De-Biasing Legal Factfinders

This chapter examines the psychological studies of biases and de-biasing measures in human decision-making with special reference to adjudicative factfinding. Research shows that factfinders are prone to cognitive biases (such as anchoring, framing, base-rate neglect, and confirmation bias) as well as social biases. Driven by this research, multiple studies have examined the extent to which those biases can be mitigated by de-biasing measures like “consider the opposite” and “give reasons.” After a brief overview of the research, the author points to the problematic evidential basis and identifies future research needs, and concludes that empirical research on de-biasing measures has so far delivered less than one would hope for.

INFLUENCER CULTURE & ALGORITHMIC APARTHEID ON GAY INSTAGRAM

Research on social media influencers has concluded that influencers’ status is always contingent upon meeting followers’ demand for specific “performances-of-self,” as well as making themselves and their content available and visible. However, few scholars have addressed the role that algorithms play in determining which influencers might be made visible, and therefore the manner by which platform design predetermines who may attain influencer status. This project seeks to address this gap in this research by providing critical ethnographic insights into what is colloquially referred to as “gay Instagram.” Throughout a three-year period, I maintained an Instagram account through which I observed trends in the production and consumption of homoerotic content utilizing a variety of methods. Of these, “algorithmic audits” provided the most insight into the manner by which algorithmic and social biases compound one another to hyper-visibilize a small minority of homoerotic content creators. Once Instagram has determined that a user is interested in homoerotic content, that user can expect the overrepresentation of white, mostly American Instagays on their “explore” page, in the promoted content featured on their feed, and so on. In effect, determinations of the most desirable homoerotic content are made through a variety of selection biases that make access to visibility an unequal enterprise on Instagram. Through these biases, white elites in Western metropolises are made more visible to Instagram users, even when others could conceivably fulfill their same representational function, troubling the notion that influencer status can be attained through an individual’s “labor” without algorithmic assistance.

Biases in COVID-19 Medical Resource Dilemmas

Accruing evidence suggest that COVID-19 is more fatal for males and minorities than other sub-populations. In this paper, we study medical dilemmas pertaining to the allocation of medical resources to evaluate whether existing social biases correspond to the demographic disparities of the pandemic. We develop and implement a choice experiment in which participants decide how to allocate scarce medical resources among COVID-19 patients with diverse demographic attributes. We find that participants violate optimal resource allocation significantly more often for the benefit of females. Males are almost half as likely to receive lifesaving resources even if these are medically more beneficial for them. We also find that participants are less likely to assign resources to patients with high compared to low income. Last, we find no evidence of patients' race affecting allocation preferences.

Eliciting Model Structures for Multivariate Probabilistic Risk Analysis

Notionally objective probabilistic risk models, built around ideas of cause and effect, are used to predict impacts and evaluate trade-offs. In this paper, we focus on the use of expert judgement to fill gaps left by insufficient data and understanding. Psychological and contextual phenomena such as anchoring, availability bias, confirmation bias and overconfidence are pervasive and have powerful effects on individual judgements. Research across a range of fields has found that groups have access to more diverse information and ways of thinking about problems, and routinely outperform credentialled individuals on judgement and prediction tasks. In structured group elicitation, individuals make initial independent judgements, opinions are respected, participants consider the judgements made by others, and they may have the opportunity to reconsider and revise their initial estimates. Estimates may be aggregated using behavioural, mathematical or combined approaches. In contrast, mathematical modelers have been slower to accept that the host of psychological frailties and contextual biases that afflict judgements about parameters and events may also influence model assumptions and structures. Few, if any, quantitative risk analyses embrace sources of uncertainty comprehensively. However, several recent innovations aim to anticipate behavioural and social biases in model construction and to mitigate their effects. In this paper, we outline approaches to eliciting and combining alternative ideas of cause and effect. We discuss the translation of ideas into equations and assumptions, assessing the potential for psychological and social factors to affect the construction of models. We outline the strengths and weaknesses of recent advances in structured, group-based model construction that may accommodate a variety of understandings about cause and effect.

Cognitive and Affective Processes Associated with Social Biases

Background Decision-making and judgments in our social activities often erroneous and irrational, known as social biases. However, cognitive and affective processes that produce such biases remain largely unknown. In this study, we investigated associations between social schemas, such as social judgment and conformity, entailing social biases and psychological measurements relevant to cognitive and affective functions. Method Forty-two healthy adult subjects were recruited in this study. A psychological test and a questionnaire were administered to assess biased social judgements by superficial attributes and social conformity by adherence to social norms, respectively, along with additional questionnaires and psychological tests for cognitive and affective measurements, including negative affects, autistic traits, and Theory of Mind (ToM). Associations of social judgment and conformity with cognitive and affective functions were examined multiple regression analysis and structural equation modeling. Results Anxiety and the cognitive realm of ToM were mutually associated with both social judgments and conformity, although social judgements and conformity were still independent processes with each other. Social judgements were also associated with autistic traits and the affective realm of ToM, whereas social conformity was associated with negative affects other than anxiety and intuitive decision-making style. Conclusions These results suggest that ToM and negative affects may play important roles in social judgements and conformity, and social biases connoted in these social schemas.

Free ranking vs. rank-choosing: New insights on the conjunction fallacy

It has been consistently shown that when asked to rank options, people often make fallacious judgements. Furthermore, such fallacies can be sensitive to presentation mode. In the first study, we explored a novel type of ranking presentation, namely, choosing between two rank orders of options. To enable a direct comparison of fallacy rates between free ranking and the new presentation mode, we calculated the frequencies of the two presented rank orders in a free ranking condition. Our analysis shows that people choose the non-fallacious rank order significantly more when asked to choose between two rank orders as compared to freely ranking the possible options. In a second study, we explored whether an agent presenting the rankings to choose from has an effect. To alleviate social biases we used videos of social robots as the presenting agents. We show that rank orders presented by social agents significantly reduce the fallacy rates, compared to rank orders presented without an agent. We discuss the results in view of social decision making theories, wherein a ranking presentation by a social agent is comparable to a consultation with another person regarding free ranking. Our results suggest that fallacious decision making can be mitigated by a social agent presenting rankings to choose from.

Love and basketball: Audience response to a professional athlete’s mental health proclamation

Objective: On 6 March 2018, the Cleveland Cavaliers power forward Kevin Love penned an op-ed for The Players’ Tribune. In big, bold letters, the title, ‘Everyone Is Going Through Something’ gave way to his proclamation of personal mental health struggles. This study assessed the impact of Love’s personal mental health testimony on the audience. Design: Content analysis of Instagram user’s responses to Love’s 2018 announcement of panic disorder. Setting: Data were collected on 19 June 2018. Comments on Love’s post were manually copied into an Excel spreadsheet. Due to Instagram’s ‘top comments’ and ‘newest first’ algorithms that determine both the visibility and order of user responses, not all comments were accessible for analysis. A total of 1,234 comments served as the sample. Method: Two trained coders analysed the comments for exhibitions of audience involvement with Love, emotional responses and perceptions of social biases surrounding mental illness. Results: Social media users who expressed hope were more likely to share their own mental health experiences. In addition, users who expressed hope were significantly more likely to mention wanting to reduce social bias surrounding discussion of mental health than posts without hope. Conclusion: Building on previous work that determined social media users can respond to a celebrity’s physical health testimonial with positive, future-oriented emotions, this study established that the same response can exist when analysing a celebrity’s mental health testimonial.

Modern Philosophy and Ancient Heresies: New Wine in Old Bottles?

This chapter revisits seventeenth- and eighteenth-century debates on the Eucharist and atomism as the backdrop of issues of credibility and authority. It explores how theology, erudition, and philosophy intertwined in defining dogma, and analyses how the Holy Office and the Congregation of the Index of Prohibited Books dealt with scientific tendencies within the Catholic world. In the light of recent scholarship underscoring the inertia of censorial mechanisms, the chapter argues that looking at censorship as a performative activity and at the social biases of doctrinal control helps to understand Rome’s attitude vis-à-vis philosophical novelties. More broadly, this new focalization highlights how truth was administrated and credibility established in the Catholic context.

It’s a Match: Task Assignment in Human–Robot Collaboration Depends on Mind Perception

Robots are becoming more available for workplace collaboration, but many questions remain. Are people actually willing to assign collaborative tasks to robots? And if so, exactly which tasks will they assign to what kinds of robots? Here we leverage psychological theories on person-job fit and mind perception to investigate task assignment in human–robot collaborative work. We propose that people will assign robots to jobs based on their “perceived mind,” and also that people will show predictable social biases in their collaboration decisions. In this study, participants performed an arithmetic (i.e., calculating differences) and a social (i.e., judging emotional states) task, either alone or by collaborating with one of two robots: an emotionally capable robot or an emotionally incapable robot. Decisions to collaborate (i.e., to assign the robots to generate the answer) rates were high across all trials, especially for tasks that participants found challenging (i.e., the arithmetic task). Collaboration was predicted by perceived robot-task fit, such that the emotional robot was assigned the social task. Interestingly, the arithmetic task was assigned more to the emotionally incapable robot, despite the emotionally capable robot being equally capable of computation. This is consistent with social biases (e.g., gender bias) in mind perception and person-job fit. The theoretical and practical implications of this work for HRI are being discussed.