Search for: All records

Abstract Framing effects play a central role in the debate regarding human rationality. They violate the normative principle ofdescription invariance, which states that merely redescribing options or outcomes in equivalent ways should not affect judgments or decisions. Description invariance is considered by many decision researchers to be “normatively unassailable”, and violations are widely regarded as demonstrations of systematic irrationality. This article develops an alternative perspective on invariance violations, applying Funder’s (1987) distinction between “errors” and “mistakes”. Description invariance implicitly assumes that (1) rational preferences must be complete and (2) frames do not convey choice-relevant information. We argue that both assumptions often do not hold. When they fail, framing effects in the laboratory are not “errors”, and they do not provide evidence for “mistakes” in natural environments. Furthermore, recent findings suggest that participants often do not regard different responses to different frames as unreasonable, and presenting them with arguments for and against description invariance has little effect on their views. Finally, we argue that similar lessons generalize to other coherence norms, such as procedure invariance and independence of irrelevant alternatives. 

Background and aimCitations in academia have long been regarded as a fundamental means of acknowledging the contribution of past work and promoting scientific advancement. The aim of this paper was to investigate the impact that misconduct allegations made against scholars have on the citations of their work, comparing allegations of sexual misconduct (unrelatedto the research merit) and allegations of scientific misconduct (directly relatedto the research merit). MethodsWe collected citation data from the Web of Science (WoS) in 2021, encompassing 31,941 publications from 172 accused and control scholars across 18 disciplines. We also conducted two studies: one on non-academics (N = 231) and one on academics (N = 240). ResultsThe WoS data shows that scholars accused of sexual misconduct incur a significant citation decrease in the three years after the accusations become public, while we do not detect a significant citation decrease for scholars accused of scientific misconduct. The study involving non-academics suggests that individuals are more averse to sexual than to scientific misconduct. Finally, contrary to the WoS data findings, a sample of academics indicates they are more likely to cite scholars accused of sexual misconduct than those accused of scientific misconduct. ConclusionsIn the first three years after accusations became public, scholars accused of sexual misconduct incur a larger citation penalty than scholars accused of scientific misconduct. However, when asked to predict their citing behavior, scholars indicated the reverse pattern, suggesting they might mis-predict their behavior or be reluctant to disclose their preferences. 

Traditional models of rational choice assume that preferences are complete, but the completeness axiom is neither normatively compelling nor psychologically plausible. Building on recent work in economics, we develop a rational analysis of decision making with incomplete preferences. The analysis sheds surprising light on a range of well-known behavioral “anomalies,” including the endowment effect, status quo maintenance, the sunk cost effect, and coherent arbitrariness. We propose a two-part division of rational choice theory—into preference theory and “implementation theory”—and show how conservative and coherently arbitrary policies can effectively implement incomplete preferences. The two-part normative framework motivates a psychological distinction between evaluation and implementation phases in decision making. We argue that the endowment effect and related phenomena, which have usually been attributed to loss aversion in the evaluation phase, are better explained by conservatism in the implementation phase. The rational analysis challenges the normative adequacy of expected utility theory and raises questions about the explanatory scope of prospect theory. It illustrates the rich interplay between psychological models of value structure and normative models of rational choice. 

A great deal of current research on moral judgments centers on moral dilemmas concerning tradeoffs between one and five lives. Whether one considers killing one innocent person to save five others to be morally required or impermissible has been taken to determine whether one is appealing to consequentialist or non-consequentialist reasoning. But this focus on tradeoffs between one and five may obscure more nuanced commitments involved in moral decision-making that are revealed when the numbers and ratio of lives to be traded off are varied, and when the probabilities of each outcome occurring are less than certain. Four studies examine participants’ reactions to scenarios that diverge in these ways from the standard ones. Study 1 examines the extent to which people are sensitive to the ratio of lives saved to lives ended by a particular action. Study 2 verifies that the ratio rather than the difference between the two values is operative. Study 3 examines whether participants treat probabilistic harm to some as equivalent to certainly harming fewer, holding expected ratio constant. Study 4 explores an analogous issue regarding the sensitivity of probabilistic saving. Participants are remarkably sensitive to expected ratio for probabilistic harms while deviating from expected value for probabilistic saving. Collectively, the studies provide evidence that people’s moral judgments are consistent with the principle of threshold deontology. 

Although all birth orders in the “birth sequence problem” are equiprobable, most participants judge the less representative order as less likely than the more representative order. But this well-known problem confounds representativeness with the direction in which birth orders are compared. We hypothesized and corroborated in three experiments (total N = 1,136) that participants pragmatically infer the birth orders’ relative prevalence from the direction of comparison. Experiment 1 found that participants judged the less representative sequence as more common when we reversed the comparison. Experiment 2 reproduced these results despite removing representativeness as a cue. In Experiment 3, participants preferred to place the relatively common sequence as the referent in an inverted “speaker” problem. Our results turn the iconic problem’s interpretation on its head: Rather than indicating flawed human cognition, the birth sequence problem illustrates people’s ability to adaptively extract subtle linguistic meaning beyond the literal content. 

Human decisions are context dependent in ways that violate classical norms of rational choice. However, these norms implicitly depend on idealized descriptive assumptions that are often unrealistic. We focus on one such assumption: that information is constant across contexts. Choice contexts often supply subtle cues—which may be embedded in frames, procedures, or menus—to which human decision makers can be highly sensitive. We review recent evidence that some important context effects reflect dynamically coherent belief and preference updating, in response to ecologically valid cues. This evidence paints a more nuanced picture of human rationality in natural choice environments and opens up prospects for nonpaternalistic forms of choice architecture. 

The normative principle of description invariance presupposes that rational preferences must be complete. The completeness axiom is normatively dubious, however, and its rejection opens the door to rational framing effects. In this commentary, we suggest that Bermúdez’s insightful challenge to the standard normative view of framing can be clarified and extended by situating it within a broader critique of completeness.