How do people perceive artificial intelligence (AI)-generated fictional stories? In Study 1, participants (1,682 adults recruited from Prolific) read either a human-written or AI-generated story. Half were told that their story was written by a human, and half were told that their story was written by ChatGPT. Participants rated the AI-generated story as higher quality and more absorbing, but they rated the story better on both measures when they were told it was written by a human. In Studies 2 and 3, participants (905 adults recruited from Prolific) read both types of story and guessed each story’s origins. Participants were no better than chance at differentiating between the human-written and AI-generated stories. Self-reported expertise with AI, but not with fictional literature, was positively correlated with correct story identification. AI programs thus can generate creative work that is viewed as being at least as good as human-generated work but are not perceived as being able to do so.

Despite its well-established predictive validity across indicators of rationality and sound decision-making, actively open-minded thinking (AOT) remains understudied in adolescent populations. Even less is known about educational activities that might cultivate AOT during adolescence. The present study had two aims: first, to examine the psychometric functioning of existing AOT items in adolescents, and second, to investigate whether participation in competitive debate is associated with longitudinal change in AOT. The first measurement point included 420 high school students (69.5% girls; M = 16.20 years). Longitudinal analyses included 161 participants (73.3% girls; M = 16.55 years), with the primary comparison focusing on 40 debaters and 101 non-debate/non-journalism students. Psychometric analyses indicated substantial item heterogeneity and method effects related to item keying. Although a shorter positively worded item subset showed good internal coherence, its construct coverage and criterion validity were more limited than expected. For the longitudinal analyses, we therefore prioritized broader construct coverage over factorial purity. Debaters showed an increase in AOT over the course of a single semester, whereas the non-debate comparison group remained essentially unchanged. Exploratory item-level analyses suggested that debate-related change was most visible in openness to opposing perspectives, engagement with disagreement, and active search for reasons why one might be wrong. Although the observational design prevents causal conclusions, the findings point to competitive debate as a promising context for fostering AOT in adolescence and highlight the need for future research using larger samples, additional measurement points, and multimethod assessments.

Decades of research in judgment and decision making highlight the significant impact of defaults, in which individuals tend to accept preselected options. While some scholars argue that individuals exhibit ‘default neglect’, failing to strategically use defaults for economic or social gain, others suggest people are default aware and capable of using defaults to their advantage. To further investigate default neglect, we conducted 3 experiments in which participants acted as default setters in a decision-making ‘card game’ involving players with aligned, misaligned, unknown, or neutral interests. Across experiments, participants consistently selected high-value cards as defaults for aligned players and low-value cards for opponents. In Experiment 2, they were also willing to override preselected defaults even at a hypothetical cost when it benefited their interests. Experiment 3 revealed that participants could accurately infer which option was better based on the default-setting behavior of others, suggesting they expect others to use defaults strategically. Across all experiments, participants chose the optimal default in over 80% of trials. Together, the results challenge the idea of default neglect and suggest that individuals are attuned to and capable of leveraging defaults to influence others.

How the odds of winning a casino game are presented impacts whether individuals understand and how they perceive those odds. When framed in terms of what the player will win (return to player), younger and middle-aged adults have lower comprehension and higher perceived chances of winning than when the same odds are framed in terms of what the casino will keep (house edge), suggesting a classic framing effect. In the current study, we tested whether older (mean age = 69.01, SD = 3.02) and younger adults (mean age = 23.50, SD = 3.02) differ in their susceptibility to this framing effect and how measures of positivity, risk propensity, and financial exploitation vulnerability impact understanding and perceptions. Across both age groups, comprehension was lower and perceived chances of winning were higher in the return-to-player condition. Age alone did not predict comprehension, but older adults were less optimistic about coming out ahead, were less risk seeking, and self-reported lower financial exploitation vulnerability than younger adults. Exploratory mediation analyses revealed that lower financial exploitation vulnerability partially explained older adults’ reduced belief that they would win, particularly in the return-to-player frame. Together, these findings suggest that age differences in gambling perceptions reflect trait-level differences in financial vigilance and affective outlook rather than chronological age alone .

From a rational choice perspective, individuals are expected to follow rules when doing so maximizes their utility, implying deliberate, case-by-case decision-making. In many contexts, however, explicit deliberation about rule following is unlikely. Instead, individuals acquire rule-based behavioral patterns through experience and the observation of others and apply these rules whenever situations appear sufficiently similar to prior learning contexts. Such routinized rule following can enhance coordination, but it may also have a downside: reduced responsiveness when rules change. In a preregistered, incentivized laboratory experiment (N = 288), we examine how quickly individuals adapt to rule change. Using a newly developed and pretested paradigm, we manipulate between-subjects the learning environment (performance feedback only versus additional peer-observation or information search) and the beneficiary of rule compliance (individuals themselves versus others). Our findings indicate that learning environment and beneficiary each affect the speed of adaptation, but do not interact: Compared to mere performance feedback as a baseline, the adaptation to a new rule is faster when individuals can observe others’ behavior in the same situation, but not when they can actively search for information about the rule. Adaptation also happens more quickly when rule compliance benefits individuals themselves rather than serving others. Overall, the findings shed light on how learning channels and incentive structures jointly shape the speed of adaptation, with implications for the effective implementation of rule changes.

In a recent article on the gambler’s fallacy, Xiang, Dorst, and Gershman (2025, Psychological Science 36(6), 451–464) concluded that when event sequences are truly random, the fallacy is observed for binary predictions but not probability judgments and that binary predictions cannot be easily predicted from probability judgments. They suggested that the gambler’s fallacy does not arise from probabilistic reasoning, calling for new theories despite having not tested any existing theory directly. However, in the present article, we show that a more thorough reanalysis of Xiang et al.’s data yields strong evidence for a gambler’s fallacy for probability judgments and undermines the claimed disconnection between probability judgments and binary predictions. We find that an improved version of their representativeness model and a version of Rabin and Vayanos’s (2010, The Review of Economic Studies 77(2), 730–778) model account for both response types reasonably well, suggesting that probabilistic theories are not seriously threatened. Finally, we propose stratified random sampling as a more efficient method for presenting participants with representative event sequences.

The decision of when to claim Social Security benefits is among the most consequential and irreversible financial choices older Americans face, yet interventions to help guide it have had limited effectiveness. Building upon past research demonstrating that people choose differently and make different predictions for others than they do for themselves, we ask whether making recommendations for another can shift one’s own claiming decision. In two high-powered experiments using national survey panels, we show that older American adults intend to claim retirement income benefits earlier (vs. later) after making guided recommendations for a person who died relatively young (vs. lived to an old age) after retiring. This effect persists when the effect of recalling such a person is disentangled from that of recommending. This research suggests that making guided recommendations for others can be a powerful intervention for one’s own choices in highly consequential domains where trade-offs are difficult to make.

The field of Judgment and Decision Making has had tremendous success in understanding and improving decision-making in several domains from business to medicine, and it is now time to turn our efforts to the legal domain. The two schools of thought that dominate our field (coherence and correspondence) have generated a set of rich ideas as well as useful methods that can be applied in this context. However, both schools have been somewhat narrow in their focus on what is a ‘good’ decision. If we care about ‘good’ decisions, then we must also care about ‘just’ decisions. We ought to use our science to scrutinize the exercise of judicial power in the search for justice, and therefore in the service of justice. The direct relevance to our field is the fact that justice is a product of decisions, and these decisions are made under conditions of risk and uncertainty. We could reorient our science and put our intellectual resources into extending and applying our theories and methods in the following three ways. (1) Focus on justice as an outcome alongside, or even above the other sorts of outcomes we routinely study. (2) Study the exercise of judicial power in all its guises. (3) Scrutinize the algorithms that increasingly influence or replace judicial power. Justice can be modelled, measured, and improved — but only if we choose to prioritize it.

What is the best way to elicit forecasts if the goal is to measure the skill of forecasters? For continuous outcomes, there are two common approaches: elicit probabilities at fixed quantiles of the outcome distribution (e.g., what is the probability gas will cost less than $4.00 per gallon in 1 month?), or elicit quantiles at fixed probabilities (e.g., what is the price such that there is a 25% chance a gallon of gas will cost less than it in 1 month?). We compare these methods using a simulation study and a longitudinal survey in which 1,194 participants provided five waves of forecasts in both formats on a total of 36 real-world questions. Although probability elicitation is more ubiquitous in the literature, we show in our simulation and survey data that quantile elicitation allows us to estimate forecaster skill more efficiently with less measurement error. While quantile elicitation is time-consuming and associated with comprehension difficulty indicated by incoherent forecasts (violations of monotonicity), there is a clear reduction in incoherence across survey waves, suggesting comprehension of the quantile elicitation format improves with practice and is trainable.

The dual-process model of moral judgment proposes that utilitarian choices arise from reflective processes that can override intuitive, rule-based responses. Yet most supporting evidence relies on sacrificial dilemmas with limited ecological validity and problematic associations with antisocial traits. This preregistered study investigated whether individual differences in reflection predict utilitarian behavior in an incentivized setting. At Time 1, 1,200 U.S. participants completed performance-based measures of reflection, reasoning ability, dispositional thinking styles, and the Oxford Utilitarianism Scale (OUS). Seventeen months later, 321 participants completed a monetary moral dilemma with real economic consequences and reported their moral justifications. Initial measures of reflection, including the Cognitive Performance Test, Raven’s matrices, and open-minded thinking, predicted utilitarian behavior. Neither impartial beneficence nor instrumental harm subscales of the OUS predicted utilitarian behavior. Exploratory analyses showed that religiosity moderated the reflection–utilitarian behavior association: reflection predicted utilitarian behavior more strongly among non-believers than believers. These findings highlight reflection as a predictor of utilitarian behavior across time, question the validity of self-report utilitarianism scales such as the OUS, and underscore the importance of religious belief in cognitive reflection when examining individual differences in moral behavior.