Risk aversion for moderate-likelihood gains is perhaps the best-known stylized fact from decision research. Though studies documenting it have focused on decisions involving money, such behavior is presumed to prevail very generally, across diverse domains. We investigate the validity of this generalization by contrasting two types of decisions. In unimodal choices, outcomes are "apples-to-apples." Consider choosing between sure and uncertain monetary payoffs, or between sure receipt of a product and a chance at several units of it. In crossmodal choices, outcomes are "apples-to-oranges." Consider choosing between sure receipt of one product and uncertain receipt of a disparate product. We observe two patterns by which risk matters less crossmodally, contrary to straightforward generalizations. First, relative to unimodal preferences involving actuarially fair risky options, corresponding crossmodal preferences exhibit less risk aversion. Second, crossmodal preferences vary less across risk levels: as the likelihood and subjective value of a risky option's outcomes become increasingly unfavorable (favorable), people do not exhibit as much additional distaste (appetite) for it. These patterns of insensitivity engender an interaction: relative to unimodal settings, crossmodal settings yield less aversion to unfavorable and fair risk but more aversion to favorable risk. To explain this interaction, we present the translate-and-accommodate (TAA) model, in which unimodal preferences follow standard accounts, but crossmodal preferences reflect different processes of (i) deterministic translation and (ii) risk accommodation. The TAA model also explains the uncertainty effect and related patterns of seemingly bizarre, dominated choices.

We present and test a model of reciprocity in which people are more likely to repay good treatment to the extent they judge it as motivated by true caring rather than tactical self-interest. The model's key contributions stem from how it handles ambiguously motivated behavior. It allows people to maintain divergent hypotheses: They can view behavior as driven by caring, self-interest, or a mix thereof. In contrast, previous analyses resolve rather than maintain ambiguity. They treat caring and self-interest as mutually exclusive hypotheses, and require that people commit to one and dismiss the other. By more realistically handling ambiguity, our model yields three benefits. First, it accommodates intuitive patterns of play that existing analyses do not and which we experimentally corroborate. These patterns reflect intermediate inclinations to reciprocate ambiguously motivated positive behavior. Second, it challenges conventional interpretations of long-studied phenomena, including unraveling in finitely iterated prisoners' dilemmas, substantial offers in ultimatum games, and gift exchange. Third, it highlights how diversity in perceptions – the same action can appear generous to one person and miserly to another – is empirically consequential. Under conventional interpretations and without accounting for diverse perceptions, the aforementioned phenomena have been viewed as inconsistent with a taste for repaying good treatment. Our model shows that they are entirely consistent with a nuanced form of this taste: a desire to repay good treatment that seems to largely reflect genuine caring.

Are people skillful in utilizing potential focal points? We find a class of situations for which the answer is negative: the presence of prominent actions appears to stymie the use of distinct actions for coordination. Across several experimental games, we consistently observe that players readily coordinate on a categorically distinct action when all available actions are non-prominent but not when some actions are prominent. For instance, given the action set {Franklin Pierce, James Buchanan, Tianjin}, most players select the Chinese city Tianjin. Yet, given {Abraham Lincoln, George Washington, Tianjin}, they are roughly equally likely to select either American president and unlikely to select Tianjin, and given {Abraham Lincoln, George Washington, Shanghai}, their choices are distributed approximately uniformly. The observation that prominence stymies reliance on distinctiveness informs cognitive hierarchy and team reasoning theories of how people recognize focality.

Decision research often takes the variability of potential outcomes as a measure of risk. It thus characterizes sure things, which, by definition, guarantee a specific outcome, as safe. But is this characterization always empirically valid? We show that, when the prevailing reference point is an uncertain option or position, sure things can be perceived as risky rather than safe. Furthermore, preferences may hinge on such perceptions: when construed as risky, sure things can be less appealing. Our findings suggest a perception-based explanation for why the classic tendency to favor sure things over uncertain options is often attenuated given an uncertain reference point. More broadly, they tap an unresolved debate about the determinants of decisions. Much research focuses on taste-based determinants, such as attitudes toward perceived risks, and thereby underplays the critical role of the underlying perceptions. Appreciating the impact of perceptions on decision making leads to novel prescriptive recommendations.

Since Schelling, it has often been assumed that players make use of salient decision labels to achieve coordination. Consistent with previous work, we find that given equal payoffs, salient labels yield frequent coordination. However, given even minutely asymmetric payoffs, labels lose much of their effectiveness and miscoordination abounds. This raises questions about the extent to which the effectiveness of focal points based on label salience persists beyond the special case of symmetric games. The patterns of miscoordination we observe vary with the magnitude of payoff differences in intricate ways that suggest nonequilibrium accounts based on "level-k" thinking and "team reasoning."

Prospect theory's S-shaped weighting function is often said to reflect the psychophysics of chance. We propose an affective rather than psychophysical deconstruction of the weighting function resting on two assumptions. First, preferences depend on the affective reactions associated with potential outcomes of a risky choice. Second, even with monetary values controlled, some outcomes are relatively affect-rich and others relatively affect-poor. Although the psychophysical and affective approaches are complementary, the affective approach has one novel implication: Weighting functions will be more S-shaped for lotteries involving affect-rich than affect-poor outcomes. That is, people will be more sensitive to departures from impossibility and certainty but less sensitive to intermediate probability variations for affect-rich outcomes. We corroborated this prediction by observing probability-outcome interactions: An affect-poor prize was preferred over an affect-rich prize under certainty, but the direction of preference reversed under low probability. We suggest that the assumption of probability-outcome independence, adopted by both expected-utility and prospect theory, may hold across outcomes of different monetary values, but not different affective values.

How does the attractiveness of a particular option depend on comparisons drawn between it and other alternatives? We observe that in many cases, comparisons hurt; When the options being compared have both meaningful advantages and meaningful disadvantages, comparison between options makes each option less attractive. The effects of comparison are crucial in choice problems involving grouped options, because the way in which options are grouped influences which comparisons are likely to be made. In particular, we propose that grouping focuses comparison, making within-group comparisons more likely than between-group comparisons. This line of reasoning suggests that grouping should hurt, and we observe that it does: An option is more likely to be chosen when alone than when part of a group.

Support theory represents probability judgment in terms of the support, or strength of evidence, of the focal relative to the alternative hypothesis. It assumes that the judged probability of an event generally increases when its description is unpacked into disjoint components (implicit subadditivity). This article presents a significant extension of the theory in which the judged probability of an explicit disjunction is less than or equal to the sum of the judged probabilities of its disjoint components (explicit subadditivity). Several studies of probability and frequency judgment demonstrate both implicit and explicit subadditivity. The former is attributed to enhanced availability, whereas the latter is attributed to repacking and anchoring.