Base Rate Information Research Articles

Adjusting Inferential Thresholds to Reflect Nonepistemic Values

Many philosophers have challenged the ideal of value-free science on the grounds that social or moral values are relevant to inferential thresholds. But given this view, how precisely and to what extent should scientists adjust their inferential thresholds in light of nonepistemic values? We suggest that signal detection theory provides a useful framework for addressing this question. Moreover, this approach opens up further avenues for philosophical inquiry and has important implications for philosophical debates concerning inductive risk. For example, the signal detection theory framework entails that considerations of inductive risk and inferential-threshold placement cannot be conducted in isolation from base-rate information.

Advancing the specification of dual process models of higher cognition: a critical test of the hybrid model view

Dual process models of higher cognition have become very influential in the cognitive sciences. The popular Default-Interventionist model has long favoured a serial view on the interaction between intuitive and deliberative processing (or System 1 and System 2). Recent work has led to an alternative hybrid model view in which people’s intuitive reasoning performance is assumed to be determined by the absolute and relative strength of competing intuitions. In the present study, we tested unique new predictions to validate the hybrid model. We adopted a two-response paradigm with popular base-rate neglect problems in which base-rate information and a stereotypical description could cue conflicting responses. By manipulating the extremity of the base-rates in our problems we aimed to affect the strength of the “logical” intuition that is hypothesised to cue selection of the base-rate response. The two-response paradigm – in which people were required to give an initial response under time-pressure and cognitive load – allowed us to identify the presumed intuitively generated response. Consistent with the hybrid model predictions, we observed that experimentally reducing the strength of the logical intuition decreased the number of initial base-rate responses when solving problems in which base-rates and stereotypical information conflicted. Critically, reasoners who gave an initial stereotypical response were less likely to register the intrinsic conflict (as reflected in decreased confidence) in this case, whereas reasoners who gave an initial base-rate response registered more conflict. Implications and remaining challenges for dual process theorising are discussed.

Model-based fMRI reveals dissimilarity processes underlying base rate neglect.

Extensive evidence suggests that people use base rate information inconsistently in decision making. A classic example is the inverse base rate effect (IBRE), whereby participants classify ambiguous stimuli sharing features of both common and rare categories as members of the rare category. Computational models of the IBRE have posited that it arises either from associative similarity-based mechanisms or from dissimilarity-based processes that may depend on higher-level inference. Here we develop a hybrid model, which posits that similarity- and dissimilarity-based evidence both contribute to the IBRE, and test it using functional magnetic resonance imaging data collected from human subjects completing an IBRE task. Consistent with our model, multivoxel pattern analysis reveals that activation patterns on ambiguous test trials contain information consistent with dissimilarity-based processing. Further, trial-by-trial activation in left rostrolateral prefrontal cortex tracks model-based predictions for dissimilarity-based processing, consistent with theories positing a role for higher-level symbolic processing in the IBRE.

The development of the representativeness heuristic in young children

In classic examinations of the representativeness heuristic, Kahneman and Tversky (1973) presented adult participants with a description of an individual who fit their stereotype of a typical engineer. Importantly, even when participants were told that the individual was drawn from a sample of 70 lawyers and 30 engineers, they estimated that the individual was an engineer at very high levels, showing that they relied almost exclusively on the personality description. Relying on the representativeness heuristic can lead to base-rate neglect and, thus, biased judgments. Two experiments provide insight into the development of the representativeness heuristic in young children using an adaptation of the classic lawyer–engineer problem. Experiment 1 (N = 96) established that 3- to 5-year-olds can use base-rate information on its own, and 4- and 5-year-olds can use individuating information on its own, to make inferences. Experiment 2 (N = 192) varied the relevance of the individuating information across conditions to assess the pervasiveness of this bias early in development. Here 5- and 6-year-olds, much like adults, continue to attempt to rely on individuating information when making classifications even if that information is irrelevant. Together, these experiments reveal how the representativeness heuristic develops across the preschool years and suggest that the bias may strengthen between 4 and 6 years of age.

Perceived Risk of Emerging Recreational Drugs: Impact of Anecdotal and Statistical Evidence

This study investigated the relative impact of personal stories and base rate evidence on the perceived risk of using two emerging recreational drugs: kratom and Spice. A 3 × 2 × 2 mixed-methods design was employed. Four hundred fifty-three young adults were randomly assigned to read internet postings that presented either 1) base rate information depicting the frequency of adverse reactions to Spice and kratom; 2) base rate information plus four personal web-postings describing beneficial reactions to Spice and kratom; or 3) base rate information plus four personal web-postings describing adverse reactions to Spice and kratom. Respondents subsequently evaluated the risk of using both drugs. Anecdotal evidence (personal stories) outweighed the impact of base rate evidence only when the personal stories described adverse drug reactions. Effective risk communication will benefit from differential use of both base rate evidence and personal stories.

I know the rule, but I'll just go with my gut: is there a rational use of intuition?

ABSTRACTResearch has established that human thinking is often biased by intuitive judgement. The base-rate neglect effect provides such an example, so named because people often support their decisions in stereotypical individuating information, neglecting base-rates. Here, we test the hypothesis that reasoners acknowledge information provided by base-rates and may use individuating information in support of a “rational” decision process. Results from four experiments show that “base-rate neglecting” occurs when participants acknowledge sample distributions; participants who prefer individuating over base-rate information perceive base-rates as less diagnostic and are more confident in their individuating-based responses; and that posterior probabilities (assigned after all relevant information is considered) predict more individuating-based responses for individuating-preference participants (suggesting a rational process). However, data also show a deeper form of base-rate neglect: even when some participants report to prefer base-rate information, define individuating information as non-diagnostic, and their posterior probabilities suggest otherwise, they still provide individuating-based responses.

Can We Identify Bad Cops Based on History? Base Rates of Historical Markers in Law Enforcement Pre-employment Evaluations

While psychometric police selection processes have progressively evolved, the efficacy of simple background information has not been extensively evaluated. This study examined the utility of base rate information to predict job performance among law enforcement officers. Pre-employment historical markers from bad hires were compared to good hires from a large sample of 1536 officers. Of this sample, 205 officers were categorized as bad hires if they were arrested, had greater than five misconduct episodes, or were terminated for cause within 5 years following hire. Base rates of 40 pre-employment history variables were compared for good and bad hires. Analysis of markers showed that bad hires had a significantly higher frequency of markers across family factors, mental health variables, conduct problems, and criminal justice outcomes compared to good hires. Among the largest differences were history of multiple physical altercations, outpatient psychiatric treatment, and events related to previous law enforcement employment such as employer reprimands, suspensions, or fitness for duty evaluations. Many other intuitive markers occurred too infrequently to analyze or showed no significant difference between good hires and bad hires in law enforcement settings. Base rates in addition to psychological test data are necessary aspects of the police pre-employment evaluation.

Customer reviews are not always informative: The impact of effortful versus heuristic processing

Online customer reviews and ratings are powerful sources of information that influence travel purchase decisions. The complexity of the online environment lends itself to heuristic processing, which aids in decision-making yet may bias perceptions. However, enlisting systematic processing reduces dependence on heuristics. Research suggests that base rate information is often neglected in favor of information that is more salient but less diagnostic. In the online review environment, base rate information takes the form of recommendation percentages, and review content varies in usefulness. This research examines the effect of base rate information, review content, and processing effort in a 2 × 2 × 2 experimental design. The results indicate that ambiguous base rate information and high processing effort cause consumers to pay greater attention to review content. In turn, this influences consumers’ perceptions, decisions, and recall. The findings deepen our understanding of dual processing systems and how they influence consumer choice.

The influence of conflict monitoring on meta-reasoning and response times in a base rate task

We examined the role of conflict monitoring processes in forming metacognitive judgements of confidence while performing base rate tasks. Recently proposed models of dual-process reasoning, as well as research, have shown that conflict detection might represent a link between Type 1 and Type 2 processing. Conflict detection has also been shown to affect metacognitive processes in reasoning tasks. By varying base rate probability and congruence, we generated base rate tasks of four distinct levels of congruence. The results of two experiments showed that participants were slower and less confident in conflict conditions regardless of their response. However, there were two distinct subsets of participants with different levels of sensitivity to conflict which resulted in different patterns of results when using low base rate ratios. In-depth analyses showed that the impact of base rate information in the formation of metacognitive judgements depended on congruence and response type. Base rate information was a more salient cue for metacognitive processes when responding according to base rates compared with responding according to belief. There is evidence that base rate information may serve as a direct cue for metacognition, independent of fluency.

Integrate instead of ignoring: Base rate neglect as a common fallacy of petroleum explorers

When petroleum explorers estimate geological probability of success (POS) and success case volumes for their prospects, they commonly suffer from various biases and fallacies that result in poor assessments. Using a data set of 25 exploration wells drilled by Lundin Petroleum on the Norwegian Continental Shelf (NCS) during the 5-yr period from 2011 to 2015, I demonstrate the base rate neglect fallacy in petroleum exploration. Lundin explorers consistently and significantly underestimated the geological POS and overestimated the success case volumes relative to the available base rate information on historical exploration success rates and sizes of previously discovered fields in the NCS area. These findings are consistent with previous studies of exploration track records of the industry and individual companies. Explorers erroneously tend to ignore the base rates likely because they do not see them as relevant to the evaluation of their specific prospects. As a result, exploration portfolios generally fail to deliver on promise. However, base rate neglect fallacy can be mitigated and eliminated. Explorers should make an effort to learn the base rates relevant to their prospects and integrate them with prospect evaluations, and managers should ensure that modeled portfolio outcomes are consistent with relevant base rates.

Retraction Note to: Investigative advising: a job for Bayes

Bayesian approaches to police decision support offer an improvement upon more commonly used statistical approaches. Common approaches to case decision support often involve using frequencies from cases similar to the case under consideration to come to an isolated likelihood that a given suspect either a) committed the crime or b) has a given characteristic or set of characteristics. The Bayesian approach, in contrast, offers formally contextualized estimates and utilizes the formal logic desired by investigators. Bayes’ theorem incorporates the isolated likelihood as one element of a three-part equation, the other parts being 1) what was known generally about the variables in the case prior to the case occurring (the scientific-theoretical priors) and 2) the relevant base rate information that contextualizes the evidence obtained (the event context). These elements are precisely the domain of decision support specialists (investigative advisers), and the Bayesian paradigm is uniquely apt for combining them into contextualized estimates for decision support. By formally combining the relevant knowledge, context, and likelihood, Bayes’ theorem can improve the logic, accuracy, and relevance of decision support statements.

Effects of exemplars and base-rate information on online physician rating sites

Effects of exemplars and base-rate information on online physician rating sites

Influence of category coherence and type of base-rate acquisition on property generalization

Category coherence involves identifying similarities between social categories, such as jobs and hobbies, and is primarily used for generalizing or drawing inferences about their properties. Category coherence is also used for drawing inferences from explicit information, such as base rates. For example, people generalize the properties of groups or brands from multiple sources (e.g., survey results and the share market in periodicals). However, generalizing the properties of categories using explicit base-rate information may not always involve category coherence; rather, it may involve directly analyzing base-rate information. Therefore, this study attempts to distinguish between conditions in which people use category coherence and explicit base-rate information. Specifically, we proposed that people acquiring explicit base-rate knowledge—described as a percentage—would generalize a category's properties according to its base rate; however, for base-rate information acquired via direct observation, the inductive strength in generalizing properties would be greater for categories with high levels of coherence. Our results (Experiment 1 and Experiment 2) validated these two predictions. Finally, the interaction between category coherence and type of base-rate acquisition was discussed.

Silver Spoons and Platinum Plans: How Childhood Environment Affects Adult Health Care Decisions

Can socioeconomic status in childhood influence desire for health coverage in adulthood? We develop and test a model that yielded two sets of findings across five experiments. First, people who grew up poor were generally less interested in health coverage compared to those who grew up wealthy. This effect was independent of people’s current level of socioeconomic status, emerged most strongly when adults were experiencing financial threat, and was mediated by differences in willingness to take risks between people from poor versus wealthy childhoods. Second, we show that this effect reverses when people are provided with base-rate information about disease. When information about the average likelihood of getting sick is made available, people who grew up poor were consistently more likely to seek health coverage than people who grew up wealthy. This effect was again strongest when people felt a sense of financial threat, and it was driven by people from poor versus wealthy childhoods differing in their perceptions of the likelihood of becoming sick. Overall, we show how, why, and when childhood socioeconomic status influences desire for health coverage.

Straight until proven gay: A systematic bias toward straight categorizations in sexual orientation judgments.

Perceivers achieve above chance accuracy judging others' sexual orientations, but they also exhibit a notable response bias by categorizing most targets as straight rather than gay. Although a straight categorization bias is evident in many published reports, it has never been the focus of systematic inquiry. The current studies therefore document this bias and test the mechanisms that produce it. Studies 1-3 revealed the straight categorization bias cannot be explained entirely by perceivers' attempts to match categorizations to the number of gay targets in a stimulus set. Although perceivers were somewhat sensitive to base rate information, their tendency to categorize targets as straight persisted when they believed each target had a 50% chance of being gay (Study 1), received explicit information about the base rate of gay targets in a stimulus set (Study 2), and encountered stimulus sets with varying base rates of gay targets (Study 3). The remaining studies tested an alternate mechanism for the bias based upon perceivers' use of gender heuristics when judging sexual orientation. Specifically, Study 4 revealed the range of gendered cues compelling gay judgments is smaller than the range of gendered cues compelling straight judgments despite participants' acknowledgment of equal base rates for gay and straight targets. Study 5 highlighted perceptual experience as a cause of this imbalance: Exposing perceivers to hyper-gendered faces (e.g., masculine men) expanded the range of gendered cues compelling gay categorizations. Study 6 linked this observation to our initial studies by demonstrating that visual exposure to hyper-gendered faces reduced the magnitude of the straight categorization bias. Collectively, these studies provide systematic evidence of a response bias in sexual orientation categorization and offer new insights into the mechanisms that produce it. (PsycINFO Database Record

Prompting deliberation increases base-rate use

AbstractPeople often base judgments on stereotypes, even when contradictory base-rate information is provided. In a sample of 438 students from two state universities, we tested several hypotheses regarding why people would prefer stereotype information over base-rates when making judgments: A) People believe stereotype information is more diagnostic than base-rate information, B) people find stereotype information more salient than base-rate information, or C) even though people have some intuitive access to base-rate information, they may need to engage in deliberation before they can make full use of it, and often fail to do so. In line with the deliberative failure account, and counter to the diagnosticity account, we found that inducing deliberation by having people evaluate statements supporting the use of base-rates increased the use of base-rate information. Moreover, counter to the salience and diagnosticity accounts, asking people to evaluate statements supporting the use of stereotypes decreased reliance on stereotype information. Additionally, more numerate subjects were more likely to make use of base-rate information.

Voluntary Sleep Choice and Its Effects on Bayesian Decisions

This study examines whether voluntary sleep restriction at commonly experienced levels impacts decision making in a Bayesian choice task. Participants recruited were largely traditional age college students from a regional state university (n = 100) and a federal military academy (n = 99; n = 56 and 43, respectively, used in final analysis). Sleep was measured by actigraphy over a one-week period, followed by performance of a decision task. The task involved two sources of information, base rate odds and sample evidence, with subjects asked to make a probability judgment. Results found that subjects with nightly sleep < 6 hr (sleep deprived = SD), relative to those with > 7 hr, placed less decision weight on new evidence, relative to base rate information, in making difficult choices. This result is strongest among female subjects. For easier choices, voluntary SD did not affect relative decision weights placed on the two sources of available information.

The neuropsychological assessment of cognitive deficits considering measures of performance variability.

Neuropsychologists often face interpretational difficulties when assessing cognitive deficits, particularly in cases of unclear cerebral etiology. How can we be sure whether a single test score below the population average is indicative of a pathological brain condition or normal? In the past few years, the topic of intra-individual performance variability has gained great interest. On the basis of a large normative sample, two measures of performance variability and their importance for neuropsychological interpretation will be presented in this paper: the number of low scores and the level of dispersion. We conclude that low scores are common in healthy individuals. On the other hand, the level of dispersion is relatively small. Here, base rate information about abnormally low scores and abnormally high dispersion across cognitive abilities are provided to improve the awareness of normal variability and to serve clinicians as additional interpretive measures in the diagnostic process.

The Impact of Base Rate Utilization and Clinical Experience on the Accuracy of Judgments Made with the HCR-20

It was hypothesized that clinical judges would make ineffective use of base rate information and that their level of experience in the forensic psychology field or with risk assessment would have no bearing on their ability to make accurate clinical judgments. A convenience sample of 43 members of two professional psychological organizations was used to determine the effects of base rates and experience on predictions made from a “blind” analysis of the Historical, Clinical, Risk-20 (HCR-20). Consistent with the hypotheses, base rate neglect was prevalent in the current study and prior experience had no effect on predictive accuracy.

RETRACTED ARTICLE: Investigative advising: a job for Bayes

Bayesian approaches to police decision support offer an improvement upon more commonly used statistical approaches. Common approaches to case decision support often involve using frequencies from cases similar to the case under consideration to come to an isolated likelihood that a given suspect either a) committed the crime or b) has a given characteristic or set of characteristics. The Bayesian approach, in contrast, offers formally contextualized estimates and utilizes the formal logic desired by investigators. Bayes’ theorem incorporates the isolated likelihood as one element of a three-part equation, the other parts being 1) what was known generally about the variables in the case prior to the case occurring (the scientific-theoretical priors) and 2) the relevant base rate information that contextualizes the evidence obtained (the event context). These elements are precisely the domain of decision support specialists (investigative advisers), and the Bayesian paradigm is uniquely apt for combining them into contextualized estimates for decision support. By formally combining the relevant knowledge, context, and likelihood, Bayes’ theorem can improve the logic, accuracy, and relevance of decision support statements.