Objective: Mindfulness-based intervention has been proposed to alleviate loneliness and improve social connectedness. Several randomized controlled trials have been conducted to evaluate the effectiveness of MBI. This study aimed to critically evaluate and determine the effectiveness and safety of MBI in alleviating the feeling of loneliness.Methods: We searched Medline, Embase, PsycInfo, Cochrane CENTRAL, and AMED for publications from inception to May 2020. We included RCTs with human subjects who were enrolled in MBI with loneliness as an outcome. The quality of (...) evidence was assessed using Cochrane's Risk of Bias tool and Grading of Recommendations Assessment, Development, and Evaluation. A random-effects model was used for meta-analysis.Results: Out of 92 articles identified, eight studies involving 815 participants were included in this study. Most trials conducted a minimum of 8 weeks of MBI. Most of the trials used UCLA-Loneliness Scale. A pooled analysis combining three trials and compared with wait-list showed significant improvement in loneliness score reduction using the UCLA-R scale with MD of −6.33 [95% confidence interval : −9.39, −3.26]. Subgroup analysis with only two Cognitively-Based Compassion Training trials also showed similar MD of −6.05. The overall quality of evidence was low.Conclusions: Mindfulness intervention with an average length of 8-week duration significantly improved the population's loneliness level with no mental health issue. However, this evidence had a low GRADE level. (shrink)

Biosemiosis is a process of choice-making between simultaneously alternative options. It is well-known that, when sufficiently young children encounter a new word, they tend to interpret it as pointing to a meaning that does not have a word yet in their lexicon rather than to a meaning that already has a word attached. In previous research, the strategy was shown to be optimal from an information theoretic standpoint. In that framework, interpretation is hypothesized to be driven by the minimization of (...) a cost function: the option of least communication cost is chosen. However, the information theoretic model employed in that research neither explains the weakening of that vocabulary learning bias in older children or polylinguals nor reproduces Zipf’s meaning-frequency law, namely the non-linear relationship between the number of meanings of a word and its frequency. Here we consider a generalization of the model that is channeled to reproduce that law. The analysis of the new model reveals regions of the phase space where the bias disappears consistently with the weakening or loss of the bias in older children or polylinguals. The model is abstract enough to support future research on other levels of life that are relevant to biosemiotics. In the deep learning era, the model is a transparent low-dimensional tool for future experimental research and illustrates the predictive power of a theoretical framework originally designed to shed light on the origins of Zipf’s rank-frequency law. (shrink)

There is an active debate within the field of phonology concerning the cognitive status of substantive phonetic factors such as ease of articulation and perceptual distinctiveness. A new framework is proposed in which substance acts as a bias, or prior, on phonological learning. Two experiments tested this framework with a method in which participants are first provided highly impoverished evidence of a new phonological pattern, and then tested on how they extend this pattern to novel contexts and novel (...) sounds. Participants were found to generalize velar palatalization (e.g., the change from [k] as in keep to [t⌢∫S] as in cheap) in a way that accords with linguistic typology, and that is predicted by a cognitive bias in favor of changes that relate perceptually similar sounds. Velar palatalization was extended from the mid front vowel context (i.e., before [e] as in cape) to the high front vowel context (i.e., before [i] as in keep), but not vice versa. The key explanatory notion of perceptual similarity is quantified with a psychological model of categorization, and the substantively biased framework is formalized as a conditional random field. Implications of these results for the debate on substance, theories of phonological generalization, and the formalization of similarity are discussed. (shrink)

It has consistently been shown that agents judge the intervals between their actions and outcomes as compressed in time, an effect named intentional binding. In the present work, we investigated whether this effect is result of prior bias volunteers have about the timing of the consequences of their actions, or if it is due to learning that occurs during the experimental session. Volunteers made temporal estimates of the interval between their action and target onset , or between two (...) events . Our results show that temporal estimates become shorter throughout each experimental block in both conditions. Moreover, we found that observers judged intervals between action and outcomes as shorter even in very early trials of each block. To quantify the decrease of temporal judgments in experimental blocks, exponential functions were fitted to participants’ temporal judgments. The fitted parameters suggest that observers had different prior biases as to intervals between events in which action was involved. These findings suggest that prior bias might play a more important role in this effect than calibration-type learning processes. (shrink)

The increasing implementation of and reliance on machine-learning (ML) algorithms to perform tasks, deliver services and make decisions in health and healthcare have made the need for fairness in ML, and more specifically in healthcare ML algorithms (HMLA), a very important and urgent task. However, while the debate on fairness in the ethics of artificial intelligence (AI) and in HMLA has grown significantly over the last decade, the very concept of fairness as an ethical value has not yet been (...) sufficiently explored. Our paper aims to fill this gap and address the AI ethics principle of fairness from a conceptual standpoint, drawing insights from accounts of fairness elaborated in moral philosophy and using them to conceptualise fairness as an ethical value and to redefine fairness in HMLA accordingly. To achieve our goal, following a first section aimed at clarifying the background, methodology and structure of the paper, in the second section, we provide an overview of the discussion of the AI ethics principle of fairness in HMLA and show that the concept of fairness underlying this debate is framed in purely distributive terms and overlaps with non-discrimination, which is defined in turn as the absence of biases. After showing that this framing is inadequate, in the third section, we pursue an ethical inquiry into the concept of fairness and argue that fairness ought to be conceived of as an ethical value. Following a clarification of the relationship between fairness and non-discrimination, we show that the two do not overlap and that fairness requires much more than just non-discrimination. Moreover, we highlight that fairness not only has a distributive but also a socio-relational dimension. Finally, we pinpoint the constitutive components of fairness. In doing so, we base our arguments on a renewed reflection on the concept of respect, which goes beyond the idea of equal respect to include respect for individual persons. In the fourth section, we analyse the implications of our conceptual redefinition of fairness as an ethical value in the discussion of fairness in HMLA. Here, we claim that fairness requires more than non-discrimination and the absence of biases as well as more than just distribution; it needs to ensure that HMLA respects persons both as persons and as particular individuals. Finally, in the fifth section, we sketch some broader implications and show how our inquiry can contribute to making HMLA and, more generally, AI promote the social good and a fairer society. (shrink)

Attentional bias to threat has been almost exclusively examined after participants experienced repeated pairings between a conditioned stimulus (CS) and an aversive unconditioned stimulus (US). This study aimed to determine whether threat-related attentional capture can result from observational learning, when participants acquire knowledge of the aversive qualities of a stimulus without themselves experiencing aversive outcomes. Non-clinical young-adult participants (N = 38) first watched a video of an individual (the demonstrator) performing a Pavlovian conditioning task in which one colour (...) was paired with shock (CS+) and another colour was neutral (CS−). They then carried out visual search for a shape-defined target. Oculomotor measures evidenced an attentional bias toward the CS+ colour, suggesting that threat-related attentional capture can ensue from observational learning. Exploratory analyses also revealed that this effect was positively correlated with empathy for the demonstrator. Our findings extend empirical and theoretical knowledge about threat-driven attention and provide valuable insights to better understand the formation of anxiety disorders. (shrink)

The use of predictive systems has become wider with the development of related computational methods, and the evolution of the sciences in which these methods are applied Solon and Selbst and Pedreschi et al.. The referred methods include machine learning techniques, face and/or voice recognition, temperature mapping, and other, within the artificial intelligence domain. These techniques are being applied to solve problems in socially and politically sensitive areas such as crime prevention and justice management, crowd management, and emotion analysis, (...) just to mention a few. However, dissimilar predictions can be found nowadays as the result of the application of these methods resulting in misclassification, for example for the case of conviction risk assessment Office of Probation and Pretrial Services or decision-making process when designing public policies Lange. The goal of this paper is to identify current gaps on fairness achievement within the context of predictive systems in artificial intelligence by analyzing available academic and scientific literature up to 2020. To achieve this goal, we have gathered available materials at the Web of Science and Scopus from last 5 years and analyzed the different proposed methods and their results in relation to the bias as an emergent issue in the Artificial Intelligence field of study. Our tentative conclusions indicate that machine learning has some intrinsic limitations which are leading to automate the bias when designing predictive algorithms. Consequently, other methods should be explored; or we should redefine the way current machine learning approaches are being used when building decision making/decision support systems for crucial institutions of our political systems such as the judicial system, just to mention one. (shrink)

This paper conducted a preliminary study of reviewing and exploring bias strategies using a framework of a different discipline: change management. The hypothesis here is: If the major problem of implicit bias strategies is that they do not translate into actual changes in behaviors, then it could be helpful to learn from studies that have contributed to successful change interventions such as reward management, social neuroscience, health behavioral change, and cognitive behavioral therapy. The result of this integrated approach (...) is: current bias strategies can be improved and new ones can be developed with insight from adjunct study fields in change management; it could be more sustainable to invest in a holistic and proactive bias strategy approach that targets the social environment, eliminating the very condition under which biases arise; and while implicit biases are automatic, future studies should invest more on strategies that empower people as “change agents” who can act proactively to regulate the very environment that gives rise to their biased thoughts and behaviors. (shrink)

Previous research suggests that implicit sequence learning is superior for believers in the paranormal and individuals with increased cerebral dopamine. Thirty-five healthy participants performed feedback-guided anticipations of four arrow directions. A 100-trial random sequence preceded two 100-trial biased sequences in which visual targets on trial t tended to be displaced 90° clockwise or counter-clockwise from those on t − 1. ISL was defined as a positive change during the course of the biased run in the difference between pro-bias (...) and counter-bias responses. It was hypothesized that this difference would be greater for believers in the paranormal than for skeptics, for those who received dopamine than for those who received placebo, and for believers who received dopamine than for the other groups. None of the hypotheses were supported by the data. It is suggested that a simple binary guessing task with a focus on prediction accuracy during early trials should be considered for future explorations. (shrink)

When applied in the health sector, AI-based applications raise not only ethical but legal and safety concerns, where algorithms trained on data from majority populations can generate less accurate or reliable results for minorities and other disadvantaged groups.

Gender bias is a recalcitrant problem in academia and society. However, dissent has been created on this issue. We focus on dissenting studies by Stephen J. Ceci and Wendy M. Williams, arguing that they reach conclusions that are unwarranted on the basis of the available evidence and that they ignore fundamental objections to their methodological decisions. Drawing on discussions from other contexts, particularly on manufactured dissent concerning anthropogenic climate change, we conclude that dissent on gender bias substantially contributes (...) to the exacerbation of biases in society and an increasing number of attacks on researchers, making it both epistemically and socially problematic. (shrink)

Purpose The online economy has not resolved the issue of racial bias in its applications. While algorithms are procedures that facilitate automated decision-making, or a sequence of unambiguous instructions, bias is a byproduct of these computations, bringing harm to historically disadvantaged populations. This paper argues that algorithmic biases explicitly and implicitly harm racial groups and lead to forms of discrimination. Relying upon sociological and technical research, the paper offers commentary on the need for more workplace diversity within high-tech (...) industries and public policies that can detect or reduce the likelihood of racial bias in algorithmic design and execution. Design/methodology/approach The paper shares examples in the US where algorithmic biases have been reported and the strategies for explaining and addressing them. Findings The findings of the paper suggest that explicit racial bias in algorithms can be mitigated by existing laws, including those governing housing, employment, and the extension of credit. Implicit, or unconscious, biases are harder to redress without more diverse workplaces and public policies that have an approach to bias detection and mitigation. Research limitations/implications The major implication of this research is that further research needs to be done. Increasing the scholarly research in this area will be a major contribution in understanding how emerging technologies are creating disparate and unfair treatment for certain populations. Practical implications The practical implications of the work point to areas within industries and the government that can tackle the question of algorithmic bias, fairness and accountability, especially African-Americans. Social implications The social implications are that emerging technologies are not devoid of societal influences that constantly define positions of power, values, and norms. Originality/value The paper joins a scarcity of existing research, especially in the area that intersects race and algorithmic development. (shrink)

Confirmation bias has been widely studied for its role in failures of reasoning. Individuals exhibiting confirmation bias fail to engage with information that contradicts their current beliefs, and, as a result, can fail to abandon inaccurate beliefs. But although most investigations of confirmation bias focus on individual learning, human knowledge is typically developed within a social structure. We use network models to show that moderate confirmation bias often improves group learning. However, a downside is (...) that a stronger form of confirmation bias can hurt the knowledge producing capacity of the community. (shrink)

A multimodal person representation contains information about what a person looks like and what a person sounds like. However, little is known about how children form these face-voice mappings. Here, we explored the possibility that two cognitive tools that guide word learning, a one-to-one mapping bias and fast mapping, also guide children’s learning about faces and voices. We taught 4- and 5-year-olds mappings between three individual faces and voices, then presented them with new faces and voices. In (...) Experiment 1, we found that children rapidly learned face-voice mappings from just a few exposures, and furthermore spontaneously mapped novel faces to novel voices using a one-to-one mapping bias (that each face can produce only one voice). In Experiment 2, we found that children’s face-voice representations are abstract, generalizing to novel tokens of a person. In Experiment 3, we found that children retained in memory the face-voice mappings that they had generated via inference (i.e., they showed evidence of fast mapping), and used these newly formed representations to generate further mappings between new faces and voices. These findings suggest that preschoolers’ rapid learning about faces and voices may be aided by biases that are similar to those that support word learning. (shrink)

In recent years, the work of organizations in the area of digitization has intensified significantly. This trend is also evident in the field of recruitment where job application tracking systems (ATS) have been developed to allow job advertisements to be published online. However, recent studies have shown that recruiting in most organizations is not inclusive, being subject to human biases and prejudices. Most discrimination activities appear early but subtly in the hiring process, for instance, exclusive phrasing in job advertisement discourages (...) qualified applicants from minority groups from applying. The existing works are limited to analyzing, categorizing and highlighting the occurrence of bias in the recruitment process. In this paper, we go beyond this and develop machine learning models for identifying and classifying biased and discriminatory language in job descriptions. We develop and evaluate a machine learning system for identifying five major categories of biased and discriminatory language in job advertisements, i.e., masculine-coded, feminine-coded, exclusive, LGBTQ-coded, demographic and racial language. We utilized the combination of linguistic features with recent state-of-the-art word embeddings representations as input features for various machine learning classifiers. Our results show that the machine learning classifiers were able to identify all the five categories of biased and discriminatory language with a decent accuracy. The Random Forest classifier with FastText word embeddings achieved the best performance with tenfolds cross-validation. Our system directly addresses the bias in the attraction phase of hiring by identifying and classifying biased and discriminatory language and thus encouraging recruiters to write more inclusive job advertisements. (shrink)

Volume 51, Issue 11, October 2019, Page 1063-1072.

This article centres on two distinct but intersecting questions: does it matter if we cannot definitively answer the question ‘what is philosophy?’ and do philosophers exhibit bias? The article will answer ‘yes’ to both questions for the following reasons. First because the uncertainty has allowed some answers to dominate. Second, because the answers necessarily demonstrate biases, and these have led to a lack of diversity in the discipline. Following this, the article will consider why philosophers have been slow or (...) reluctant to accept that bias plays a role in what we do. This lack of recognition has, I suggest, limited opportunity for diverse approaches, methods, and standpoints to flourish. The discipline has suffered as a result, and so too have diverse philosophers. Many fields in science, technology, engineering and mathematics meanwhile have sought to tackle the lack of diversity head on, and philosophers would do well to heed their examples. (shrink)

Machine Learning (ML) is on the rise in medicine, promising improved diagnostic, therapeutic and prognostic clinical tools. While these technological innovations are bound to transform health care, they also bring new ethical concerns to the forefront. One particularly elusive challenge regards discriminatory algorithmic judgements based on biases inherent in the training data. A common line of reasoning distinguishes between justified differential treatments that mirror true disparities between socially salient groups, and unjustified biases which do not, leading to misdiagnosis and (...) erroneous treatment. In the curation of training data this strategy runs into severe problems though, since distinguishing between the two can be next to impossible. We thus plead for a pragmatist dealing with algorithmic bias in healthcare environments. By recurring to a recent reformulation of William James’s pragmatist understanding of truth, we recommend that, instead of aiming at a supposedly objective truth, outcome-based therapeutic usefulness should serve as the guiding principle for assessing ML applications in medicine. (shrink)

The bias–variance tradeoff is a theoretical concept that suggests machine learning algorithms are susceptible to two kinds of error, with some algorithms tending to suffer from one more than the other. In this letter, we claim that the bias–variance tradeoff is a general concept that can be applied to human cognition as well, and we discuss implications for research in cognitive science. In particular, we show how various strands of research in cognitive science can be interpreted in (...) light of the bias–variance tradeoff, giving insight into individual differences in learning, the nature of cognitive processes, and debates in cognitive science research. (shrink)

Often machine learning programs inherit social patterns reflected in their training data without any directed effort by programmers to include such biases. Computer scientists call this algorithmic bias. This paper explores the relationship between machine bias and human cognitive bias. In it, I argue similarities between algorithmic and cognitive biases indicate a disconcerting sense in which sources of bias emerge out of seemingly innocuous patterns of information processing. The emergent nature of this bias obscures (...) the existence of the bias itself, making it difficult to identify, mitigate, or evaluate using standard resources in epistemology and ethics. I demonstrate these points in the case of mitigation techniques by presenting what I call ‘the Proxy Problem’. One reason biases resist revision is that they rely on proxy attributes, seemingly innocuous attributes that correlate with socially-sensitive attributes, serving as proxies for the socially-sensitive attributes themselves. I argue that in both human and algorithmic domains, this problem presents a common dilemma for mitigation: attempts to discourage reliance on proxy attributes risk a tradeoff with judgement accuracy. This problem, I contend, admits of no purely algorithmic solution. (shrink)

Recently, amid growing awareness that computer algorithms are not neutral tools but can cause harm by reproducing and amplifying bias, attempts to detect and prevent such biases have intensified. An approach that has received considerable attention in this regard is the Value Sensitive Design (VSD) methodology, which aims to contribute to both the critical analysis of (dis)values in existing technologies and the construction of novel technologies that account for specific desired values. This article provides a brief overview of the (...) key features of the Value Sensitive Design approach, examines its contributions to understanding and addressing issues around bias in computer systems, outlines the current debates on algorithmic bias and fairness in machine learning, and discusses how such debates could profit from VSD-derived insights and recommendations. Relating these debates on values in design and algorithmic bias to research on cognitive biases, we conclude by stressing our collective duty to not only detect and counter biases in software systems, but to also address and remedy their societal origins. (shrink)

Humans typically display hindsight bias. They are more confident that the evidence available beforehand made some outcome probable when they know the outcome occurred than when they don't. There is broad consensus that hindsight bias is irrational, but this consensus is wrong. Hindsight bias is generally rationally permissible and sometimes rationally required. The fact that a given outcome occurred provides both evidence about what the total evidence available ex ante was, and also evidence about what that evidence (...) supports. Even if you in fact evaluate the ex ante evidence correctly, you should not be certain of this. Then, learning the outcome provides evidence that if you erred, you are more likely to have erred low rather than high in estimating the degree to which the ex ante evidence supported the hypothesis that that outcome would occur. (shrink)

Some cultural evolution researchers have argued for the importance of prestige bias as a systematic and widespread social learning bias, that structures human social learning and cultural transmission patterns. Broadly speaking, prestige bias accounts understand it as a bias towards copying ‘prestigious’ individuals. Prestige bias, along with other social learning biases, has been argued to pay a crucial role in allowing cumulative cultural selection to take place, thereby generating adaptations that are key (...) to our success as a species. However, I argue for skepticism about the plausibility and scope of a prestige bias account. I argue that although an account of prestige bias seems plausible or compelling on their face, it is committed to a particular view of the cognition underpinning the bias, and therefore to predictions regarding its flexibility and context-sensitivity. Given this, current empirical evidence gives us reason to doubt the explanatory value of a prestige bias account over a naive, goal-directed agent account. Additionally, the way that prestige is defined in empirical work is in tension with a general understanding of prestige, casting doubt upon its status as evidence of prestige bias. I examine two studies cited as evidence of prestige bias, arguing that in these cases we cannot clearly favour a prestige bias explanation over a goal-directed agent explanation. (shrink)

It is often thought that traditional recidivism prediction tools used in criminal sentencing, though biased in many ways, can straightforwardly avoid one particularly pernicious type of bias: direct racial discrimination. They can avoid this by excluding race from the list of variables employed to predict recidivism. A similar approach could be taken to the design of newer, machine learning-based (ML) tools for predicting recidivism: information about race could be withheld from the ML tool during its training phase, ensuring (...) that the resulting predictive model does not use race as an explicit predictor. However, if race is correlated with measured recidivism in the training data, the ML tool may ‘learn’ a perfect proxy for race. If such a proxy is found, the exclusion of race would do nothing to weaken the correlation between risk (mis)classifications and race. Is this a problem? We argue that, on some explanations of the wrongness of discrimination, it is. On these explanations, the use of an ML tool that perfectly proxies race would (likely) be more wrong than the use of a traditional tool that imperfectly proxies race. Indeed, on some views, use of a perfect proxy for race is plausibly as wrong as explicit racial profiling. We end by drawing out four implications of our arguments. (shrink)

Are human beings irredeemably irrational? If so, why? In this article, I suggest that we need a broader appreciation of thought and reasoning to understand why people get things wrong. Although we can never escape cognitive bias, learning to recognize and understand it can help us push back against its dangers – and in particular to do so collectively and collaboratively.

이 글은 인공지능에 의한 차별과 그 책임 논의를 철학적 차원에서 본격적으로 연구하기에 앞선 예비적 고찰이다. 인공지능에 의한 차별을 철학자들의 연구를 요하는 당면 ‘문제’로 제기하고, 이를 위해 ‘인공지능에 의한 차별’이라는 문제의 성격과 원인을 규명하는 것이 이 글의 주된 목적이다. 인공지능은 기존 차별을 그대로 반복하여 현존하는 차별의 강화 및 영속화를 야기할 수 있으며, 이는 먼 미래의 일이 아니다. 이러한 문제는 현재 발생 중이며 공동체적 대응을 요구한다. 그러나 철학자의 입장에서 그와 관련한 책임 논의를 다루기는 쉽지 않다. 그 이유는 크게 인공지능의 복잡한 기술적 문제와 (...) 전통적 책임 개념의 한계라 할 수 있다. 여기서는 특히 인공지능과 관련된 기술적 개념을 중심으로 문제를 파악하고, 이와 관련하여 인간 행위자의 책임 및 그 역량을 규명한다. 통칭 인공지능으로 이해되는 것은 알고리즘에 기반한 자동화 시스템이며, 알고리즘의 편향성이 인공지능에 의한 차별을 야기할 수 있다. 알고리즘이 겉보기에 인간 행위자와 독립적으로 작동하는 것처럼 보일지라도, 인간 행위자가 그에 개입하고 있음을 밝힌다. 인간 행위자가 편향된 입력 자료를 사용하거나 편향된 설계를 할 수 있다는 점에서, 알고리즘의 편향성은 인간 행위자의 편향성과 불가분의 관계이기 때문이다. 그러므로 인간 행위자는 인공지능에 의한 차별문제에 대응할 책임이 있으며, 책임 역량을 갖고 있다. 마지막으로 인공지능에 의한 차별 논의를 위한 철학 연구의 추후 과제를 정리한다. (shrink)

This chapter focuses on one of the common fallacies in Western philosophy, “confirmation bias”. Confirmation bias is the human tendency only to look for evidence that confirms what one wants to believe or what one already thinks is true. Usually people are not too keen to look for evidence against what they want to believe is true. The human propensity for self‐delusion is strong. When one is confronted with sufficient evidence against some belief that one holds, what one (...) should do is reject that belief ‐ at least if one is concerned with truth. When we surround ourselves with people and views that already agree with our own, we can draw erroneous conclusions. Confirmation bias greatly hinders one's ability to find the truth. Confirmation bias and evidence denial is why we start “seeing something everywhere” once you learn about it. The more scrutiny a belief can withstand, the more likely it is that it's true. (shrink)

The present comment concerns a recent AI & Ethics article which purports to report evidence of speciesist bias in various popular computer vision (CV) and natural language processing (NLP) machine learning models described in the literature. I examine the authors' analysis and show it, ironically, to be prejudicial, often being founded on poorly conceived assumptions and suffering from fallacious and insufficiently rigorous reasoning, its superficial appeal in large part relying on the sequacity of the article's target readership.

Our understanding of implicit bias and how to measure it has yet to be settled. Various debates between cognitive scientists are unresolved. Moreover, the public’s understanding of implicit bias tests continues to lag behind cognitive scientists’. These discrepancies pose potential problems. After all, a great deal of implicit bias research has been publicly funded. Further, implicit bias tests continue to feature in discourse about public- and private-sector policies surrounding discrimination, inequality, and even the purpose of science. (...) We aim to do our part by reconstructing some of the recent arguments in ordinary language and then revealing some of the operative norms or values that are often hidden beneath the surface of these arguments. This may help the public learn more about the science of implicit bias. It may also help both laypeople and scientists reflect on the values, interests, and stakeholders involved in establishing, justifying, and communicating scientific research. (shrink)

In this paper, we show how game theoretic work on conversation combined with a theory of discourse structure provides a framework for studying interpretive bias and how bias affects the production and interpretation of linguistic content. We model the influence of author bias on the discourse content and structure of the author’s linguistic production and interpreter bias on the interpretation of ambiguous or underspecified elements of that content and structure. Interpretive bias is an essential feature (...) of learning and understanding but also something that can be exploited to pervert or subvert the truth. We develop three types of games to understand and to analyze a range of interpretive biases, the factors that contribute to them, and their strategic effects. (shrink)

When making decisions, humans can observe many kinds of information about others' activities, but their effects on performance are not well understood. We investigated social learning strategies using a simple problem-solving task in which participants search a complex space, and each can view and imitate others' solutions. Results showed that participants combined multiple sources of information to guide learning, including payoffs of peers' solutions, popularity of solution elements among peers, similarity of peers' solutions to their own, and relative (...) payoffs from individual exploration. Furthermore, performance was positively associated with imitation rates at both the individual and group levels. When peers' payoffs were hidden, popularity and similarity biases reversed, participants searched more broadly and randomly, and both quality and equity of exploration suffered. We conclude that when peers' solutions can be effectively compared, imitation does not simply permit scrounging, but it can also facilitate propagation of good solutions for further cumulative exploration. (shrink)

In this paper we make the case for the emergence of novel kind of bias with the use of algorithmic decision-making systems. We argue that the distinctive generative process of feature creation, characteristic of machine learning (ML), contorts feature parameters in ways that can lead to emerging feature spaces that encode novel algorithmic bias involving already marginalized groups. We term this bias _assembled bias._ Moreover, assembled biases are distinct from the much-discussed algorithmic bias, both (...) in source (training data versus feature creation) and in content (mimics of extant societal bias versus reconfigured categories). As such, this problem is distinct from issues arising from bias-encoding training feature sets or proxy features. Assembled bias is not epistemically transparent in source or content. Hence, when these ML models are used as a basis for decision-making in social contexts, algorithmic fairness concerns are compounded. (shrink)

Prediction-based decisions, which are often made by utilizing the tools of machine learning, influence nearly all facets of modern life. Ethical concerns about this widespread practice have given rise to the field of fair machine learning and a number of fairness measures, mathematically precise definitions of fairness that purport to determine whether a given prediction-based decision system is fair. Following Reuben Binns (2017), we take ‘fairness’ in this context to be a placeholder for a variety of normative egalitarian (...) considerations. We explore a few fairness measures to suss out their egalitarian roots and evaluate them, both as formalizations of egalitarian ideas and as assertions of what fairness demands of predictive systems. We pay special attention to a recent and popular fairness measure, counterfactual fairness, which holds that a prediction about an individual is fair if it is the same in the actual world and any counterfactual world where the individual belongs to a different demographic group (cf. Kusner et al. (2018)). (shrink)

Behavior in one-shot coordination games with common knowledge labels can be described by theories of salience and focal points. Behavior in repeated games, including coordination games, can be explained by theories of learning. This paper considers games in which both theories apply, repeated coordination games with common knowledge labels. The research question asks how players combine the two sources of information—salience and the history of play—when making their choices. We specifically ask whether salience, normally considered as a one-shot strategy, (...) continues to influence players’ actions beyond the first round, even while the player might learn from the history of play. We explore two possible mechanisms for such a continuing effect of salience: via an influence on prior beliefs, and/or via a bias, given beliefs. Regression analysis of individual-level choices shows that salience, normally considered only in the context of one-shot games, does exert a lasting effect, with the precise mechanism depending on the details of the game. (shrink)

This study investigates the acquisition of integrated object manipulation and categorization abilities through a series of experiments in which human adults and artificial agents were asked to learn to manipulate two-dimensional objects that varied in shape, color, weight, and color intensity. The analysis of the obtained results and the comparison of the behavior displayed by human and artificial agents allowed us to identify the key role played by features affecting the agent/environment interaction, the relation between category and action development, and (...) the role of cognitive biases originating from previous knowledge. (shrink)