In many learning or inference tasks human behavior approximates that of a Bayesian ideal observer, suggesting that, at some level, cognition can be described as Bayesian inference. However, a number of findings have highlighted an intriguing mismatch between human behavior and standard assumptions about optimality: People often appear to make decisions based on just one or a few samples from the appropriate posterior probability distribution, rather than using the full distribution. Although sampling-based approximations are a common way to implement Bayesian (...) inference, the very limited numbers of samples often used by humans seem insufficient to approximate the required probability distributions very accurately. Here, we consider this discrepancy in the broader framework of statistical decision theory, and ask: If people are making decisions based on samples—but as samples are costly—how many samples should people use to optimize their total expected or worst-case reward over a large number of decisions? We find that under reasonable assumptions about the time costs of sampling, making many quick but locally suboptimal decisions based on very few samples may be the globally optimal strategy over long periods. These results help to reconcile a large body of work showing sampling-based or probability matching behavior with the hypothesis that human cognition can be understood in Bayesian terms, and they suggest promising future directions for studies of resource-constrained cognition. (shrink)

Previous research suggests that individuals faced with risky choices seek ways to actively reduce their risks. The risk defusing operators that are identified through these searches can be used to prevent or compensate for negative outcomes. Although several factors that affect RDO selection have been identified, they are limited to static decisions conducted during descriptive tasks. The factors that influence RDO selection in dynamically unfolding environments are unknown, and the relationship between task characteristics and RDO selection has yet to be (...) mapped. We used a videogame environment to conduct two experiments to address these issues and found that experienced losses impacted risk mitigation strategy: when the task was difficult, participants experienced greater losses and were more likely to select preventive RDOs. Additionally, risk mitigation behavior stabilized as participants gained experience with the task and could be shifted by making an RDO easier to use. Exploratory analyses suggested that these risk mitigation choices were not driven by judgments of difficulty, even though participants’ JODs were accurate and aligned with task difficulty. This research suggests that while people seek preventive RDOs when tasks are difficult and risky, risk mitigation strategy is shaped by experienced losses; decision makers do not use JODs to anticipate future risks and inform risk mitigation decisions. (shrink)

Prior research in judgment and decision making (JDM) has investigated the effect of problem framing on human preferences. Furthermore, research in JDM documented the absence of such reversal of preferences when making decisions from experience. However, little is known about the effect of context on preferences under the combined influence of problem framing and problem format. Also, little is known about how cognitive models would account for human choices in different problem frames and types (general/specific) in the experience format. One (...) of the primary objectives of this research is to investigate the presence of preference reversals under the influence of problem framing (gain/loss), problem format (experience/description), and problem type (general/specific). Another objective of this research is to develop cognitive models to account for human choices across different problem frames and types in the experience format. A total of 320 participants from India were randomly assigned to one of eight between-subjects conditions that differed in problem frame, format, and type. Results revealed preference reversals in the description condition; however, they were absent in the experience condition. Moreover, preference reversals were less pronounced in the general problem framing compared to the specific problem framing. Furthermore, specific problems influenced risk-seeking behavior among participants. We developed cognitive and heuristics models using instance-based learning theory and natural mean heuristic. Results reveal models’ dependency on recent and frequent observations during information sampling. These experience-based cognitive models could help build artificial intelligence models with fewer preference reversals. (shrink)

The aim of this study was to investigate the effect of numeracy and the emotion of fear on the decision-making process. While previous research demonstrated that these factors are independently related to search effort, search policy and choice in a decision from experience task, less is known about how their interaction contributes to processing information under uncertainty. We attempted to address this problem and to fill this gap. In the present study, we hypothesized that more numerate people would sample more (...) information about a decision problem and that the effect of fear would depend on the source of this emotion: whether it is integral (i.e., relevant) or incidental (i.e., irrelevant) to a decision problem. Additionally, we tested how these factors predict choices. We addressed these hypotheses in a series of two experiments. In each experiment, we used a sampling paradigm to measure search effort, search policy and choice in nine binary problems included in a decision from experience task. In Experiment 1, before the sampling task we elicited incidental fear by asking participants to recall fearful events from their life. In Experiment 2, integral fear was elicited by asking participants to make choices concerning medical treatment. Decision problems and their payoff distributions were the same in the two experiments and across each condition. In both experiments, we assessed objective statistical numeracy and controlled for a change in the current emotional state. We found that more numerate people sampled more information about a decision problem and switched less frequently between alternatives. Incidental fear marginally predicted search effort. Integral fear led to larger sample sizes, but only among more numerate people. Neither numeracy nor fear were related to the number of choices that maximized expected values. However, across two experiments sample sizes predicted the number of choices that maximized experienced mean returns. The findings suggest that people with higher numeracy may be more sensitive to integral emotions; this may result in more effortful sampling of relevant information leading to choices maximizing experienced returns. (shrink)

Prospect theory predicts a four-fold risk attitude, which means that people are risk seeking for low-probability gain and high-probability loss and risk averse for low-probability loss and high-probability gain because they overweight probability when it is low. The four-fold pattern of risk attitude has been supported by several former studies with mainstream industrialized populations but has never previously been tested in a non-industrialized society. In this work, we examined the robustness of the four-fold risk attitude in the DuLong minority ethnicity (...) in China, which is a small society with only 4,000 members that is isolated from modern civilization. We used simple lotteries for gain and loss with different probabilities to elicit the risk attitude of 37 DuLong villagers. Our results support prospect theory predictions in that DuLong people are risk seeking for low-probability gain and risk averse for low-probability loss. However, although they showed a tendency to decrease their degree of risk seeking for gain, their risk attitude did not reverse when the probability of the prospect increased to 50%. In summary, our results suggest a right-shifted weighting function in this non-industrialized small society. The deviation might be caused by the particular living situation of the DuLong people, their sensitivity to monetary payoffs, and the elicitation procedure. (shrink)

Recent studies indicate that indicative conditionals like "If people wear masks, the spread of Covid-19 will be diminished" require a probabilistic dependency between their antecedents and consequents to be acceptable (Skovgaard-Olsen et al., 2016). But it is easy to make the slip from this claim to the thesis that indicative conditionals are acceptable only if this probabilistic dependency results from a causal relation between antecedent and consequent. According to Pearl (2009), understanding a causal relation involves multiple, hierarchically organized conceptual dimensions: (...) prediction, intervention, and counterfactual dependence. In a series of experiments, we test the hypothesis that these conceptual dimensions are differentially encoded in indicative and counterfactual conditionals. If this hypothesis holds, then there are limits as to how much of a causal relation is captured by indicative conditionals alone. Our results show that the acceptance of indicative and counterfactual conditionals can become dissociated. Furthermore, it is found that the acceptance of both is needed for accepting a causal relation between two co-occurring events. The implications that these findings have for the hypothesis above, and for recent debates at the intersection of the psychology of reasoning and causal judgment, are critically discussed. Our findings are consistent with viewing indicative conditionals as answering predictive queries requiring evidential relevance (even in the absence of direct causal relations). Counterfactual conditionals in contrast target causal relevance, specifically. Finally, we discuss the implications our results have for the yet unsolved question of how reasoners succeed in constructing causal models from verbal descriptions. (shrink)

One of the paradigms in judgment and decision-making involves decision-makers sample information before making a final consequential choice. In the sampling paradigm, certain computational models have been proposed where a set of single or distribution parameters is calibrated to the choice proportions of a group of participants. However, currently little is known on how aggregate and hierarchical models would account for choices made by individual participants in the sampling paradigm. In this paper, we test the ability of aggregate and hierarchical (...) models to explain choices made by individual participants. Several models, Ensemble, Cumulative Prospect Theory, Best Estimation and Simulation Techniques, Natural-Mean Heuristic, and Instance-Based Learning, had their parameters calibrated to individual choices in a large dataset involving the sampling paradigm. Later, these models were generalized to two large datasets in the sampling paradigm. Results revealed that the aggregate models accounted for individual choices better than hierarchical models upon generalization to problems that were like those encountered during calibration. Furthermore, the CPT model, which relies on differential valuing of gains and losses, respectively, performed better than other models during calibration and generalization on datasets with similar set of problems. The IBL model, relying on recency and frequency of sampled information, and the NMH model, relying on frequency of sampled information, performed better than other models during generalization to a challenging dataset. Sequential analyses of results from different models showed how these models accounted for transitions from the last sample to final choice in human data. We highlight the implications of using aggregate and hierarchical models in explaining individual choices from experience. (shrink)

À la fin des années 1990, plusieurs tendances convergentes en économie, psychologie et neuroscience ont préparé le terrain pour la naissance d’un nouveau champ scientifique qualifié de « neuroéconomie ». Comme pour toute discipline émergente – pensons par exemple à l’économie mathématique, l’économétrie ou l’économie expérimentale en d’autres temps – la neuroéconomie est plutôt controversée en économie. Elle soulève un grand nombre de questions d’ordre empirique, méthodologique et philosophiques donnant lieu à des débats et controverses que l’article identifie et discute (...) à travers trois questions majeures relatives aux connaissances produites : Sont-elles fiables? Sont-elles scientifiquement légitimes? Sont-elles utiles pour les économistes? Sans prétendre à l’exhaustivité, ce cadre analytique permet de comprendre l’essence des principales critiques adressées à la neuroéconomie. Dans le même temps, l’article éclaire la nature des réponses susceptibles d’être apportées à ces critiques au vu des progrès conceptuels et empiriques qu’a connus ce champ de recherche depuis son émergence et à la lumière du tournant apparu dans l’agenda de la neuroéconomie au cours de la dernière décennie. JEL Codes : A12, B41, C51, C81, C9, D03, D87. (shrink)

In spite of the remarkable progress made in the burgeoning field of social neuroscience, the neural mechanisms that underlie social encounters are only beginning to be studied and could —paradoxically— be seen as representing the ‘dark matter’ of social neuroscience. Recent conceptual and empirical developments consistently indicate the need for investigations, which allow the study of real-time social encounters in a truly interactive manner. This suggestion is based on the premise that social cognition is fundamentally different when we are in (...) interaction with others rather than merely observing them. In this article, we outline the theoretical conception of a second-person approach to other minds and review evidence from neuroimaging, psychophysiological studies and related fields to argue for the development of a second-person neuroscience, which will help neuroscience to really go social; this may also be relevant for our understanding of psychiatric disorders construed as disorders of social cognition. (shrink)

This paper discusses methodological issues of two recent experiments conducted by Christopher Bartel, and Elzė S. Mikalonytė and Vilius Dranseika, respectively, about the repeatability and individuation of musical works. I argue, first, that the reliability of their results about people’s intuitions in our everyday musical practices can be questioned due to the use of descriptions instead of musical stimuli of the works and performances involved in the cases tested. This procedure is prone to place participants in an epistemic situation in (...) which their mental capacities for their everyday judgments are not deployed and makes their judgments more vulnerable to the problem of “filling in”. I provide empirical data of an experiment that implements musical stimuli, obtaining large effect sizes regarding the results obtained by Bartel, Mikalonytė and Dranseika. In addition, I argue that the results of three scenarios of the experiment conducted by Mikalonytė and Dranseika are suspicious of being influenced by presentation effects that favor the contextualist intuition that the composer is a relevant parameter in work-individuation. Formal and superficial aspects of the description of their cases would cause these effects. I supply empirical data of an experiment that neutralizes those effects, obtaining moderate and large effect sizes regarding the results of Mikalonytė and Dranseika. (shrink)

Using belief elicitation, the paper investigates the process of belief formation and evolution in a signaling game in which a common prior is not induced. Both prior and posterior beliefs of Receivers about Senders’ types are elicited, as well as beliefs of Senders about Receivers’ strategies. In the experiment, subjects often start with diffuse uniform beliefs and update them in view of observations. However, the speed of updating is influenced by the strength of initial beliefs. An interesting result is that (...) beliefs about the prior distribution of types are updated slower than posterior beliefs, which incorporate Senders’ strategies. In the medium run, for some specifications of game parameters, this leads to outcomes being significantly different from the outcomes of the game in which a common prior is induced. It is also shown that elicitation of beliefs does not considerably change the pattern of play in this game. (shrink)

“Decisions from experience” (DFE) refers to a body of work that emerged in research on behavioral decision making over the last decade. One of the major experimental paradigms employed to study experienced-based choice is the “sampling paradigm”, which serves as a model of decision making under limited knowledge about the statistical structure of the world. In this paradigm respondents are presented with two payoff distributions, which, in contrast to standard approaches in behavioral economics, are specified not in terms of explicit (...) outcome-probability information, but by the opportunity to sample outcomes from each distribution without economic consequences. Participants are encouraged to explore the distributions until they feel confident enough to decide from which they would prefer to draw from in a final trial involving real monetary payoffs. One commonly employed measure to characterize the behavior of participants in the sampling paradigm is the sample size, that is, the number of outcome draws which participants choose to obtain from each distribution prior to terminating sampling. A natural question that arises in this context concerns the “optimal” sample size, which could be used as a normative benchmark to evaluate human sampling behavior in DFE. In this theoretical manuscript, we relate the DFE sampling paradigm to the classical statistical decision theoretic literature and, under a probabilistic inference assumption, evaluate optimal sample sizes for decisions from experience. In our treatment we go beyond analytically established results by showing how the classical statistical decision theoretic framework can be used to derive optimal sample sizes under arbitrary, but numerically evaluable, constraints. Finally, we critically evaluate the value of deriving optimal sample sizes under this framework as testable predictions for the experimental study of sampling behavior in DFE. (shrink)

How can we study bounded rationality? We answer this question by proposing rational task analysis —a systematic approach that prevents experimental researchers from drawing premature conclusions regarding the rationality of agents. RTA is a methodology and perspective that is anchored in the notion of bounded rationality and aids in the unbiased interpretation of results and the design of more conclusive experimental paradigms. RTA focuses on concrete tasks as the primary interface between agents and environments and requires explicating essential task elements, (...) specifying rational norms, and bracketing the range of possible performance, before contrasting various benchmarks with actual performance. After describing RTA’s core components we illustrate its use in three case studies that examine human memory updating, multitasking behavior, and melioration. We discuss RTA’s characteristic elements and limitations by comparing it to related approaches. We conclude that RTA provides a useful tool to render the study of bounded rationality more transparent and less prone to theoretical confusion. (shrink)

Perceptual Confidence is the view that our conscious perceptual experiences assign confidence. In previous papers, I motivated it using first-personal evidence (Morrison, 2016), and Jessie Munton motivated it using normative evidence (Munton, 2016). In this paper, I will consider the extent to which it is motivated by third-personal evidence. I will argue that the current evidence is supportive but not decisive. I will then describe experiments that might provide stronger evidence. I hope to thereby provide a roadmap for future research.

Recent developmental models assume a higher tendency to take risks in mid-adolescence, while the empirical evidence for this assumption is rather mixed. Most of the studies applied quite different tasks to measure risk-taking behavior and used a narrow age range. The main goal of the present study was to examine risk-taking behavior in four task settings, the Treasure Hunting Task (THT) in a gain and a loss domain, the Balloon Analogue Risk Task (BART) and the STOPLIGHT task. These task settings (...) differ in affective task moderators, like descriptive vs. experienced outcomes, anticipation of gains vs. losses, static vs. dynamic risk presentation and time pressure vs. no-time pressure and were applied in a sample of 187 participants from age 9 to 18. Beneath age trends, we were interested in their association with individual differences in approach behavior, venturesomeness, impulsivity and empathy above age, gender and fluid intelligence. Our findings revealed that risk-taking behavior is only low to moderately correlated between the four task contexts, suggesting that they capture different aspects of risk-taking behavior. Accordingly, a mid-adolescent peak in risk-propensity was only found under time pressure in the STOPLIGHT that was associated with higher impulsivity and empathy. In contrast, risky decisions decreased with increasing age in task settings, in which losses were anticipated (THT Loss), and this was associated with higher cognitive abilities. We found no age differences when gains were anticipated, neither in a static (THT Gain) nor in a dynamic task setting (BART). These findings clearly suggest the need to consider affective task moderators, as well as individual differences in temperament and cognitive abilities, in actual models about adolescent development. (shrink)

Despite the ubiquity of uncertainty, scientific attention has focused primarily on probabilistic approaches, which predominantly rely on the assumption that uncertainty can be measured and expressed numerically. At the same time, the increasing amount of research from a range of areas including psychology, economics, and sociology testify that in the real world, people’s understanding of risky and uncertain situations cannot be satisfactorily explained in probabilistic and decision-theoretical terms. In this article, we offer a theoretical overview of an alternative approach to (...) uncertainty developed in the framework of the ecological rationality research program. We first trace the origins of the ecological approach to uncertainty in Simon’s bounded rationality and Brunswik’s lens model framework and then proceed by outlining a theoretical view of uncertainty that ensues from the ecological rationality approach. We argue that the ecological concept of uncertainty relies on a systemic view of uncertainty that features it as a property of the organism–environment system. We also show how simple heuristics can deal with unmeasurable uncertainty and in what cases ignoring probabilities emerges as a proper response to uncertainty. (shrink)

In the last decade, a number of studies in the behavioural sciences, particularly in psychology and economics, have explored the complexity of individual risk behaviour and its underlying factors. Most previous studies have examined the influences of various socio-economic, cognitive, biological and psychological factors on human decision-making however, the relationship between the decision-makers’ risk preferences and occupational background has not received much empirical attention. Accordingly, in the current study, we investigated how occupational background, together with decision-making framing (e.g., variations in (...) decision domain, context, presentation of risk and utility ratios), influence participants’ risk preferences for decision options with equivalent expected utility. Our novel findings indicate that risk preferences may vary among individuals from different occupational backgrounds. As such, when the task was framed in gain terms, participants who mostly deal with health/safety-related risks on a day-to-day basis (high-risk occupations) were predominantly risk-averse (avoiding risky options), while participants who mostly deal with financial/social risks (white-collar occupations) were prone to risk-seeking behaviour (avoiding certain options). Specifically, in “high-risk” occupations, participants’ pattern of choices changed from risk-averse in gain scenarios to risk-seeking in loss scenarios. However, the opposite pattern of risk preferences was found in participants with “white-collar” occupations. Our findings indicate that decision-makers’ occupational backgrounds influence risk preferences under some circumstances. (shrink)

Most investigations into how people make risky choices have employed a simple drosophila: monetary gambles involving stated outcomes and probabilities. People are asked to make decisions from description . When people decide whether to back up their computer hard drive, cross a busy street, or go out on a date, however, they do not enjoy the convenience of stated outcomes and probabilities. People make such decisions either in the void of ignorance or in the twilight of their own often limited (...) experience of such real-world options. In the latter case, they make decisions from experience . Recent research has consistently documented that decisions from description and decisions from experience can lead to substantially different choices. Key in this description–experience gap is people’s treatment of rare events. In this paper, I briefly review studies that have documented the description–experience gap, offer several explanations for this gap, and discuss to what extent people’s decisions from experience are in conflict with benchmarks of rationality. (shrink)

The Iowa Gambling Task (IGT) is widely used to study decision‐making within healthy and psychiatric populations. However, the complexity of the IGT makes it difficult to attribute variation in performance to specific cognitive processes. Several cognitive models have been proposed for the IGT in an effort to address this problem, but currently no single model shows optimal performance for both short‐ and long‐term prediction accuracy and parameter recovery. Here, we propose the Outcome‐Representation Learning (ORL) model, a novel model that provides (...) the best compromise between competing models. We test the performance of the ORL model on 393 subjects' data collected across multiple research sites, and we show that the ORL reveals distinct patterns of decision‐making in substance‐using populations. Our work highlights the importance of using multiple model comparison metrics to make valid inference with cognitive models and sheds light on learning mechanisms that play a role in underweighting of rare events. (shrink)

The statistics of small samples are often quite different from those of large samples, and this needs to be taken into account in assessing the rationality of human behavior. Specifically, in evaluating human responses to environmental statistics, it is the effective environment that matters; that is, the environment actually experienced by the agent needs to be considered, not simply long‐run frequencies. Significant deviations from long‐run statistics may arise through experiential limitations of the agent that stem from resource constraints and/or information‐processing (...) bounds. The article draws together recent work from a number of areas in judgment and decision making ranging from randomness perception (Hahn & Warren, ), information sampling (Hertwig & Pleskac, ; Kareev et al., ), and consequences of choice for exploration or exploitation (e.g., Denrell, ) to demonstrate how proper consideration of these deviations leads to reevaluation of behaviors that are otherwise deemed irrational. (shrink)

If citizens’ behavior threatens to harm others or seems not to be in their own interest, it is not uncommon for governments to attempt to change that behavior. Governmental policy makers can apply established tools from the governmental toolbox to this end. Alternatively, they can employ new tools that capitalize on the wealth of knowledge about human behavior and behavior change that has been accumulated in the behavioral sciences. Two contrasting approaches to behavior change are nudge policies and boost policies. (...) These policies rest on fundamentally different research programs on bounded rationality, namely, the heuristics and biases program and the simple heuristics program, respectively. This article examines the policy–theory coherence of each approach. To this end, it identifies the necessary assumptions underlying each policy and analyzes to what extent these assumptions are implied by the theoretical commitments of the respective research program. Two key results of this analysis are that the two policy approaches rest on diverging assumptions and that both suffer from disconnects with the respective theoretical program, but to different degrees: Nudging appears to be more adversely affected than boosting does. The article concludes with a discussion of the limits of the chosen evaluative dimension, policy–theory coherence, and reviews some other benchmarks on which policy programs can be assessed. (shrink)

Our understanding of the decisions made under scenarios where both descriptive and experience-based information are available is very limited. Underweighting of small probabilities was observed in the gain domain when both description and experience were provided. The divergence observed from the prospect theory suggests a need for a separate or modified theory of decision making under risk. Recent studies suggest a possible role of probability weighting in the choice behaviour under risk. We investigated both gain and loss domains with and (...) without feedback for small and large probability conditions. We characterized the shape of the probability weighting function by a two-parameter functional form representing discriminability (concave-convex shape) and attractiveness (level of absolute weights relative to objective probability). We replicated four-fold pattern of risk attitude on non-WEIRD population. We find that feedback leads to underweighting of small probabilities and overweighting of large probabilities in the gain domain and overall underweighting of probabilities in the loss domain. We find that underweighting of small probabilities is driven by changes in discriminability and attractiveness components in the gain domain and changes in the attractiveness component in the loss domain. We have interpreted the results by proposing an updated belief-based account of decisions under uncertainty in which feedback, when available, influences the probability weighting mediating the choice behaviour. (shrink)

Topics in Cognitive Science, Volume 14, Issue 1, Page 14-30, January 2022.

A framing bias shows risk aversion in problems framed as “gains” and risk seeking in problems framed as “losses,” even when these are objectively equivalent and probabilities and outcomes values are explicitly provided. We test this framing bias in situations where decision makers rely on their own experience, sampling the problem's options and seeing the outcomes before making a choice. In Experiment 1, we replicate the framing bias in description-based decisions and find risk indifference in gains and losses in experience-based (...) decisions. Predictions of an Instance-Based Learning model suggest that objective probabilities as well as the number of samples taken are factors that contribute to the lack of framing effect. We test these two factors in Experiment 2 and find no framing effect when a few samples are taken but when large samples are taken, the framing effect appears regardless of the objective probability values. Implications of behavioral results and cognitive modeling are discussed. (shrink)