Human behavioral risk-seeking tendencies differ across content domains. How can such behavioral differences be reliably produced by the cognitive system? This article presents an explorative analysis of the reasons for and cognitive mechanisms underlying different risk propensities across 10 evolutionary domains. We investigate three cognitive process models: Tally, Take The First, and Most Relevant. Tally assumes decision-makers use majority rules. Take The First assumes decision makers rely on the first piece of information that comes to mind. Most Relevant assumes decision makers rely on information that is important in their environment. A survey with a total of N ϭ 120 individuals in the United States gathered 1,598 self-reported memory-based attributes of risky situations in 10 evolutionary content domains. The explorative analysis of the cognitive processes underlying the domain differences suggest that the Most Relevant strategy is most closely related to the shifts in risk seeking across content domains, and that Take The First is also related, but the Tally process is not related to domain differences in risk propensities. This means that a cognitive process that relies on the first or frequent pieces of information from the environment may be underlying domain differences in risk taking.

Humans excel in categorization. Yet from a computational standpoint, learning a novel probabilistic classification task involves severe computational challenges. The present paper investigates one way to address these challenges: assuming class-conditional independence of features. This feature independence assumption simplifies the inference problem, allows for informed inferences about novel feature combinations, and performs robustly across different statistical environments. We designed a new Bayesian classification learning model (the dependence-independence structure and category learning model, DISC-LM) that incorporates varying degrees of prior belief in class-conditional independence, learns whether or not independence holds, and adapts its behavior accordingly. Theoretical results from two simulation studies demonstrate that classification behavior can appear to start simple, yet adapt effectively to unexpected task structures. Two experiments-designed using optimal experimental design principles-were conducted with human learners. Classification decisions of the majority of participants were best accounted for by a version of the model with very high initial prior belief in class-conditional independence, before adapting to the true environmental structure. Class-conditional independence may be a strong and useful default assumption in category learning tasks.

The term process model is widely used, but rarely agreed upon. This paper proposes a framework for characterizing and building cognitive process models. Process models model not only inputs and outputs but also model the ongoing information transformations at a given level of abstraction. We argue that the following dimensions characterize process models: They have a scope that includes different levels of abstraction. They specify a hypothesized mental information transformation. They make predictions not only for the behavior of interest but also for processes. The models’ predictions for the processes can be derived from the input, without reverse inference from the output data. Moreover, the presumed information transformation steps are not contradicting current knowledge of human cognitive capacities. Lastly, process models require a conceptual scope specifying levels of abstraction for the information entering the mind, the proposed mental events, and the behavior of interest. This framework can be used for refining models before testing them or after testing them empirically, and it does not rely on specific modeling paradigms. It can be a guideline for developing cognitive process models. Moreover, the framework can advance currently unresolved debates about which models belong to the category of process models.

Digital contact-tracing applications (DCTAs) can help control the spread of epidemics, such as the coronavirus disease 2019 pandemic. But people in Western societies fail to install DCTAs. Understanding the low use rate is key for policy makers who support DCTAs as a way to avoid harsh nationwide lockdowns. In a preregistered study in a representative German-speaking Swiss sample (N = 757), the roles of individual risk perceptions, risk preferences, social preferences, and social values in the acceptance of and compliance with DCTA were compared. The results show a high compliance with the measures recommended by DCTAs but a comparatively low acceptance of DCTAs. Risk preferences and perceptions, but not social preferences, influenced accepting DCTAs; a high health-risk perception and a low data-security-risk perception increased acceptance. Additionally, support of political measures, technical abilities, and understanding the DCTA functionality had large effects on accepting DCTAs. Therefore, we recommend highlighting personal health risks and clearly explaining DCTAs, focusing on data security, to enhance DCTA acceptance.