, we test the hypothesis that mental exploration during planning is uncertainty-driven, such that people will exhibit a tendency to explore parts of the plan space that have high epistemic uncertainty. We developed a route-planning task, displayed as a binary tree, where participants were instructed to collect as many treats (rewards) as possible by traversing the tree. By separating the planning and execution phases, we encouraged participants to externalize their planning process. We manipulated uncertainty by varying the number of potential future states available from each current state. Across two studies, the data suggest that people preferred to explore options with more successor states after controlling for value differences, supporting the uncertainty-driven planning hypothesis. We also found that uncertainty played a larger role during the planning phase than during the execution phase, consistent with the hypothesis that the uncertainty effect primarily reflects a property of human planning algorithms rather than an intrinsic preference for uncertainty.
Intertemporal choice tasks are used to measure how people make decisions between outcomes spaced out in time. Measurements from those tasks are often used to study impulsivity and self-control, concepts central in theories and empirical research concerning addiction, substance use, psychopathology, and organisational behaviour, among many. Accordingly, preferences for smaller rewards received sooner, over larger rewards received later, have been linked to higher impulsivity and less self-control. This paper is a critique of that approach. To present our argument, we first provide a historical overview of research on time preferences tracking the theoretical link between intertemporal choice, impulsivity, and self-control. Our subsequent conceptual analysis reveals that impulsivity concerns a lack of reflection on one’s choices, not a lack of concern with the future, and self-control concerns internal conflict due to temptation, rather than future-orientedness. People may, and do, use self-control to choose a "smaller-sooner" reward or impulsively select a "larger-later" reward. We also address technical limitations pertaining to the reliability, external and predictive validity of intertemporal choice tasks. We argue that, therefore, impulsivity and self-control cannot be measured using a standard intertemporal choice task. We canvass possible future directions for decision-making models in this area, providing the basis for new understanding of how impulsivity, self-control, and time preferences influence behaviour across different domains. We suggest that to study impulsivity and self-control in a temporal context, more information is needed about agents’ motivation and deliberative process.
Despite the importance of predictive judgments, individual human forecasts are frequently less accurate than those of even simple prediction algorithms. At the same time, not all forecasts are amenable to algorithmic prediction. Here, we describe the evaluation of an automated prediction tool that enabled participants to create simple rules that monitored relevant indicators (e.g., commodity prices) to automatically update forecasts. We examined these rules in both a pool of previous participants in a geopolitical forecasting tournament (Study 1) and a na & iuml;ve sample recruited from Mechanical Turk (Study 2). Across the two studies, we found that automated updates tended to improve forecast accuracy relative to initial forecasts and were comparable to manual updates. Additionally, making rules improved the accuracy of manual updates. Crowd forecasts likewise benefitted from rule-based updates. However, when presented with the choice of whether to accept, reject or adjust an automatic forecast update, participants showed little ability to discriminate between automated updates that were harmful versus beneficial to forecast accuracy. Simple prospective rule-based tools are thus able to improve forecast accuracy by offering accurate and efficient updates, but ensuring forecasters make use of tools remains a challenge.
Product reviews on e-commerce platforms can have a pronounced effect on consumers' decisions.Less is known, however, whether the reviews written by others can shape a person's own written opinion of a product.We hypothesized that people who compose reviews on digital storefronts will try to imitate successful reviews, such that their content will show similarity with other reviews displayed at the time of writing.More specifically, we predicted that reviews will be more semantically similar to the most successful, salient, and readily accessible reviews written by others.To investigate this hypothesis, we extracted over 3 million reviews from a major online distribution platform and traced the reviews that were displayed at the time when each review was being composed.Using word embeddings from a pre-trained language model, we quantified the semantic similarity between a given review and other reviews that were visible (or not) to a user.We found that reviewers imitate the most helpful reviews written by others, especially those that are visually salient.Their reviews, in turn, gather more helpfulness ratings in the future, leading to a cascade of similar reviews.Our findings suggest that the default sorting and display format of reviews on online platforms will have a pronounced effect on the style and content of new reviews.
Despite the volume of research examining overprecision, the underlying drivers of individual differences in probability estimations remain elusive. I propose that visual processing through mental simulation of small samples is a cognitive mechanism influencing the relative degrees of over and underprecision. I conducted three preregistered experiments contrasting probability estimates between a control group and a treatment group, where participants were prompted to engage in greater visual processing by mentally simulating outcomes. In Study 1, participants estimate a binomial distribution of a ball drop machine. I find that engaging in visual simulation led to higher estimates of values near the distributions' center, while the control group provided higher estimates for the distributions' tails. Although the control group is more underprecise, visual simulation could arguably increase estimation accuracy and not overconfidence. To separate these effects, I modify the ball drop mechanism in Study 2 to produce a flatter distribution. The results show that the control group is well-calibrated, but the visual simulation group is overprecise. Study 3 investigates a boundary condition where participants mentally simulated multiple outcomes. The results demonstrate that an increase in the variance of imagined outcomes lowers subjective estimates near the center of the distribution, diminishing the treatment effect.
