Hearing hooves, thinking zebras: A review of the inverse base-rate effect

Abstract

People often fail to use base-rate information appropriately in decision-making. This is evident in the inverse base-rate effect, a phenomenon in which people tend to predict a rare outcome for a new and ambiguous combination of cues. While the effect was first reported in 1988, it has recently seen a renewed interest from researchers concerned with learning, attention and decisionmaking. However, some researchers have raised concerns that the effect arises in specific circumstances and is unlikely to provide insight into general learning and decision-making processes. In this review, we critically evaluate the evidence for and against the main explanations that have been proposed to explain the effect, and identify where this evidence is currently weak. We argue that concerns about the effect are not well supported by the data. Instead, the evidence supports the conclusion that the effect is a result of general mechanisms that provides a useful opportunity to understand the processes involved in learning and decision making. We discuss gaps in our knowledge and some promising avenues for future research, including the relevance of the effect to models of attentional change in learning, an area where the phenomenon promises to contribute new insights.

A cognitive category-learning model of rule abstraction, attention learning, and contextual modulation.

“…The theoretical framework that guided the implementation of CAL’s hypotheses is applicable beyond the paradigms considered in this article. For instance, the hypotheses of attention-driven learning and contextual modulation could predict the frequently studied phenomenon known as the inverse base-rate effect (IBRE; Medin & Edelson, 1988; for a comprehensive review, see Don et al, 2021). Consider a learning phase that pairs singleton stimuli (e.g., S1 or S2) with outcomes (e.g., O1 or O2), presenting S1 and S2 → O1 three times and S1 and S3 → O2 once.…”

“…The theoretical framework that guided the implementation of CAL's hypotheses is applicable beyond the paradigms considered in this article. For instance, the hypotheses of attention-driven learning and contextual modulation could predict the frequently studied phenomenon known as the inverse base-rate effect (Medin & Edelson, 1988; for a comprehensive review see Don, Worthy, & Livesey, 2021 The question of whether modulators (e.g., S3) may generalize to correlated dimensions is speculative at this stage. However, it has been argued that probabilistic errors (e.g., defined as S1 leads to O1 in 75% of the cases) trigger attention to correlated stimulus features (Little & Lewandowsky, 2009a).…”

g-distance: On the comparison of model and human heterogeneity

“…Although multiple published replications of the inverse base-rate effect exist (see the review of Don et al, 2021), none were suitable for the current application. This was for two reasons.…”

“…However, when participants are sequentially trained and tested on a combination of conflicting cues, the group-level result is that participants prefer the rare category (Kruschke, 1996). The IBRE has been observed in many experiments across a range of G-DISTANCE 20 scenarios (e.g., Shanks, 1992;Don & Livesey, 2017;Inkster, Milton, Edmunds, Benattayallah, & Wills, 2021;Inkster, Mitchell, Schlegelmilch, & Wills, 2022;Johansen, Fouquet, & Shanks, 2007, 2010Kalish, 2001;Sherman et al, 2009;Wills, Lavric, Croft, & Hodgson, 2007); for an excellent recent review, see Don, Worthy, and Livesey (2021).…”

Effect of a context shift on the inverse base-rate effect

“…The Inverse Base Rate Effect (IBRE; Medin & Edelson, 1988) is a non-rational learning phenomenon that has generated considerable debate within the literature (Bohil, Markman, & Maddox, 2005;Don, Worthy, & Livesey, 2021;Kruschke, 1996Kruschke, , 2001bKruschke, , 2003Winman, Wennerholm, & Juslin, 2003). In its canonical form, participants are asked to diagnose fictitious patients under a simulated medical diagnosis procedure.…”