Search

Simple, transparent rules are often frowned upon while complex, black-box models are seen as holding greater promise. Yet in quickly changing situations, simple rules can protect against overfitting and adapt quickly. We show that the surprisingly simple recency heuristic forecasts more accurately than Google Flu Trends (GFT) which used big data analytics and a black-box algorithm. This heuristic predicts that “this week’s proportion of flu-related doctor visits equals the proportion from the most recent week.” It is based on psychological theory of how people deal with rapidly changing situations. Other theory-inspired heuristics have outperformed big data models in predicting outcomes, such as U.S. presidential elections, or other uncertain events, such as consumer purchases, patient hospitalizations, and terrorist attacks. Heuristics are transparent, clearly communicating the underlying rationale for their predictions. We advocate taking into account psychological principles that have evolved over millennia and using these as a benchmark when testing big data models.

Distinguishing between risk and uncertainty, this article draws on the psychological literature on heuristics to consider whether and when simpler approaches may outperform more complex methods for modeling and regulating the financial system. We find that: simple methods can sometimes dominate more complex modeling approaches for calculating banks’ capital requirements, especially when data are limited or underlying risks are fat-tailed; simple indicators often outperformed more complex metrics in predicting individual bank failure during the global financial crisis; when combining different indicators to predict bank failure, simple and easy-to-communicate “fast-and-frugal” decision trees can perform comparably to standard, but more information-intensive, regressions. Taken together, our analyses suggest that because financial systems are better characterized by uncertainty than by risk, simpler approaches to modeling and regulating financial systems can usefully complement more complex ones and ultimately contribute to a safer financial system.

We structure this response to the commentaries to our article “Transparent modeling of influenza incidence: Big data or a single data point from psychological theory?” around the concept of psychological AI, Herbert Simon's classic idea of using insights from how people make decisions to make computers smart. The recency heuristic in Katsikopoulos, Şimşek, Buckmann, and Gigerenzer (2021) is one example of psychological AI. Here we develop another: the trend-recency heuristic. While the recency heuristic predicts that the next observation will equal the most recent observation, the trend-recency heuristic predicts that the next trend will equal the most recent trend. We compare the performance of these two recency heuristics with forecasting models that use trend damping for predicting flu incidence. Psychological AI prioritizes ecological rationality and transparency, and we provide a roadmap of how to study such issues. We also discuss how this transparency differs from explainable AI and how ecological rationality focuses on the comparative empirical study and theoretical analysis of different types of models.

The five of us research and teach risk analysis with an eye towards decision support. Our work has been dedicated to taming risks and helping to make challenging decisions. But nothing had prepared us for the Covid-19 pandemic. We first had to grapple with the news coming from abroad, including, for some of us, our home countries. Then, some information and research, but mostly opinions, started coming in from our academic community, and we felt the tensions. Finally, the UK went into an unofficial and then official lockdown, and all University staff were asked to redirect their research capacity so as to support the national effort for risk analysis and decision support. As we write this on the 20th of April, many countries, including the UK, are starting to consider how to get out of lockdown. Like the previous stages of the pandemic, there is little data, perhaps a bit more research, surely many more opinions, and definitely an overwhelming amount of personal experiences and thoughts. Here we reflect on all of the above, just in case it can help the readers of this Minds in Society flash editorial to think and act, or at least, to not have to do so entirely on their own. As it can be expected, our collage introduces more questions than it can answer.

Rules for building formal models that use fast-and-frugal heuristics, extending the psychological study of classification to the real world of uncertainty. This book focuses on classification—allocating objects into categories—“in the wild,” in real-world situations and far from the certainty of the lab. In the wild, unlike in typical psychological experiments, the future is not knowable and uncertainty cannot be meaningfully reduced to probability. Connecting the science of heuristics with machine learning, the book shows how to create formal models using classification rules that are simple, fast, and transparent and that can be as accurate as mathematically sophisticated algorithms developed for machine learning. The authors—whose individual expertise ranges from empirical psychology to mathematical modeling to artificial intelligence and data science—offer real-world examples, including voting, HIV screening, and magistrate decision making; present an accessible guide to inducing the models statistically; compare the performance of such models to machine learning algorithms when applied to problems that include predicting diabetes or bank failure; and discuss conceptual and historical connections to cognitive psychology. Finally, they analyze such challenging safety-related applications as decreasing civilian casualties in checkpoints and regulating investment banks.

PurposeThis research explores how investment and central bankers cope with strategic uncertainty when they anticipate prices. The uncertainty originates from others' decisions and their consequences, and cannot be meaningfully reduced to risk. The authors postulate that, in order to cope with this type of uncertainty, bankers use simple rules, also called heuristics. This study aims to identify such heuristics and the psychological processes that underlie them.Design/methodology/approachThe authors interviewed 22 managers of teams tasked to anticipate prices, in two leading investment and central banks. The primary data came from in-depth, semi-structured interviews lasting 30–60 min, supplemented by our observations during the on-site visits, emails and phone calls when preparing the interviews, and reports published by the banks. Data were coded and heuristics were induced over multiple rounds by multiple researchers.FindingsBankers (1) construct simple game representations of markets, (2) make inferences to gauge opponents, (3) become alert when they see too much agreement and (4) communicate coherent narratives. Heuristics (1)–(3) are employed when the pace of decision-making is fast, whereas (4) is used for longer time scales. In sum, bankers exhibit reciprocal bounded rationality, wherein interaction partners are mutually aware of and adapted to the fundamental uncertainty of the task and their limited resources.Originality/valueHeuristics for anticipating prices have not been studied empirically outside the lab. The findings may help integrate conceptualizations of heuristics in the simple-rules and fast-and-frugal-heuristics research programs and improve market efficiency.

This paper presents a survey of the literature on the application of Operational Research (OR) in healthcare, with a particular focus on behavioural considerations. In order to explore the extent to which behavioural aspects are included, we perform a search of the most relevant OR journals for articles with content related to the representation of behaviour in models, evidence of behavioural change using models, and the impact on organisations beyond the use of a model. A detailed analysis of 130 articles is presented and shows that the majority are focused on improving service delivery at an organisational level. The most common OR methods depicting behaviour are simulation and qualitative methods, but there is evidence of the use across a range of methods. However, in many cases, authors do not necessarily acknowledge the behavioural aspects in their papers. Given many aspects of healthcare are influenced by human behaviour, it is important that that future work makes more explicit the assumptions used to represent behaviour, test the sensitivity of models to different behavioural assumptions, and offer more information about how users employ models to make decisions.

Models offer many services, of which ‘intermediate services’ are often neglected but nevertheless, we will argue, very useful for Operational Research. Examples of intermediate services of models are to provide insights, platforms for further discussion, and coherent stories, all of which allow practical and theoretical work to accumulate and continue. We argue that models which are transparent are more likely to deliver useful intermediate services than models which appear to be black boxes to practitioners or academics. The argument entails two points: (1) a definition of model transparency in behavioral sciences such as economics and psychology and its adjustment for Operational Research and (2) a discussion of some particular applications of behavioral models in Operational Research, which reveals the merits of transparency.

Achievements in various fields of creativity are resulting more and more from collaborative teams. This research investigated the role of interpersonal process variables, namely closeness and perspective taking in group creativity, with a 2 by 2 experimental design. Sixty-one 3-person groups assigned to 4 conditions (a: closeness and perspective taking, b: perspective taking, c: no closeness and no perspective taking, d: closeness). Group members collaboratively wrote stories that were rated by 3 independent expert judges. There was a positive main effect of closeness and negative main effect of perspective taking on group creativity scores. Moreover, the significant interaction between perspective taking and closeness displayed that combination of closeness with perspective taking negatively affect group creativity. These results indicate that closeness might be beneficial for group creativity only when it is not accompanied with perspective taking.

Many decisions can be analyzed and supported by quantitative models. These models tend to be complex psychologically in that they require the elicitation and combination of quantities such as probabilities, utilities, and weights. They may be simplified so that they become more transparent, and lead to increased trust, reflection, and insight. These potential benefits of simplicity should be weighed against its potential costs, notably possible decreases in performance. We review and synthesize research that has used mathematical analyses and computer simulations to investigate if and when simple models perform worse, equal, or better than more complex models. Various research strands have pursued this, but have not reached the same conclusions: Work on frequently repeated decisions as in inference and forecasting—which typically are operational and involve one or a few decision makers—has put forth conditions under which simple models are more accurate than more complex ones, and some researchers have proposed that simple models should be preferred. On the other hand, work on more or less one-off decisions as in preference and multi-criteria analysis—which typically are strategic and involve group decision making and multiple stakeholders—has concluded that simple models can at best approximate satisfactorily the more complex models. We show how these conclusions can be reconciled. Additionally, we discuss the theory available for explaining the relative performance of simple and more complex models. Finally, we present an aid to help determine if a simple model should be used, or not, for a particular type of decision problem.

Nudge and boost are two competing approaches to applying the psychology of reasoning and decision making to improve policy. Whereas nudges rely on manipulation of choice architecture to steer people towards better choices, the objective of boosts is to develop good decision-making competences. Proponents of both approaches claim capacity to enhance social welfare through better individual decisions. We suggest that such efforts should involve a more careful analysis of how individual and social welfare are related in the policy context. First, individual rationality is not always sufficient or necessary for improving collective outcomes. Second, collective outcomes of complex social interactions among individuals are largely ignored by the focus of both nudge and boost on individual decisions. We suggest that the design of mechanisms and social norms can sometimes lead to better collective outcomes than nudge and boost, and present conditions under which the three approaches (nudge, boost, and design) can be expected to enhance social welfare.

The search for parking space in busy urban districts is one of those routine human activities that are expected to benefit from the widespread adoption of pervasive sensing and radio communication technologies. Proposed parking assistance solutions combine sensors, either as part of fixed infrastructure or onboard vehicles, wireless networking technologies and mobile social applications running on smartphones to collect, share and present to drivers real-time information about parking demand and availability. One question that arises is how does (and should) the driver actually use such information to take parking decisions, e.g. , whether to search for on-street parking space or drive to a parking lot and, in the latter case, which one. The paper is, hence, a performance analysis study that seeks to capture the highly behavioral and heuristic dimension of drivers’ decisions and its impact on the efficiency of the parking search process. To this end, and in sharp contrast with the existing literature, we model drivers as agents of bounded rationality and assume that their choices are directed by lexicographic heuristics , an instance of the fast and frugal heuristics developed in behavioral sciences such as psychology and biology. We analyze the performance of the search process under these heuristics and compare it against the predictions of normative game-theoretic models that assume fully rational strategically acting agents. We derive conditions under which the game-theoretic norms turn out to be more pessimistic than the simpler heuristic choice rules and show that these are fulfilled for a broad range of scenarios concerning the fees charged for the parking resources and their distance from the destinations of the drivers’ trips. The practical implications of these results for parking assistance solutions are identified and thoroughly discussed.

Medical decision-making is a complex process that often takes place during uncertainty, that is, when knowledge, time, and resources are limited. How can we ensure good decisions? We present research on heuristics-simple rules of thumb-and discuss how medical decision-making can benefit from these tools. We challenge the common view that heuristics are only second-best solutions by showing that they can be more accurate, faster, and easier to apply in comparison to more complex strategies. Using the example of fast-and-frugal decision trees, we illustrate how heuristics can be studied and implemented in the medical context. Finally, we suggest how a heuristic-friendly culture supports the study and application of heuristics as complementary strategies to existing decision rules.

Research on bounded rationality has two cultures, which I call ‘idealistic’ and ‘pragmatic’. Technically, the cultures differ on whether they (1) build models based on normative axioms or empirical facts, (2) assume that people's goal is to optimize or to satisfice, (3) do not or do model psychological processes, (4) let parameters vary freely or fix them, (5) aim at explanation or prediction and (6) test models from one or both cultures. Each culture tells a story about people's rationality. The story of the idealistic culture is frustrating, with people in principle being able to know what they should do, but in practice systematically failing to do it. This story makes one hide in books for intellectual solace or surrender to the designs of someone smarter. The story of the pragmatic culture is empowering: If people are educated to use the right tool in the right situation, they do well.

This paper presents an axiomatic framework for the priority heuristic, a model of bounded rationality in Selten’s (in: Gigerenzer and Selten (eds.) Bounded rationality: the adaptive toolbox, 2001) spirit of using empirical evidence on heuristics. The priority heuristic predicts actual human choices between risky gambles well. It implies violations of expected utility theory such as common consequence effects, common ratio effects, the fourfold pattern of risk taking and the reflection effect. We present an axiomatization of a parameterized version of the heuristic which generalizes the heuristic in order to account for individual differences and inconsistencies. The axiomatization uses semiorders (Luce, Econometrica 24:178–191, 1956), which have an intransitive indifference part and a transitive strict preference component. The axiomatization suggests new testable predictions of the priority heuristic and makes it easier for theorists to study the relation between heuristics and other axiomatic theories such as cumulative prospect theory.

Whether to forgive is a key decision supporting cooperation. Like many other evolutionarily recurrent decisions, it is made under uncertainty and requires the trade-off of costs and benefits. This decision can be conceptualized as a signal detection or error management task: Forgiving is adaptive if a relationship with the "harmdoer" will be fitness enhancing and not adaptive if the relationship will be fitness reducing, and the decision should be biased toward lowering the likelihood of the more costly error, which depending on the context may be either erroneously not forgiving or forgiving. Building on such conceptualization, we developed two cognitive models and examined how well they described participants' forgiveness decisions in hypothetical scenarios and predicted their decisions in recalled real-life incidents. We found that the models performed similarly and generally well—around 80% in describing and 70% in prediction. Moreover, this modeling approach allowed us to estimate the decision bias of each participant; we found that the biases were generally consistent with the prescriptions of signal detection theory and were directed at reducing the more costly error. In addition to testing mechanistic models of the forgiveness decision, our study also contributes to forgiveness research by applying a novel experimental method that studied both hypothetical and real-life decisions in tandem. These models and experimental methods could be used to study other evolutionarily recurrent problems, advancing understanding of how they are solved in the mind.