The past decade has witnessed a growing awareness of conceptual and methodological hurdles within psychology and neuroscience that must be addressed for taxonomic and explanatory progress in understanding psychological functions to be possible. In this paper, I evaluate several recent knowledge-building initiatives aimed at overcoming these obstacles. I argue that while each initiative offers important insights about how to facilitate taxonomic and explanatory progress in psychology and neuroscience, only a “coordinated pluralism” that incorporates positive aspects of each initiative will have (...) the potential for success. (shrink)

Optogenetic techniques are described as “revolutionary” for the unprecedented causal control they allow neuroscientists to exert over neural activity in awake behaving animals. In this paper, I demonstrate by means of a case study that optogenetic techniques will only illuminate causal links between the brain and behavior to the extent that their error characteristics are known and, further, that determining these error characteristics requires comparison of optogenetic techniques with techniques having well known error characteristics and consideration of the broader neural (...) and behavioral context in which the targets of optogenetic interventions are situated. (shrink)

Crime is a serious social problem, but its causes are not exclusively social. There is growing consensus that explaining and preventing it requires interdisciplinary research efforts. Indeed, the landscape of contemporary criminology includes a variety of theoretical models that incorporate psychological, biological and sociological factors. These multi-disciplinary approaches, however, have yet to radically advance scientific understandings of crime and shed light on how to manage it. In this paper, using conceptual tools on offer in the philosophy of science in combination (...) with theoretical work represented in this special volume of Psychology, Crime and Law, I provide some perspective on why explanatory progress in criminology has remained elusive and evaluate some positive proposals for attaining it. -/- . (shrink)

In this chapter I investigate the kinds of changes that psychiatric kinds undergo when they become explanatory targets of areas of sciences that are not “mature” and are in the early stages of discovering mechanisms. The two areas of science that are the targets of my analysis are cognitive neuroscience and cognitive neurobiology.

The Morris water maze has been put forward in the philosophy of neuroscience as an example of an experimental arrangement that may be used to delineate the cognitive faculty of spatial memory (e.g., Craver and Darden, Theory and method in the neurosciences, University of Pittsburgh Press, Pittsburgh, 2001; Craver, Explaining the brain: Mechanisms and the mosaic unity of neuroscience, Oxford University Press, Oxford, 2007). However, in the experimental and review literature on the water maze throughout the history of its use, (...) we encounter numerous responses to the question of “what” phenomenon it circumscribes ranging from cognitive functions (e.g., “spatial learning”, “spatial navigation”), to representational changes (e.g., “cognitive map formation”) to terms that appear to refer exclusively to observable changes in behavior (e.g., “water maze performance”). To date philosophical analyses of the water maze have not been directed at sorting out what phenomenon the device delineates nor the sources of the different answers to the question of what. I undertake both of these tasks in this paper. I begin with an analysis of Morris’s first published research study using the water maze and demonstrate that he emerged from it with an experimental learning paradigm that at best circumscribed a discrete set of observable changes in behavior. However, it delineated neither a discrete set of representational changes nor a discrete cognitive function. I cite this in combination with a reductionist-oriented research agenda in cellular and molecular neurobiology dating back to the 1980s as two sources of the lack of consistency across the history of the experimental and review literature as to what is under study in the water maze. (shrink)

This paper offers a critique of an account of explanatory integration that claims that explanations of cognitive capacities by functional analyses and mechanistic explanations can be seamlessly integrated. It is shown that achieving such explanatory integration requires that the terms designating cognitive capacities in the two forms of explanation are stable but that experimental practice in the mind-brain sciences currently is not directed at achieving such stability. A positive proposal for changing experimental practice so as to promote such stability is (...) put forward and its implications for explanatory integration are briefly considered. (shrink)

To develop rehabilitative treatment programs for persons who have committed crimes, correctional psychologists build theoretical structures that weld theoretical ideas about the causes of criminal behavior, theoretical perspectives about appropriate targets for correctional intervention and normative assumptions about crime and the aims of correctional intervention. To differentiate the tri-partite theoretical structure with which correctional program designers' work, Ward and Durrant (2021) introduce the metatheoretical concept of “practice frameworks”. In this paper, I describe and evaluate this concept, situating my analysis within (...) the broader contexts of philosophical work on scientific theories and models and metatheoretical work on the structure of explanatory and intervention theories in correctional psychology. (shrink)

The use of neuroscientific evidence in criminal trials has been steadily increasing. Despite progress made in recent decades in understanding the mechanisms of psychological and behavioral functioning, neuroscience is still in an early stage of development and its potential for influencing legal decision-making is highly contentious. Scholars disagree about whether or how neuroscientific evidence might impact prescriptions of criminal culpability, particularly in instances in which evidence of an accused’s history of mental illness or brain abnormality is offered to support a (...) plea of not criminally responsible. In the context of these debates, philosophers and legal scholars have identified numerous problems with admitting neuroscientific evidence in legal contexts. To date, however, less has been said about the challenges of evaluating the evidence upon which integrative mechanistic explanations that bring together evidence from different areas of neuroscience are based. As we explain, current criteria for evaluating such evidence to determine its admissibility in legal contexts are inadequate. Appealing to literature in the philosophy of scientific experimentation and theoretical work in the social, cognitive and behavioral sciences, we lay the groundwork for reforming these criteria and identify some of the implications of modifying them. (shrink)

In the quest to discover the neural bases of cognition, rigorous behavioral tools are equally as important as sophisticated tools for neural intervention. This paper evaluates several episodes in the development of a novel behavioral tool for rodent cognitive testing, the rodent touchscreen operant chamber. Using conceptual tools on offer in the philosophical literature on exploratory experimentation and control, I illuminate how optimization of this behavioral tool and an understanding of the causal knowledge it may be used to generate historically (...) has involved dynamic interplay between community-driven exploratory practices operating in parallel with causal hypothesis-driven research. (shrink)

In 2007, ten world-renowned neuroscientists proposed “A Decade of the Mind Initiative.” The contention was that, despite the successes of the Decade of the Brain, “a fundamental understanding of how the brain gives rise to the mind [was] still lacking” (2007, 1321). The primary aims of the decade of the mind were “to build on the progress of the recent Decade of the Brain (1990-99)” by focusing on “four broad but intertwined areas” of research, including: healing and protecting, understanding, enriching, (...) and modeling the mind. These four aims were to be the result of “transdisciplinary and multiagency” research spanning “across disparate fields, such as cognitive science, medicine, neuroscience, psychology, mathematics, engineering, and computer science.” The proposal for a decade of the mind prompted many questions (See Spitzer 2008). In this chapter, I address three of them: (1) How do proponents of this new decade conceive of the mind? (2) Why should a decade be devoted to understanding it? (3) What should this decade look like? (shrink)

Neuroscience is a laboratory-based science that spans multiple levels of analysis from molecular genetics to behavior. At every level of analysis experiments are designed in order to answer empirical questions about phenomena of interest. Understanding the nature and structure of experimentation in neuroscience is fundamental for assessing the quality of the evidence produced by such experiments and the kinds of claims that are warranted by the data. This article provides a general conceptual framework for thinking about evidence and experimentation in (...) neuroscience with a particular focus on two research areas: cognitive neuroscience and cognitive neurobiology. (shrink)

Ian Hacking instigated a revolution in 20th century philosophy of science by putting experiments (“interventions”) at the top of a philosophical agenda that historically had focused nearly exclusively on representations (“theories”). In this paper, I focus on a set of conceptual tools Hacking (1992) put forward to understand how laboratory sciences become stable and to explain what such stability meant for the prospects of unity of science and kind discovery in experimental science. I first use Hacking’s tools to understand sources (...) of instability and disunity in rodent behavioral neuroscience. I then use them to understand recent grass-roots collaborative open science initiatives aimed at establishing stability in this research area and tease out some implications for unity of science and kind creation and discovery in cognitive neuroscience. -/- . (shrink)

What role does the concept of representation play in the contexts of experimentation and explanation in cognitive neurobiology? In this article, a distinction is drawn between minimal and substantive roles for representation. It is argued by appeal to a case study that representation currently plays a role in cognitive neurobiology somewhere in between minimal and substantive and that this is problematic given the ultimate explanatory goals of cognitive neurobiological research. It is suggested that what is needed is for representation to (...) instead play a more substantive role. (shrink)

Many neurodegenerative and neuropsychiatric diseases and other brain disorders are accompanied by impairments in high-level cognitive functions including memory, attention, motivation, and decision-making. Despite several decades of extensive research, neuroscience is little closer to discovering new treatments. Key impediments include the absence of validated and robust cognitive assessment tools for facilitating translation from animal models to humans. In this review, we describe a state-of-the-art platform poised to overcome these impediments and improve the success of translational research, the Mouse Translational Research (...) Accelerator Platform (MouseTRAP), which is centered on the touchscreen cognitive testing system for rodents. It integrates touchscreen-based tests of high-level cognitive assessment with state-of-the art neurotechnology to record and manipulate molecular and circuit level activity in vivo in animal models during human-relevant cognitive performance. The platform also is integrated with two Open Science platforms designed to facilitate knowledge and data-sharing practices within the rodent touchscreen community, touchscreencognition and mousebytes. Touchscreencognition includes the Wall, showcasing touchscreen news and publications, the Forum, for community discussion, and Training, which includes courses, videos, SOPs, and symposia. To get started, interested researchers simply create user accounts. We describe the origins of the touchscreen testing system, the novel lines of research it has facilitated, and its increasingly widespread use in translational research, which is attributable in part to knowledge-sharing efforts over the past decade. We then identify the unique features of MouseTRAP that stand to potentially revolutionize translational research, and describe new initiatives to partner with similar platforms such as McGill’s M3 platform. (shrink)

In this chapter, I argue that scientific practice in the neurosciences of cognition is not conducive to the discovery of natural kinds of cognitive capacities. The “neurosciences of cognition” include cognitive neuroscience and cognitive neurobiology, two research areas that aim to understand how the brain gives rise to cognition and behavior. Some philosophers of neuroscience have claimed that explanatory progress in these research areas ultimately will result in the discovery of the underlying mechanisms of cognitive capacities. Once such mechanistic understanding (...) is achieved, cognitive capacities purportedly will be relegated into natural kind categories that correspond to real divisions in the causal structure of the world. I provide reasons here, however, in support of the claim that the neurosciences of cognition currently are not on a trajectory for discovering natural kinds. As I explain, this has to do with how mechanistic explanations of cognitive capacities are developed. Mechanistic explanations and the kinds they explain are abstract representational byproducts of the conceptual, experimental and integrative practices of neuroscientists. If these practices are not coordinated towards developing mechanistic explanations that mirror the causal structure of the world, then natural kinds of cognitive capacities will not be discovered. I provide reasons to think that such coordination is currently lacking in the neurosciences of cognition and indicate where changes in these practices appropriate to the natural kinds ideal would be required if achieving this ideal is indeed the goal. However, I suggest that an evaluation of current practices in these research areas is suggestive that discovering natural kinds of cognitive capacities is not the goal. (shrink)

This article investigates several consequences of a recent trend in philosophy of mind to shift the relata of realization from mental state–physical state to function‐mechanism. It is shown, by applying both frameworks to the neuroscientific case study of memory consolidation, that, although this shift can be used to avoid the immediate antireductionist consequences of the traditional argument from multiple realizability, what is gained is a far more modest form of reductionism than recent philosophical accounts have intimated and neuroscientists themselves have (...) claimed. (shrink)

In this volume, leading philosophers of psychiatry examine psychiatric classification systems, including the Diagnostic and Statistical Manual of Mental Disorders, asking whether current systems are sufficient for effective diagnosis, treatment, and research. Doing so, they take up the question of whether mental disorders are natural kinds, grounded in something in the outside world. Psychiatric categories based on natural kinds should group phenomena in such a way that they are subject to the same type of causal explanations and respond similarly to (...) the same type of causal interventions. When these categories do not evince such groupings, there is reason to revise existing classifications. The contributors all question current psychiatric classifications systems and the assumptions on which they are based. They differ, however, as to why and to what extent the categories are inadequate and how to address the problem. Topics discussed include taxometric methods for identifying natural kinds, the error and bias inherent in DSM categories, and the complexities involved in classifying such specific mental disorders as "oppositional defiance disorder" and pathological gambling. -/- Contributors George Graham, Nick Haslam, Allan Horwitz, Harold Kincaid, Dominic Murphy, Jeffrey Poland, Nancy Nyquist Potter, Don Ross, Dan Stein, Jacqueline Sullivan, Serife Tekin, Peter Zachar. (shrink)

In the last decade alone, a growing body of preliminary evidence suggests that classic psychedelics (CPs) can rapidly and durably ameliorate symptoms and cognitive deficits associated with depression. However, the mechanisms by which CPs work in the brain are not well understood. Rodent translational research, in which experimental findings from rodents are translated to humans, is fundamental in achieving this goal. This chapter focuses on a representative subset of human and rodent studies investigating CPs for depression, including the various lines (...) of research that have been initiated to understand how they work. Our aim is to show that in addition to epistemic challenges that scientists face in translating findings from rodents to humans, there is also mismatch between experimental approaches used to investigate CPs in humans and rodents. We thus show current experimental practices are not conducive to mechanistic discovery. We end with a set of positive proposals to expedite the drive to translate CPs into effective treatments for depression. -/- Keywords: classic psychedelics, psychiatry, clinical trials, collaboration, extrapolation, major depressive disorder, open science, philosophy of science, rodent models, scientific practice, translation, validity . (shrink)

Scientists represent their world, grouping and organizing phenomena into classes by means of concepts. Philosophers of science have historically been interested in the nature of these concepts, the criteria that inform their application and the nature of the kinds that the concepts individuate. They also have sought to understand whether and how different systems of classification are related and more recently, how investigative practices shape conceptual development and change. Our aim in this paper is to provide a critical overview of (...) some of the key developments in this philosophical literature and identify some interesting issues it raises about the prospects of the so-called “special sciences”, including psychiatry, psychology, and the mind-brain sciences more generally, to discover natural kinds. (shrink)

Biomedical science has been remarkably successful in explaining illness by categorizing diseases and then by identifying localizable lesions such as a virus and neoplasm in the body that cause those diseases. Not surprisingly, researchers have aspired to apply this powerful paradigm to addiction. So, for example, in a review of the neuroscience of addiction literature, Hyman and Malenka (2001, p. 695) acknowledge a general consensus among addiction researchers that “[a]ddiction can appropriately be considered as a chronic medical illness.” Like other (...) diseases, “Once addiction has taken hold, it tends to follow a chronic course.” (Koob and La Moal 2006, p. ?). Working from this perspective, much effort has gone into characterizing the symptomology of addiction and the brain changes that underlie them. Evidence for involvement of dopamine transmission changes in the ventral tegmental area (VTA) and nucleus accumbens (NAc) have received the greatest attention. Kauer and Malenka (2007, p. 844) put it well: “drugs of abuse can co-opt synaptic plasticity mechanisms in brain circuits involved in reinforcement and reward processing”. Our goal in this chapter to provide an explicit description of the assumptions of medical models, the different forms they may take, and the challenges they face in providing explanations with solid evidence of addiction. <br>. (shrink)

Do neurobiologists aim to discover natural kinds? I address this question in this chapter via a critical analysis of classification practices operative across the 43-year history of research on long-term potentiation. I suggest that this 43-year history supports the idea that the structure of scientific practice surrounding LTP research has remained an obstacle to the discovery of natural kinds as philosophers of science have traditionally conceived them.

Efforts by governments, firms, and patients to deliver pioneering drugs for critical health needs face a challenge of diminishing efficiency in developing those medicines. While multi-sectoral collaborations involving firms, researchers, patients, and policymakers are widely recognized as crucial for countering this decline, existing incentives to engage in drug development predominantly target drug manufacturers and thereby do little to stimulate collaborative innovation. In this mini review, we consider the unexplored potential within pharmaceutical regulations to create novel incentives to encourage a diverse (...) set of actors from the public and private spheres to engage in the kind of collaborative knowledge exchange requisite for fostering enhanced innovation in early drug development. (shrink)

One might be inclined to assume, given the mouse donning its cover, that the behavior of interest in Nicole Nelson's book Model Behavior (2018) is that of organisms like mice that are widely used as “stand-ins” for investigating the causes of human behavior. Instead, Nelson's ethnographic study focuses on the strategies adopted by a community of rodent behavioral researchers to identify and respond to epistemic challenges they face in using mice as models to understand the causes of disordered human behaviors (...) associated with mental illness. Although Nelson never explicitly describes the knowledge production activities in which her behavioral geneticist research subjects engage as “exemplary”, the question of whether or not these activities constitute “model behavior(s)”—generalizable norms for engaging in scientific research—is one of the many thought-provoking questions raised by her book. As a philosopher of science interested in this question, I take it up here. (shrink)

Do neurobiologists aim to discover natural kinds? I address this question in this chapter via a critical analysis of classification practices operative across the 43-year history of research on long-term potentiation (LTP). I argue that this 43-year history supports the idea that the structure of scientific practice surrounding LTP research has remained an obstacle to the discovery of natural kinds.

Descriptive accounts of the nature of explanation in neuroscience and the global goals of such explanation have recently proliferated in the philosophy of neuroscience and with them new understandings of the experimental practices of neuroscientists have emerged. In this paper, I consider two models of such practices; one that takes them to be reductive; another that takes them to be integrative. I investigate those areas of the neuroscience of learning and memory from which the examples used to substantiate these models (...) are culled, and argue that the multiplicity of experimental protocols used in these research areas presents specific challenges for both models. In my view, these challenges have been overlooked largely because philosophers have hitherto failed to pay sufficient attention to fundamental features of experimental practice. I demonstrate that when we do pay attention to such features, evidence for reduction and integrative unity in neuroscience is simply not borne out. I end by suggesting some new directions for the philosophy of neuroscience that pertain to taking a closer look at the nature of neuroscientific experiments. (shrink)

In this article, I explain why stabilizing constructs is important to the success of the Research Domain Criteria Project and identify one measure for facilitating such stability.

This chapter has two aims. The first aim is to compare and contrast three different conceptual-explanatory models for thinking about mental illness with an eye towards identifying the assumptions upon which each model is based, and exploring the model’s advantages and limitations in clinical contexts. Major Depressive Disorder is used as an example to illustrate these points. The second aim is to address the question of what conceptual-theoretical framework for thinking about mental illness is most likely to facilitate the discovery (...) of causes and treatments of mental illness in research contexts. To this end, the National Institute of Mental Health’s Research Domain Criteria (RDoC) Project is briefly considered. (shrink)

This study aimed to understand the preserved elements of self-identity in persons with moderate to severe dementia attributable to Alzheimer’s disease. A semi-structured interview was developed to explore the narrative self among residents with dementia in a residential care facility and residents without dementia in an independent living setting. The interviews were transcribed verbatim from audio recordings and analyzed for common themes, while being sensitive to possible differences between the groups. The participants with dementia showed evidence of self-reference even though (...) losses in explicit memory were evident. The most noticeable difference between the two groups was time frame reference. Nonetheless, all participants showed understanding of their role in relationships and exhibited concrete preferences. Our findings suggest that memory loss and other cognitive deficits associated with moderate to severe dementia do not necessarily lead to a loss of “self.”. (shrink)

This paper is a response to a commentary by Walter Glannon (2016, Frontiers in Human Neuroscience) on my paper "Stabilizing Constructs Across Research Fields as a Way to Foster the Integrative Approach of the Research Domain Criteria Project".

In her Remarks Upon Some Writers (1743), Catharine Trotter Cockburn takes a seemingly radical stance by asserting that it is possible for atheists to be virtuous. In this paper, I examine whether or not Cockburn’s views concerning atheism commit her to a naturalistic ethics and a so-called radical enlightenment position on the independence of morality and religion. First, I examine her response to William Warburton’s critique of Pierre Bayle’s arguments concerning the possibility of a society of virtuous atheists. I argue (...) that this response shows Cockburn vacillating between a moral naturalism, on the one hand, and a theistic morality, on the other. Second, I draw on Cockburn’s letters to her niece Ann Arbuthnot, and her opinions concerning mystical ideas about “the will of God” in north-east Scotland in the mid-eighteenth century. I maintain that these letters give us a fuller appreciation of Cockburn’s naturalistic position. My conclusion is that Cockburn’s ideas concerning atheism prompt us to consider the close interplay between secular and religious principles in so-called radical ideas of the period. (shrink)

Little is known about the shaping and development of Anne Conway’s thought in relation to her early modern contemporaries. In one part of her only surviving treatise, The Principles, Conway criticises “those doctors” who uphold a dualist theory of soul and body, a mechanist conception of body (as dead and inert), and the view that the soul is “intimate present” in the body. In this paper, I argue that here she targets Walter Charleton, a well-known defender of Epicurean atomism (...) in mid-seventeenth-century England. My intention is to highlight the sophistication of Conway’s theory of soul-body relations vis-a-vis that of Charleton. (shrink)

In his De la recherche de la vérité (The Search after Truth) of 1674-75, Nicolas Malebranche makes a number of apparently contradictory remarks about women and their capacity for pure intellectual thought. On the one hand, he seems to espouse a negative biological determinism about women’s minds, and on the other, he suggests that women have the free capacity to attain truth and happiness, regardless of their physiology. In the early eighteenth-century, four English women thinkers – Anne Docwra (c. (...) 1624-1710), Mary Astell (1666-1731), Damaris Masham (1659-1708), and Mary Chudleigh (1656-1710) – engaged with Malebranche’s ideas. Their writings reveal how we might dispel the apparent contradictions in Malebranche’s thinking about women, and reaffirm the liberating potential of Cartesian philosophy for women in the early modern period. (shrink)

In The Women of Grub Street (1998), Paula McDowell highlighted the fact that the overwhelming majority of women’s texts in early modern England were polemical or religio-political in nature rather than literary in content. Since that time, the study of early modern women’s political ideas has dramatically increased, and there have been a number of recent anthologies, modern editions, and critical analyses of female political writings. As a result of Patricia Springborg’s research, Mary Astell (1668-1731) has risen to prominence as (...) one of the most astute female political commentators of her day. Springborg argues for Astell’s importance as one of the first systematic critics of the philosopher John Locke, and she interprets Astell’s political thought as a reaction to Lockean Whig politics in the early reign of Queen Anne. But some scholars claim that it is a mistake to think that Astell was chiefly preoccupied with Locke, and others suggest that reading Astell in relation to Locke alone can distort our understanding of her larger political theory. In this chapter, I argue that a more accurate picture of Astell’s political outlook emerges by examining a little studied essay, ‘A Prefatory Discourse to Dr D’Aveanant’ in her 1704 work Moderation Truly Stated. In this critique of Charles Davenant’s Essays upon Peace at Home, and War Abroad (1704), Astell provides a commentary on the political ruthlessness of Niccolo Machiavelli in his Discourses on Livy. Astell’s comments reveal that, in an age in which the political vocabulary was turning to the language of rights, property, and liberty, she still subscribed to the ancient world view, and the ethical-political language of virtue and the good. My intention is to show that these ethical dimensions of Astell’s political thought are consonant with her wider philosophy. (shrink)

Laura Specker Sullivan (2022) makes a fairly compelling case for the value of the perspectives of Buddhist practitioners in neuroethics. In this study, Tibetan Buddhist monks have been asked, among other things, whether consciousness, in brain-injured patients in a minimally conscious state, entails a duty to preserve life. In our view, some of the participants’ responses could be used to inform the bioethical debate on death determination.

Simple idealized models seem to provide more understanding than opaque, complex, and hyper-realistic models. However, an increasing number of scientists are going in the opposite direction by utilizing opaque machine learning models to make predictions and draw inferences, suggesting that scientists are opting for models that have less potential for understanding. Are scientists trading understanding for some other epistemic or pragmatic good when they choose a machine learning model? Or are the assumptions behind why minimal models provide understanding misguided? In (...) this paper, using the case of deep neural networks, I argue that it is not the complexity or black box nature of a model that limits how much understanding the model provides. Instead, it is a lack of scientific and empirical evidence supporting the link that connects a model to the target phenomenon that primarily prohibits understanding. (shrink)

What can contemporary machine learning (ML) models do? Given the proliferation of ML models in society, answering this question matters to a variety of stakeholders, both public and private. The evaluation of models' capabilities is rapidly emerging as a key subfield of modern ML, buoyed by regulatory attention and government grants. Despite this, the notion of an ML model possessing a capability has not been interrogated: what are we saying when we say that a model is able to do something? (...) And what sorts of evidence bear upon this question? -/- In this paper, we aim to answer these questions, using the capabilities of large language models (LLMs) as a running example. Drawing on the large philosophical literature on abilities, we develop an account of ML models' capabilities which can be usefully applied to the nascent science of model evaluation. Our core proposal is a conditional analysis of model abilities (CAMA): crudely, a machine learning model has a capability to X just when it would reliably succeed at doing X if it 'tried'. The main contribution of the paper is making this proposal precise in the context of ML, resulting in an operationalisation of CAMA applicable to LLMs. We then put CAMA to work, showing that it can help make sense of various features of ML model evaluation practice, as well as suggest procedures for performing fair inter-model comparisons. (shrink)

Despite its centrality in the philosophy of cognitive science, there has been little prior philosophical work engaging with the notion of representation in contemporary NLP practice. This paper attempts to fill that lacuna: drawing on ideas from cognitive science, I introduce a framework for evaluating the representational claims made about components of neural NLP models, proposing three criteria with which to evaluate whether a component of a model represents a property and operationalising these criteria using probing classifiers, a popular analysis (...) technique in NLP (and deep learning more broadly). The project of operationalising a philosophically-informed notion of representation should be of interest to both philosophers of science and NLP practitioners. It affords philosophers a novel testing-ground for claims about the nature of representation, and helps NLPers organise the large literature on probing experiments, suggesting novel avenues for empirical research. (shrink)

The development of personal technologies has recently shifted from devices that seek to capture user attention to those that aim to improve user well-being. Digital wellness technologies use the same attractive qualities of other persuasive apps to motivate users towards behaviors that are personally and socially valuable, such as exercise, wealth-management, and meaningful communication. While these aims are certainly an improvement over the market-driven motivations of earlier technologies, they retain their predecessors’ focus on influencing user behavior as a primary metric (...) of success. Digital wellness technologies are still persuasive technologies, and they do not evade concerns over whether their influence on users is ethically justified. In this paper, we describe several ethical frameworks with which to assess the justification of digital wellness technologies’ influence on users. We propose that while some technologies help users to complete tasks and satisfy immediate preferences, other technologies encourage users to reflect on the values underlying their habits and teach them to evaluate their lives’ competing demands. While the former approach to digital wellness technology is not unethical, we propose that the latter approach is more likely to lead to skillful user engagement with technology. (shrink)

In a recent letter, Dillion et. al (2023) make various suggestions regarding the idea of artificially intelligent systems, such as large language models, replacing human subjects in empirical moral psychology. We argue that human subjects are in various ways indispensable.

Under what conditions does machine learning (ML) model opacity inhibit the possibility of explaining and understanding phenomena? In this article, I argue that nonepistemic values give shape to the ML opacity problem even if we keep researcher interests fixed. Treating ML models as an instance of doing model-based science to explain and understand phenomena reveals that there is (i) an external opacity problem, where the presence of inductive risk imposes higher standards on externally validating models, and (ii) an internal opacity (...) problem, where greater inductive risk demands a higher level of transparency regarding the inferences the model makes. (shrink)

Some scholars have identified a puzzle in the writings of Mary Astell (1666–1731), a deeply religious feminist thinker of the early modern period. On the one hand, Astell strongly urges her fellow women to preserve their independence of judgement from men; yet, on the other, she insists upon those same women maintaining a submissive deference to the Anglican church. These two positions appear to be incompatible. In this paper, I propose a historical-contextualist solution to the puzzle: I argue that the (...) seeming inconsistency can be dispelled through a close examination of (i) the concepts of selfhood and self-government in Anglican women’s devotional texts of the period, and of (ii) the role that these concepts play in Astell’s feminist arguments. (shrink)

We argue that the trend toward providing users with feasible and actionable explanations of AI decisions—known as recourse explanations—comes with ethical downsides. Specifically, we argue that recourse explanations face several conceptual pitfalls and can lead to problematic explanation hacking, which undermines their ethical status. As an alternative, we advocate that explanations of AI decisions should aim at understanding.

Explainable AI (xAI) methods are important for establishing trust in using black-box models. However, recent criticism has mounted against current xAI methods that they disagree, are necessarily false, and can be manipulated, which has started to undermine the deployment of black-box models. Rudin (2019) goes so far as to say that we should stop using black-box models altogether in high-stakes cases because xAI explanations ‘must be wrong’. However, strict fidelity to the truth is historically not a desideratum in science. Idealizations (...) – the intentional distortions introduced to scientific theories and models – are commonplace in the natural sciences and are seen as a successful scientific tool. Thus, it is not falsehood qua falsehood that is the issue. In this paper, I outline the need for xAI research to engage in idealization evaluation. Drawing on the use of idealizations in the natural sciences and philosophy of science, I introduce a novel framework for evaluating whether xAI methods engage in successful idealizations or deceptive explanations (SIDEs). SIDEs evaluates whether the limitations of xAI methods, and the distortions that they introduce, can be part of a successful idealization or are indeed deceptive distortions as critics suggest. I discuss the role that existing research can play in idealization evaluation and where innovation is necessary. Through a qualitative analysis we find that leading feature importance methods and counterfactual explanations are subject to idealization failure and suggest remedies for ameliorating idealization failure. (shrink)

In this paper, we explain and showcase the promising methodology of testimonial network analysis and visualization for experimental epistemology, arguing that it can be used to gain insights and answer philosophical questions in social epistemology. Our use case is the epistemic community that discusses vaccine safety primarily in English on Twitter. In two studies, we show, using both statistical analysis and exploratory data visualization, that there is almost no neutral or ambivalent discussion of vaccine safety on Twitter. Roughly half the (...) accounts engaging with this topic are pro-vaccine, while the other half are con-vaccine. We also show that these two camps rarely engage with one another, and that the con-vaccine camp has greater epistemic reach and receptivity than the pro-vaccine camp. In light of these findings, we question whether testimonial networks as they are currently constituted on popular fora such as Twitter are living up to their promise of delivering the wisdom of crowds. We conclude by pointing to directions for further research in digital social epistemology. (shrink)

In this chapter we argue that some beliefs present a problem for the truth-aim teleological account of belief, according to which it is constitutive of belief that it is aimed at truth. We draw on empirical literature which shows that subjects form beliefs about the real world when they read fictional narratives, even when those narratives are presented as fiction, and subjects are warned that the narratives may contain falsehoods. We consider Nishi Shah’s teleologist’s dilemma and a response to it (...) from Asbjørn Steglich-Petersen which appeals to weak truth regulation as a feature common to all belief. We argue that beliefs from fiction indicate that there is not a basic level of truth regulation common to all beliefs, and thus the teleologist’s dilemma remains. We consider two objections to our argument. First, that the attitudes gained through reading fiction are not beliefs, and thus teleologists are not required to account for them in their theory. We respond to this concern by defending a doxastic account of the attitudes gained from fiction. Second, that these beliefs are in fact appropriately truth-aimed, insofar as readers form beliefs upon what they take to be author testimony. We respond to this concern by suggesting that the conditions under which one can form justified beliefs upon testimony are not met in the cases we discuss. Lastly, we gesture towards a teleological account grounded in biological function, which is not vulnerable to our argument. We conclude that beliefs from fiction present a problem for the truth-aim teleological account of belief. (shrink)

Recently, many have argued that there are certain kinds of abstract mathematical explanations that are noncausal. In particular, the irrelevancy approach suggests that abstracting away irrelevant causal details can leave us with a noncausal explanation. In this paper, I argue that the common example of Renormalization Group explanations of universality used to motivate the irrelevancy approach deserves more critical attention. I argue that the reasons given by those who hold up RG as noncausal do not stand up to critical scrutiny. (...) As a result, the irrelevancy approach and the line between casual and noncausal explanation deserves more scrutiny. (shrink)

Ethical analyses of the effects of neural interventions commonly focus on changes to personality and behavior, interpreting these changes in terms of authenticity and identity. These phenomena have led to debate among ethicists about the meaning of these terms for ethical analysis of such interventions. While these theoretical approaches have different criteria for ethical significance, they agree that patients’ reports are concerning because a sense of self is valuable. In this paper, I question this assumption. I propose that the Buddhist (...) theory of no‐self offers a novel approach to making ethical sense of patients’ claims following deep brain stimulation. This alternative approach is based on the value of insight into patterns of cause and effect among mental states and actions. (shrink)

Recent studies have explored the effectiveness of open-label placebos for a variety of conditions, including chronic pain, cancer-related fatigue and irritable bowel syndrome. OLPs are thought to sidestep traditional ethical worries about placebos because they do not involve deception: with an OLP, patients or subjects are told outright that they are not given an active substance. As deception is framed as the primary hurdle to ethical placebo use, the door is ostensibly opened to ethical studies of OLPs. In this article, (...) I suggest that even though OLPs seemingly do not involve deception, there are other ethical considerations in their clinical investigation and subsequent use. Research ethics often focusses on informed consent—of which, deception and honesty are a piece—as a means to justify research practices with human subjects. Yet, it is but one of the ethical considerations that should be taken into account. With research into placebo effects in particular, I argue that the history of clinical placebo use grounds special considerations for OLP research that go beyond respect for the autonomy of individual patients through informed consent and encompass structural concerns about the type of patient for whom a placebo has historically been thought appropriate. (shrink)

In this paper I argue against two prevailing views of Kant’s Religion. Against commentators such as Michalson and Quinn, who have argued that Kant’s project in Religion is riddled with inconsistencies and circularities, I show that a proper understanding of Kant’s views on grace reveals these do not exist. And contra commentators that attribute to Kant at best a minimalist conception of grace, I show that Kant’s view of it is remarkably robust. I argue that Kant works with three different (...) conceptions of grace. These are: a) grace and the God within, b) grace and the transformation of the fundamental orientation, and c) grace that can be laid hold of; the first and the last play a significant role in his philosophy of religion. (shrink)