Monday, November 2, 2020

Physical attractiveness: Not all implicit biases are negative, some traits foster positive unfairness, advantaging some candidates over others; mask wearing is appropriate may play some role in mitigating attractiveness bias

Guarding Against Implicit Bias: Attractiveness. Harrison L. Love, Richard B. Gunderman. Journal of the American College of Radiology, October 29, 2020.

Abstract: Most of the literature on implicit bias naturally focuses on negative stereotypes: categories such as race and gender that lead to discrimination against some candidates in job interview invitations, offers of employment, and promotions. Examples of implicit bias abound; for example, girls are often assumed to be less capable than boys in engineering and math, and parents rate these abilities in their daughters lower than in their sons, even when they perform equally in school [ 1 ].

Yet not all implicit biases are negative, and in fact some traits foster positive unfairness, advantaging some candidates over others. One such trait is physical attractiveness. Recognizing and compensating for this form of bias is equally important for members of the radiology profession.


Some traits tend to be viewed as attractive across all cultures. Amongthese are facial symmetry, clear complexion, and a narrow waistline [2].Other features tend to apply to particular genders. For example, men tend to be attracted to women who appear young and have full breasts and lips. Women tend to be attracted tomen who are taller than they are andwho have broad shoulders. Viewers appraise such features at a subliminallevel, gauging attractiveness afterviewing a photograph for only 1/100thof a second [2].There is evidence that the prefer-ence for attractiveness is innate [3]. Forexample, 1-year-old infants play longerand more intensely, experience lessdistress, and even appear to exhibitmore pleasure when playing withattractive people. Even infants asyoung as 2 months old gaze longer atattractive faces than unattractive ones.The corollary of attractiveness bias is abias against the unattractive [4].Hence it is no surprise that mostpeople take care of how they appear toothers, relying to some degree on attire,cosmetics, and grooming to create a morefavorable impression. Very few peoplewould willingly show up for a job inter-view appearing as they do the momentthey climb out of bed in the morning.That attractiveness should beappealing is a bit tautological, but manytheorists have speculated that it mayserve as an indicator of geneticfitness [5].For example, body build may offer cluesabout health, and general appearancemay serve as an indicator ofsocioeconomic status, both which maybe desirable to prospective mates.Somehaveargued,forexample,thatthe male preference for youth reflects aconcern with reproductivefitness, whilethe traits preferred by women serve asindicators of the capacity to offer protec-tion and provide resources. Even at theneurologic level, viewing the faces ofattractive people has been shown to acti-vate areas of the brain associated withreward [6].


As this bias toward attractiveness appliesacross genders, ages, and sexual orientations [6], there is ample evidence that it manifests in the workplace [7]. For example, such individuals are morelikely to be interviewed and hired, andthey tend to earn higher wages than individuals judged to be unattractive. Attractive wait staff in restaurants havebeen shown to receive higher tips,regardless of the quality of service theyprovide.And such effects appear not to beaccounted for by other traits; forexample, even when self-confidence istaken into account, attractive individualsretain an advantage in compensation.Likewise, studies in the legal professionhave shown that physically attractive defendants tend to receive more favor-able judgments. Such effects are likely toredound to the detriment of the biased.The underlying reasons for such bia-ses are complex, but it appears thatattractive people are, on balance, perceived by others to be friendlier, healthier, more intelligent, more competent, more generous, and more trustworthy than unattractive people, who tend to beperceived as duller, more introverted, and less generous and trustworthy. There is also reason to think that thebias toward physical attractiveness is at least somewhat unfair. Forexample, some studies suggest that the skill andproductivity of workers is not correlated with their attractiveness [6]. Likewise, here is little evidencethata ttractivepeople are more likely than others to becooperative, generous, or trustworthy.


The association between attractivenessand academic performance is especiallygermane to radiologists who are select-ing candidates for residency, fellowship,and post-training employment oppor-tunities. There appear to be strongpositive correlations between attractive-ness and such characteristics as perceivedintelligence, perceived academic perfor-mance, and perceived conscientiousness.However, there is no strong positivecorrelation between attractiveness andactual academic performance [4].Attractiveness influences what actu-ally happens in contexts such as schoolsand workplaces. Attractive students aremore likely to get into university becausethey are deemed more intelligent andconscientious. They are also likely toreceive better grades. Likewise, moreattractive people are more likely to gethired and be retained. And salariesappear to be 10% to 15% higher forindividuals deemed to be attractive,which is similar to wage differentialsassociated with gender and race.In some contexts, the implicit biastoward attractiveness may not even beunfair. Consider, for example, lines ofwork such as modeling and acting, inwhich the attractiveness of workers is likelyto enhance sales of products and tickets.


How should radiologists and otherradiology personnel who seek to eval-uate candidates, learners, and colleaguesfairly approach the problem of the pos-itive bias toward physical attractiveness?One part of the solution may be simplyto recognize that such biases exist.Knowing that they are biased to-ward attractive candidates, committeescharged with residency and fellowshipselection, searching and screening forjob candidates, and promotion andtenure can consciously question thedegree to which their deliberations arebiased by attractiveness. Where suchbiases are identified, attempts can bemade to compensate for them.Another way to reduce the effects ofattractiveness bias is to take the physicalappearance of those being evaluated outof the equation. In some cases, candi-date photographs can be excluded fromassessment. During interviews, evalua-tors can be blinded, for example byproviding one or more members of aselection committee with only the audiocomponent of an interview.It is worth noting that the currentcoronavirus pandemic and other situations in which mask wearing is appropriate may play some role in mitigating attractiveness bias. If part of a candi-dates face is covered when a photo-graph is taken or during a remote or in-person interview, the potential for facialattractiveness is diminished.Some have even suggested that arti-ficial intelligence might play a role inmitigating attractiveness bias. As suchbias is to some degree subjective, perhapscomputers could provide a more objec-tive assessment of candidates. Yet therecent travails of facial recognition tech-nology serve as an important reminderthat even seemingly objective computers may harborbiases based on theirpro-gramming and the data sets they havebeen tasked to learn from.In thefinal analysis, the challenge of implicit bias toward physical <attractiveness is not one that can beeliminated but must instead be managed. To begin with, we need toacknowledge that such biases exist.Moreover, we must recognize that in many situations, it is unfair to allowour expectations and evaluations to beshaped by how a person looks. Finally, we need to do our best toensure that such biases do not undulyinfluence our decision making, allow-ing traits that are only skin deep toobscure our perception of far moreimportant personal attributes such asintelligence, character, and experience,which in the long run are likelier topredict performance.

Combine people’s 1st estimate with their 2d estimate made from the perspective of a person they often disagree with; this produces highly accurate inner crowds, as compared to when people simply make a second guess

van de Calseyde, Philippe, and Emir Efendic. 2019. “Taking a Disagreeing Perspective Improves the Accuracy of People’s Quantitative Estimates.” PsyArXiv. November 15. doi:10.31234/

Abstract: Many decisions rest upon people’s ability to make estimates of some unknown quantities. In these judgments, the aggregate estimate of the group is often more accurate than most individual estimates. Remarkably, similar principles apply when aggregating multiple estimates made by the same person – a phenomenon known as the “wisdom of the inner crowd”. The potential contained in such an intervention is enormous and a key challenge is to identify strategies that improve the accuracy of people’s aggregate estimates. Here, we propose the following strategy: combine people’s first estimate with their second estimate made from the perspective of a person they often disagree with. In five pre-registered experiments (total N = 6425, with more than 53,000 estimates), we find that such a strategy produces highly accurate inner crowds (as compared to when people simply make a second guess, or when a second estimate is made from the perspective of someone they often agree with). In explaining its accuracy, we find that taking a disagreeing perspective prompts people to consider and adopt second estimates they normally would not consider as viable option, resulting in first- and second estimates that are highly diverse (and by extension more accurate when aggregated). However, this strategy backfires in situations where second estimates are likely to be made in the wrong direction. Our results suggest that disagreement, often highlighted for its negative impact, can be a powerful tool in producing accurate judgments.

We already knew... Leaving the Loners Alone: Preference for Solitude Evokes Ostracism

Leaving the Loners Alone: Dispositional Preference for Solitude Evokes Ostracism. Dongning Ren, Anthony M. Evans. Personality and Social Psychology Bulletin, November 2, 2020.

Abstract: What are the interpersonal consequences of seeking solitude? Leading theories in developmental research have proposed that having a general preference for solitude may incur significant interpersonal costs, but empirical studies are still lacking. In five studies (total N = 1,823), we tested whether target individuals with a higher preference for solitude were at greater risk for ostracism, a common, yet extremely negative, experience. In studies using self-reported experiences (Study 1) and perceptions of others’ experiences (Study 2), individuals with a stronger preference for solitude were more likely to experience ostracism. Moreover, participants were more willing to ostracize targets with a high (vs. low) preference for solitude (Studies 3 and 4). Why do people ostracize solitude-seeking individuals? Participants assumed that interacting with these individuals would be aversive for themselves and the targets (Study 5; preregistered). Together, these studies suggest that seeking time alone has important (and potentially harmful) interpersonal consequences.

Keywords: preference for solitude, ostracism, exclusion, person perception

Across five studies, we found consistent evidence that individuals who voluntarily seek solitude are at greater risk for ostracism. This conclusion is based on correlational evidence, using participants’ self-reported experiences and their perceptions of others’ experiences (Studies 1 and 2), as well as experimental evidence using verbal descriptions and simulated personality profiles (Studies 3–5). These findings were robust across the contexts of data collection: the United States and the Netherlands; online and in a laboratory; from college students and MTurk.

Importantly, our final study provides some insight into why people ostracize targets with higher preference for solitude. Ostracism intentions are related to both self-interested and other-regarding motives (Study 5). This finding supports and builds on Williams’ theorizing that people may use ostracism preemptively to avoid any aversive outcomes (Williams, 1997). Comparing the two motives further revealed that self-interest, wanting to avoid an unpleasant social interaction, was the primary motive underlying participants’ ostracizing intentions.

Our work also provides insights into general beliefs about solitude-seeking individuals (Studies 3–5): participants considered solitude-seeking individuals to be low in the need to belong, indifferent to belonging events, cold, competent, and introverted. While most of these evaluations are intuitive, given the conceptual link between preference for solitude and low sociality, the positive relationship between preference for solitude and competence is surprising. Past studies found that people believe that loneliness and introversion are associated with incompetence (Anderson & Kilduff, 2009Lau & Gruen, 1992), suggesting that lay people are able to distinguish preference for solitude from loneliness and introversion. Why do people perceive high (vs. low) solitude preference targets to be more competent? One possible interpretation is that preference for solitude is perceived as a sign of maturity, given that as people transition from adolescent to adulthood, choosing to spend time in solitude becomes more normative and purposeful (Coplan, Ooi, et al., 2019). Another possible interpretation is that preference for solitude is linked with independence in lay beliefs. In fact, wanting to be alone is termed as a motivation for independence in the fundamental social motives framework (Neel et al., 2016). Future research should try to better understand the intriguing link between preference for solitude and competence in lay beliefs.

Theoretical Contributions

The current research contributes to the growing literature on voluntary solitude. To date, there is a general lack of studies on the voluntary preference for solitude, and there are even fewer studies using adult samples or providing causal evidence (Coplan, Ooi, et al., 2019). Our research contributes to this literature by presenting clear evidence that having a strong preference for solitude is consequential in the interpersonal domain. The desire for “me time” is commonly experienced (e.g., Larson, 1990), and there are many potential benefits that voluntary solitude affords (e.g., Long et al., 2003). However, our research sheds light on potential barriers (and consequences) to seeking solitude—the risk of being ostracized and stigmatized.

The current studies suggest that the link between preference for solitude and ostracism could be dynamic and recursive. Targets of ostracism may withdraw from social interactions to minimize risk of additional social pain (Richman & Leary, 2009Van Kleef et al., 2010). In past experiments, targets of ostracism (vs. inclusion) indicated stronger intentions to disengage from social situations (Pfundmair et al., 2015), more positive ratings of physical spaces that hinder social interaction (Meagher & Marsh, 2017), and, importantly, a higher preference for being alone in the following activity (Ren et al., 2020). Here, we showed that, ironically, the very response to ostracism (i.e., preference for solitude) may put targets at higher risk for ostracism in future social interactions. To fully establish this bidirectional causal link between preference for solitude and ostracism, future work should track participants longitudinally.

The current studies also broaden our understanding of who is ostracized. Focusing on the broad Big Five dimensions, past studies identified two risk factors: low agreeableness and low conscientiousness (Rudert et al., 2020). Notably, narrow traits are often able to better predict domain-specific behavioral outcomes, even when controlling for global traits (Dudley et al., 2006Paunonen et al., 2003). Here, we focus on preference for solitude, a narrow, domain-specific trait, as both preference for solitude and ostracism are conceptually related to absence of social interactions (although in the case of ostracism, the absence is involuntary). We found that preference for solitude was associated with general ostracism experience, even while controlling for the Big Five traits (Study 1); in addition, participants did not consistently infer agreeableness or conscientiousness from targets’ preference for solitude (Studies 3–5: analyses on perceptions of targets). Taken together, these findings demonstrated that a narrow trait—preference for solitude—put individuals at heightened risk for ostracism above and beyond the known dispositional factors of agreeableness and conscientiousness. An interesting direction for future research is to explore other narrow, domain-specific traits (e.g., trait aggression) and examine multiple risk factors for ostracism in one study. This would allow researchers to examine the relative importance of each risk factor and any potential interaction effects between these factors.

More broadly, the current studies shed light on the question of why people ostracize others. Empirical attention has been given to self-interested or malicious reasons such as using ostracism to punish deviant or burdensome behaviors (Schachter, 1951Wesselmann et al., 2013Wirth et al., 2020). Our research adds to this literature by showing that people may have self-interested and other-regarding reasons for ostracizing others. This other-regarding motive for ostracism is in fact not uncommon in our daily life: people may stay silent during an argument with their partner to avoid saying anything harmful, refrain from inviting a busy coworker out for drinks so as not to distract them, or withhold information from a friend when they believe the information may hurt their feelings (a form of partial ostracism; Jones & Kelly, 2013). All these behaviors, albeit motivated by genuine concerns for the target individual, are still examples of the act of ostracizing.

Limitations and Additional Future Directions

In Studies 3 to 5, we used hypothetical profiles to manipulate preference for solitude. This approach is limited in two ways. First, the profiles (e.g., the verbal descriptions in Study 3) may not represent the actual levels of preference for solitude of individuals people encounter in their social environment. Recognizing this potential issue, in Studies 4 and 5, we used a data-driven approach of generating the hypothetical profiles. Second, the profiles made the information of preference for solitude explicit to the participants. In real life, people sometimes indeed make interpersonal decisions based on explicit personality information, for instance, in the domains of personnel selection (Ones & Viswesvaran, 1996) and romantic partner choice (Hall et al., 2010). Yet, at other times, people lack explicit knowledge of their interaction partners. Is preference for solitude a visible trait in social interaction? In other words, can people accurately infer others’ preference for solitude?

Past studies have not examined this question directly. However, there is suggestive evidence that people readily detect the preference for solitude in others. For example, children are able to recognize their peers’ preference for solitude and interact with these individuals accordingly (e.g., overlook them; Harrist et al., 1997). Similarly, adult participants can detect their friends’ motivation to spend time alone (referred to as independence) with some accuracy (Huelsnitz et al., 2020). Generally, people accurately detect personality traits in a target person based on brief interactions or minimal information (Connelly & Ones, 2010Tskhay & Rule, 2014). Moreover, compared with other traits, extroversion (a related construct) is more visible and more accurately rated by perceivers (Connelly & Ones, 2010).

Another limitation in our experiments is that we measured participants’ ostracism intentions; yet intentions do not necessarily predict actual behavior (e.g., Ajzen, 1991). However, we speculate that there is a relatively strong link between ostracism intentions and behavior (vs. other active forms of exclusion such as physical aggression; Kerr & Levine, 2008). Ostracism does not require an action (Williams, 19972009). In fact, it may take minimal effort to engage in ostracizing (e.g., not saying hello; Kerr & Levine, 2008). In addition, the ambiguous nature of ostracism makes it hard to be documented and thus sources may not be held accountable. It has been shown that ostracism (vs. harassment) is perceived to be more socially acceptable and less regulated at the workplace (O’Reilly et al., 2015). Finally, people tend to underestimate others’ social sufferings caused by ostracism (Nordgren et al., 2011), suggesting that the act of ostracizing is believed to be relatively inconsequential, which may further contribute to the link between intentions and behavior.

In addition to the limitations of the experiments, we collected data from Western countries (the United States and the Netherlands) in all five studies. This puts constraints on the generalizability of the results to other cultural contexts. It has been observed that people in Western cultures are more encouraged to be sociable and expressive, whereas people from East Asian cultures are more encouraged to be shy and self-reflective (Chen, 2010Ding et al., 2015Oyserman et al., 2002). Consistent with this observation, past research has suggested that solitude is more valued and experienced more positively in East Asian cultures than in Western cultures (Jiang et al., 2019). Thus, solitude-seeking individuals might be perceived more positively and at less risk for ostracism in East Asian cultures versus Western cultures. These ideas point to a fruitful avenue for future research.

Finally, future research should examine whether or not people’s judgments of those who prefer solitude are accurate. Participants in our studies assumed that preference for solitude is an undesirable disposition in social interactions. They anticipated interactions to be unpleasant for themselves and for the target individual. But are these valid concerns? We speculate that people might over-rely on preference for solitude as a predictor of social interaction outcomes. Because preference for solitude is not an indicator of a lack of interest in social interactions (Coplan, Ooi, et al., 2019), the need to affiliate is a basic need that applies to everyone (Baumeister & Leary, 1995), and that the immediate impact of ostracism is universally aversive regardless of dispositional characteristics of the targeted individual (McDonald & Donnellan, 2012), it is likely that individuals with a high preference for solitude would enjoy social interactions as much as others. Dispelling these misconceptions of solitude-seeking individuals might be an effective strategy to promote inclusive behaviors.

Novel techniquest (exorcism) instead of the Salem witchcraft convictions could have conferred even greater wealth for the Puritan church; but exorcism was inferior to executions for the congregant-maximizing Puritan ministers

The economics of Puritanism’s treatment of bewitchment: exorcism as a potential market-pull innovation? Franklin G. Mixon Jr. & Kamal P. Upadhyaya. European Journal of Law and Economics volume 50, pp 203–222 (2020). May 26 2020.

Abstract: A long history of research on the witchcraft hysteria in Massachusetts Bay Colony in 1692 contends that a group of Puritan ministers, including Salem Village’s Samuel Parris, developed and used the witchcraft hysteria in order to boost religiosity and church attendance in an effort to augment corporate and personal wealth. In carrying out this effort, these ministers pitted churched colonists against unchurched colonists, resulting in the wrongful convictions of 20 American colonials. This study argues that it might have ended without the executions of the colonists, and perhaps in even greater corporate wealth for the Puritan church, had Puritanism been receptive to the potential market-pull innovation represented by exorcism. Scrutiny of this proposition through the lens of rational choice theory suggests, however, that exorcism was inferior to executions as a technology choice for the congregant-maximizing Puritan ministers in Salem Village in 1692.

The Accuracy and of Personality Impressions from Faces, and the meta-accuracy (i.e., whether they are aware of their judgment accuracy), are both low

Jaeger, Bastian, Willem Sleegers, Julia Stern, Lars Penke, and Alex L. Jones. 2020. “The Accuracy and Meta-accuracy of Personality Impressions from Faces.” PsyArXiv. November 2. doi:10.31234/

Rolf Degen's take:

Abstract: People spontaneously judge others’ personality based on their facial appearance and these impressions guide many important decisions. Although the consequences of personality impressions are well documented, studies on the accuracy of personality impressions have yielded mixed results. Moreover, little is known about people’s meta-accuracy (i.e., whether they are aware of their judgment accuracy). Even if accuracy is generally low, meta-accuracy would allow people to rely on their impressions in the right situations. In two studies (one preregistered), we examined the accuracy and meta-accuracy of personality impressions. We addressed three crucial limitations of previous studies (a) by incentivizing accuracy and meta-accuracy, (b) by relying on substantially larger samples of raters and targets (646 participants rating 1,660 faces), and (c) by conducting Bayesian analyses to also quantify evidence for the null hypothesis. Our findings consistently suggest that people show neither accuracy nor meta-accuracy when forming face-based personality impressions.