Scientists cannot devise theories, construct models, propose explanations, make predictions, or even carry out observations, without first classifying their subject matter. The goal of scientific taxonomy is to come up with classification schemes that conform to nature's own. Another way of putting this is that science aims to devise categories that correspond to 'natural kinds.' The interest in ascertaining the real kinds of things in nature is as old as philosophy itself, but it takes on a different guise when one (...) adopts a naturalist stance in philosophy, that is when one looks closely at scientific practice and takes it as a guide for identifying natural kinds and investigating their general features. This Element surveys existing philosophical accounts of natural kinds, defends a naturalist alternative, and applies it to case studies in a diverse set of sciences. This title is also available as Open Access on Cambridge Core. (shrink)

Detailed examinations of scientific practice have revealed that the use of idealized models in the sciences is pervasive. These models play a central role in not only the investigation and prediction of phenomena, but in their received scientific explanations as well. This has led philosophers of science to begin revising the traditional philosophical accounts of scientific explanation in order to make sense of this practice. These new model-based accounts of scientific explanation, however, raise a number of key questions: Can the (...) fictions and falsehoods inherent in the modeling practice do real explanatory work? Do some highly abstract and mathematical models exhibit a noncausal form of scientific explanation? How can one distinguish an exploratory "how-possibly" model explanation from a genuine "how-actually" model explanation? Do modelers face tradeoffs such that a model that is optimized for yielding explanatory insight, for example, might fail to be the most predictively accurate, and vice versa? This chapter explores the various answers that have been given to these questions. (shrink)

This chapter subsumes David Marr’s levels of analysis account of explanation in cognitive science under the framework of mechanistic explanation: Answering the questions that define each one of Marr’s three levels is tantamount to describing the component parts and operations of mechanisms, as well as their organization, behavior, and environmental context. By explicating these questions and showing how they are answered in several different cognitive science research programs, this chapter resolves some of the ambiguities that remain in Marr’s account, and (...) shows that many different areas and traditions of cognitive scientific research can be unified under the mechanistic framework. (shrink)

In this paper, we explore the link between understanding and transformative decisions. Paul (2014) suggests that one important aspect of making some decisions is that we make them not just on the basis of what data from other people tell us, but based on our own acquaintance with how the decision affects us. In this paper, we draw out a parallel between the sort of reasoning that Paul argues is required for authentic decision making and the sort of epistemic grasp (...) of a subject matter that is required for understanding. The central claim of this paper is that one’s ability to make a decision authentically is proportional to one’s ability to model what it is like to have the possible resultant experience. We endorse a broad notion according to which one can model something by thinking about relevantly similar things. (shrink)

This paper makes a phenomenological contribution to the distinction between personal and subpersonal types of explanation. I expound the little-known fact that Husserl gives an account of personal level explanation via his exposition of our capacity to express the understanding of another’s motivational nexus when we are in the personalistic attitude. I show that Husserl’s unique exposition of the motivational nexus conveys its concrete, internally coherent, and intentional nature, involving relationships amongst the sense contents of acts of consciousness. Moreover, the (...) motivational nexus is a generative space of possibility and choice. I show that, for these reasons, motivational explanation is not causal, nor deductive nomological, nor does it reduce to subpersonal explanation. I finish with the comment that the uniqueness of personal level explanation points towards the possibility that the human sciences ought also employ types of explanations not found in natural sciences. (shrink)

Four experiments investigate the folk concept of “understanding,” in particular when and why it is deployed differently from the concept of knowledge. We argue for the positions that people have higher demands with respect to explanatory depth when it comes to attributing understanding, and that this is true, in part, because understanding attributions play a functional role in identifying experts who should be heeded with respect to the general field in question. These claims are supported by our findings that people (...) differentially withhold attributions of understanding when the object of attribution has minimal explanatory information. We also show that this tendency significantly correlates with people’s willingness to defer to others as potential experts. This work bears on a pressing issue in epistemology concerning the place and value of understanding. Our results also provide reason against positing a simple equation of knowledge and understanding. We contend that, because deference plays a crucial role in many aspects of everyday reasoning, the fact that we use understanding attributions to demarcate experts reveals a potential mechanism for achieving our epistemic aims in many domains. (shrink)

Neuroscience constrains psychology, but stating these constraints with precision is not simple. Here I consider whether mechanistic analysis provides a useful way to integrate models of cognitive and neural structure. Recent evidence suggests that cognitive systems map onto overlapping, distributed networks of brain regions. These highly entangled networks often depart from stereotypical mechanistic behaviors. While this casts doubt on the prospects for classical mechanistic integration of psychology and neuroscience, I argue that it does not impugn a realistic interpretation of either (...) type of model. Cognitive and neural models may depict different, but equally real, causal structures within the mind/brain. (shrink)

I propose a cautionary assessment of the recent debate concerning the impact of the dynamical approach on philosophical accounts of scientific explanation in the cognitive sciences and, particularly, the cognitive neurosciences. I criticize the dominant mechanistic philosophy of explanation, pointing out a number of its negative consequences: In particular, that it doesn’t do justice to the field’s diversity and stage of development, and that it fosters misguided interpretations of dynamical models’ contribution. In order to support these arguments, I analyze a (...) case study in computational neuroethology and show why it should not be understood through a mechanistic lens; I specially focus on Zednik’s mechanistic interpretation of the case study. In addition, I argue for a greater appreciation of the relation between explanation and other epistemic goals in the field, as well as an increased sensitivity towards the associated contextual factors. (shrink)

Recent extensions of mechanistic explanation into psychology suggest that cognitive models are only explanatory insofar as they map neatly onto, and serve as scaffolding for more detailed neural models. Filling in those neural details is what these accounts take the integration of cognitive psychology and neuroscience to mean, and they take this process to be seamless. Critics of this view have given up on cognitive models possibly explaining mechanistically in the course of arguing for cognitive models having explanatory value independent (...) of how well they align with neural mechanisms. We can have things both ways, however. The problem with seamless integration accounts is their seamlessness, not that they take cognitive models to be mechanistic. A non-componential view of mechanisms allows for cognitive and neural models that cross cut one another, and for cognitive models that don’t decompose into parts. I illustrate the inadequacy of seamless accounts of integration by contrasting how “filter” models of attention in psychology and of sodium channels in neuroscience developed; by questioning whether the mappings generated by neuroimaging subtraction studies achieve integration; and by reinterpreting the evidence for cognitive models of memory having been successfully integrated with neural models. I argue that the integrations we can realistically expect are more partial, patchy, and full of loose threads than the mosaic unity Craver describes. (shrink)

In light of recent criticisms by Woodward (2017) and Rescorla (2018), we examine the relationship between mechanistic explanation and phenomenological laws. We disambiguate several uses of the phrase “phenomenological law” and show how a mechanistic theory of explanation sorts them into those that are and are not explanatory. We also distinguish the problem of phenomenological laws from arguments about the explanatory power of purely phenomenal models, showing that Woodward and Rescorla conflate these problems. Finally, we argue that the temptation to (...) pit mechanistic and interventionist theories of explanation against one another occludes important and scientifically relevant research questions. (shrink)

ABSTRACT Proponents of mechanistic explanation have recently suggested that all explanation in the cognitive sciences is mechanistic, even functional explanation. This last claim is surprising, for functional explanation has traditionally been conceived as autonomous from the structural details that mechanistic explanations emphasize. I argue that functional explanation remains autonomous from mechanistic explanation, but not for reasons commonly associated with the phenomenon of multiple realizability. 1Introduction 2Mechanistic Explanation: A Quick Primer 3Functional Explanation: An Example 4Autonomy as Lack of Constraint 5The Price (...) of Autonomy 6Another Argument against Autonomy 7Conclusion: Autonomy and Multiple Realization. (shrink)

An increasing number of philosophers have promoted the idea that mechanism provides a fruitful framework for thinking about the explanatory contributions of computational approaches in cognitive neuroscience. For instance, Piccinini and Bahar :453–488, 2013) have recently argued that neural computation constitutes a sui generis category of physical computation which can play a genuine explanatory role in the context of investigating neural and cognitive processes. The core of their proposal is to conceive of computational explanations in cognitive neuroscience as a subspecies (...) of mechanistic explanations. This paper identifies several challenges facing their mechanistic account and sketches an alternative way of thinking about the epistemic roles of computational approaches used in the study of brain and cognition. Drawing on examples from both low-level and systems-level computational neuroscience, I argue that at least some computational explanations of neural and cognitive processes are partially independent from mechanistic constraints. (shrink)

I note the multitude of ways in which, beginning with the classic paper by Machamer et al., the mechanists have qualify their methodological dicta, and limit the vulnerability of their claims by strategic vagueness regarding their application. I go on to generalize a version of the mechanist requirement on explanations due to Craver and Kaplan :601–627, 2011) in cognitive and systems neuroscience so that it applies broadly across the life sciences in accordance with the view elaborated by Craver and Darden (...) in In Search of Mechanisms. I then go on to explore what ramifications their mechanist requirement on explanations may have for explanatory “dependencies” reported in biology and the special sciences. What this exploration suggests is that mechanism threatens to eliminate instead of underwrite a large number of such “dependencies” reported in higher-levels of biology and the special sciences. I diagnose the source of this threat in mechanism’s demand that explanations identify nested causal differences makers in mechanisms, their components, the components further components, and so forth. Finally, I identify the “love–hate” relationship mechanism must have with functional explanation, and show how it makes mechanism an extremely interesting thesis indeed. (shrink)

The medium‐independence of computational descriptions has shaped common conceptions of computational explanation. So long as our goal is to explain how a system successfully carries out its computations, then we only need to describe the abstract series of operations that achieve the desired input–output mapping, however they may be implemented. It is argued that this abstract conception of computational explanation cannot be applied to so‐called real‐time computing systems, in which meeting temporal deadlines imposed by the systems with which a device (...) interfaces are constitutive of the computing tasks that a device performs. Instead, real‐time computing reveals the need for alternative conceptions of computational explanation, as well as computational implementation, that eschew medium‐independence. (shrink)

Interventionism is a theory of causal explanation developed by Woodward and Hitchcock. I defend an interventionist perspective on the causal explanations offered within scientific psychology. The basic idea is that psychology causally explains mental and behavioral outcomes by specifying how those outcomes would have been different had an intervention altered various factors, including relevant psychological states. I elaborate this viewpoint with examples drawn from cognitive science practice, especially Bayesian perceptual psychology. I favorably compare my interventionist approach with well-known nomological and (...) mechanistic theories of psychological explanation. (shrink)

An important strand in philosophy of science takes scientific explanation to consist in the conveyance of some kind of information. Here I argue that this idea is also implicit in some core arguments of mechanists, some of whom are proponents of an ontic conception of explanation that might be thought inconsistent with it. However, informational accounts seem to conflict with some lay and scientific commonsense judgments and a central goal of the theory of explanation, because information is relative to the (...) background knowledge of agents. Sometimes we make lay judgments about whether a model is an explanation simpliciter, not just an explanation relative to some particular agent. And as philosophers of explanation, we would like a philosophical account to tell us when a model is an explanation simpliciter, not just when a model is an explanation relative to some particular agent. Thus, even if one’s account of explanation is not concerned with explanation qua communicative or speech act, the account’s reliance on the concept of information generates a prima facie conflict between the claims that 1) explanation is the conveyance of information, 2) information is relative to the background knowledge of an agent, and 3) some models are explanations not relative to the background knowledge of any particular agent. I sketch a solution to this puzzle by distinguishing informationally what I call “explanation simpliciter” from what I call “explanation-to,” relativizing the latter to an individual’s background knowledge and the former to what I call “total scientific background knowledge”. (shrink)

Mechanistic explanations satisfy widely held norms of explanation: the ability to manipulate and answer counterfactual questions about the explanandum phenomenon. A currently debated issue is whether any nonmechanistic explanations can satisfy these explanatory norms. Weiskopf argues that the models of object recognition and categorization, JIM, SUSTAIN, and ALCOVE, are not mechanistic yet satisfy these norms of explanation. In this article I argue that these models are mechanism sketches. My argument applies recent research using model-based functional magnetic resonance imaging, a novel (...) neuroimaging method whose significance for current debates on psychological models and mechanistic explanation has yet to be explored. (shrink)

This article demonstrates that non-mechanistic, dynamical explanations are a viable approach to explanation in the special sciences. The claim that dynamical models can be explanatory without reference to mechanisms has previously been met with three lines of criticism from mechanists: the causal relevance concern, the genuine laws concern, and the charge of predictivism. I argue, however, that these mechanist criticisms fail to defeat non-mechanistic, dynamical explanation. Using the examples of Haken et al.’s model of bimanual coordination, and Thelen et al.’s (...) dynamical field model of infant perseverative reaching, I show how each mechanist criticism fails once the standards of Woodward’s interventionist framework are applied to dynamical models. An even-handed application of Woodwardian interventionism reveals that dynamical models are capable of producing genuine explanations without appealing to underlying mechanistic details. 1Introduction2Interventionism and Mechanistic Explanation 2.1Causal relevance and ideal interventions2.2Invariance2.3Explanation3Covering-Laws and Dynamical Explanation 3.1Dynamical models3.2Covering-law explanation3.3Prediction4Causal Relevance 4.1The causal relevance concern4.2Intervening on dynamical models4.3Test case I: The Haken–Kelso–Bunz model4.4Test case II: Dynamical field model5Genuine Laws 5.1The genuine laws concern5.2Using invariance in place of laws5.3Test case I: The Haken–Kelso–Bunz model5.4Test case II: Dynamical field model6Prediction 6.1Predictivism6.2Crude and invariant prediction7Interventionist Criticism of the Haken–Kelso–Bunz Model8Dynamical Explanation8Conclusion. (shrink)

It is argued that once biological systems reach a certain level of complexity, mechanistic explanations provide an inadequate account of many relevant phenomena. In this article, I evaluate such claims with respect to a representative programme in systems biological research: the study of regulatory networks within single-celled organisms. I argue that these networks are amenable to mechanistic philosophy without need to appeal to some alternate form of explanation. In particular, I claim that we can understand the mathematical modelling techniques of (...) systems biologists as part of a broader practice of constructing and evaluating mechanism schemas. This argument is elaborated by considering the case of bacterial chemotactic networks, where some research has been interpreted as explaining phenomena by means of abstract design principles. (shrink)

We analyse Hutto & Myin's three arguments against computationalism [Hutto, D., E. Myin, A. Peeters, and F. Zahnoun. Forthcoming. “The Cognitive Basis of Computation: Putting Computation In Its Place.” In The Routledge Handbook of the Computational Mind, edited by M. Sprevak, and M. Colombo. London: Routledge.; Hutto, D., and E. Myin. 2012. Radicalizing Enactivism: Basic Minds Without Content. Cambridge, MA: MIT Press; Hutto, D., and E. Myin. 2017. Evolving Enactivism: Basic Minds Meet Content. Cambridge, MA: MIT Press]. The Hard Problem (...) of Content targets computationalism that relies on semantic notion of computation, claiming that it cannot account for the natural origins of content. The Intentionality Problem is targeted against computationalism using non-semantic accounts of computation, arguing that it fails in explaining intentionality. Theion Problem claims that causal interaction between concrete physical processes and abstract computational properties is problematic. We argue that these a... (shrink)

The philosophical literature on scientific explanation in neuroscience has been dominated by the idea of mechanisms. The mechanist philosophers often claim that neuroscience is in the business of finding mechanisms. This view has been challenged in numerous ways by showing that there are other successful and widespread explanatory strategies in neuroscience. However, the empirical evidence for all these claims was hitherto lacking. Empirical evidence about the pervasiveness and uses of various explanatory strategies in neuroscience is particularly needed because examples and (...) case studies that are used to illustrate philosophical claims so far tend to be hand-picked. The risk of confirmation bias is therefore considerable: when looking for white swans, all one finds is that swans are white. The more systematic quantitative and qualitative bibliometric study of a large body of relevant literature that we present in this paper can put such claims into perspective. Using bibliometric tools, we identify the typical linguistic patterns used in the alleged mechanistic, dynamical, and topological explanations in the literature, their preponderance and how they change over time. Our findings show abundant use of mechanistic language, but also the presence of a significant neuroscience literature using topological and dynamical explanatory language, which grows over time and increasingly differentiates from each other and from mechanistic explanations. (shrink)

Recently, it has been provocatively claimed that dynamical modeling approaches signal the emergence of a new explanatory framework distinct from that of mechanistic explanation. This paper rejects this proposal and argues that dynamical explanations are fully compatible with, even naturally construed as, instances of mechanistic explanations. Specifically, it is argued that the mathematical framework of dynamics provides a powerful descriptive scheme for revealing temporal features of activities in mechanisms and plays an explanatory role to the extent it is deployed for (...) this purpose. It is also suggested that more attention should be paid to the distinctive methodological contributions of the dynamical framework including its usefulness as a heuristic for mechanism discovery and hypothesis generation in contemporary neuroscience and biology. (shrink)

Pseudo‐mechanistic explanations in psychology and cognitive neuroscienceThis paper focuses on the level of systems/cognitive neuroscience. It argues that the great majority of explanations in psychology and cognitive neuroscience is “pseudo‐mechanistic.” On the basis of various case studies, Hommel argues that cognitive neuroscience should move beyond what he calls an “Aristotelian phase” to become a mature “Galilean” science seeking to discover actual mechanisms of cognitive phenomena.

This paper advances the view that some explanations in cognitive science are extra-mathematical explanations. Demonstrating the plausibility of this interpretation centers around certain efficient coding cases which ineliminably enlist information theoretic laws, facts and theorems to identify in-principle, mathematical constraints on neuronal information processing capacities. The explanatory structure in these cases is shown to parallel other putative instances of mathematical explanation. The upshot for cognitive mathematical explanations is thus two-fold: first, the view capably rebuts standard mechanistic objections to non-mechanistic explanation; (...) second, this view serves to importantly widen the possibility space for non-mechanistic forms of explanation in cognitive science. (shrink)

In this paper, I explicate and discuss performance-similarity reasoning as a strategy for mechanism schema evaluation, understood in Lindley Darden’s sense. This strategy involves inferring hypotheses about the mechanism responsible for cognitive capacities from premises describing the performance of those capacities; performance-similarity reasoning is a type of Inference to the Best Explanation, or IBE. Two types of such inferences are distinguished: one in which the performance of two systems is compared, and another when the performance of two systems under intervention (...) is compared. Both types of inferences are illustrated with examples taken from cognitive science. I conclude that performance-similarity reasoning is an important strategy for evaluating mechanistic hypotheses. (shrink)

Explanations are supposed to provide us with understanding. It is common to make a distinction between genuine, scientific understanding, and the phenomenological, or ‘aha’ notion of understanding, where the former is considered epistemically relevant, the latter irrelevant. I argue that there is a variety of phenomenological understanding that does play a positive epistemic role. This phenomenological understanding involves a similarity between bodily sensations that is used as evidence for mechanistic hypotheses. As a case study, I will consider 17th and 18th (...) century research into the mechanism behind the electric eel’s power to shock. Se supone que las explicaciones nos dotan de comprensión. Es habitual hacer una distinción entre compresión cientifica, genuina, y la noción fenomenológica, o ‘ahá’, de comprensión, donde la primera se considera epistemológicamente relevante y esta última irrelevante. Argumento que hay un tipo de compresión fenomenológica que sí juega un papel positivo epistémico. Esta comprensión epistemológica conlleva cierta semejanza con las sensaciones corporales que se usan como evidencia de algunas hipótesis mecanicistas. Como caso de estudio, consideraré la investigación en los siglos xvii y xviii del mecanismo tras la capacidad de la anguila eléctrica para provocar un shock. (shrink)

A common misunderstanding of the selected effects theory of function is that natural selection operating over an evolutionary time scale is the only functionbestowing process in the natural world. This construal of the selected effects theory conflicts with the existence and ubiquity of neurobiological functions that are evolutionary novel, such as structures underlying reading ability. This conflict has suggested to some that, while the selected effects theory may be relevant to some areas of evolutionary biology, its relevance to neuroscience is (...) marginal. This line of reasoning, however, neglects the fact that synapses, entire neurons, and potentially groups of neurons can undergo a type of selection analogous to natural selection operating over an evolutionary time scale. In the following, I argue that neural selection should be construed, by the selected effect theorist, as a distinct type of function-bestowing process in addition to natural selection. After explicating a generalized selected effects theory of function and distinguishing it from similar attempts to extend the selected effects theory, I do four things. First, I show how it allows one to identify neural selection as a distinct function-bestowing process, in contrast to other forms of neural structure formation such as neural construction. Second, I defend the view from one major criticism, and in so doing I clarify the content of the view. Third, I examine drug addiction to show the potential relevance of neural selection to neuroscientific and psychological research. Finally, I endorse a modest pluralism of function concepts within biology. (shrink)

In this paper, I defend an updated account of functional kinds, initially presented by Daniel Weiskopf, from the criticism that functional kinds will not qualify as scientific kinds. An important part of Weiskopf’s account is that functional kinds are multiply realisable. The criticisms I consider avoid discussion of multiple realisability. Instead, it is argued that functional kinds carry inferior counterfactual profiles when compared to other accounts of kinds. I respond to this charge by arguing that this criticism fails to take (...) into consideration the role that multiple realisability can play in providing important explanatory counterfactuals. I do so by highlighting some points made by Lauren Ross that highlight where multiple realisability is explanatorily pertinent. I then argue that the criticisms of Weiskopf’s account fail to establish the explanatory inferiority of functional kinds. (shrink)

Which notion of computation (if any) is essential for explaining cognition? Five answers to this question are discussed in the paper. (1) The classicist answer: symbolic (digital) computation is required for explaining cognition; (2) The broad digital computationalist answer: digital computation broadly construed is required for explaining cognition; (3) The connectionist answer: sub-symbolic computation is required for explaining cognition; (4) The computational neuroscientist answer: neural computation (that, strictly, is neither digital nor analogue) is required for explaining cognition; (5) The extreme (...) dynamicist answer: computation is not required for explaining cognition. The first four answers are only accurate to a first approximation. But the “devil” is in the details. The last answer cashes in on the parenthetical “if any” in the question above. The classicist argues that cognition is symbolic computation. But digital computationalism need not be equated with classicism. Indeed, computationalism can, in principle, range from digital (and analogue) computationalism through (the weaker thesis of) generic computationalism to (the even weaker thesis of) digital (or analogue) pancomputationalism. Connectionism, which has traditionally been criticised by classicists for being non-computational, can be plausibly construed as being either analogue or digital computationalism (depending on the type of connectionist networks used). Computational neuroscience invokes the notion of neural computation that may (possibly) be interpreted as a sui generis type of computation. The extreme dynamicist argues that the time has come for a post-computational cognitive science. This paper is an attempt to shed some light on this debate by examining various conceptions and misconceptions of (particularly digital) computation. (shrink)

A leading theoretical framework for naturalistic explanation of mind holds that we explain the mind by positing progressively "stupider" capacities ("homunculi") until the mind is "discharged" by means of capacities that are not intelligent at all. The so-called homuncular fallacy involves violating this procedure by positing the same capacities at subpersonal levels. I argue that the homuncular fallacy is not a fallacy, and that modern-day homunculi are idle posits. I propose an alternative view of what naturalism requires that reflects how (...) the cognitive sciences are actually integrating mind and matter. (shrink)

In this essay I propose that what design principles in systems biology and systems neuroscience do is to present abstract characterizations of mechanisms, and thereby facilitate mechanistic explanation. To show this, one design principle in systems neuroscience, i.e., the multilayer perceptron, is examined. However, Braillard contends that design principles provide a sort of non-mechanistic explanation due to two related reasons: they are very general and describe non-causal dependence relationships. In response to this, I argue that, on the one hand, all (...) mechanisms are more or less general, and on the other, many design principles are causal systems. (shrink)

It is generally accepted that, in the cognitive and neural sciences, there are both computational and mechanistic explanations. We ask how computational explanations can integrate into the mechanistic hierarchy. The problem stems from the fact that implementation and mechanistic relations have different forms. The implementation relation, from the states of an abstract computational system to the physical, implementing states is a homomorphism mapping relation. The mechanistic relation, however, is that of part/whole; the explaining features in a mechanistic explanation are the (...) components of the explanandum phenomenon and their causal organization. Moreover, each component in one level of mechanism is constituted and explained by components of an underlying level of mechanism. Hence, it seems, computational variables and functions cannot be mechanistically explained by the medium-dependent states and properties that implement them. How then, do the computational and the implementational integrate to create the mechanistic hierarchy? After explicating the general problem, we further demonstrate it through a concrete example, of reinforcement learning, in the cognitive and neural sciences. We then examine two possible solutions. On one solution, the mechanistic hierarchy embeds at the same levels computational and implementational properties. This picture fits with the view that computational explanations are mechanistic sketches. On the other solution, there are two separate hierarchies, one computational and another implementational, which are related by the implementation relation. This picture fits with the view that computational explanations are functional and autonomous explanations. It is less clear how these solutions fit with the view that computational explanations are full-fledged mechanistic explanations. Finally, we argue that both pictures are consistent with the reinforcement learning example, but that scientific practice does not align with the view that computational models are merely mechanistic sketches. (shrink)

Philosophers have traditionally treated physicalism as an empirically informed metaphysical thesis. This approach faces a well-known problem often referred to as Hempel’s dilemma: formulations of physicalism tend to be either false or indeterminate. The generally preferred strategy to address this problem involves an appeal to a hypothetical complete and ideal physical theory. After demonstrating that this strategy is not viable, I argue that we should redefine physicalism as an interdisciplinary research program seeking to explain the mental in terms of the (...) physical that encompasses the physical sciences, the psychological and brain sciences, and philosophy. Redefining physicalism in this way improves upon previous reconstructive accounts while avoiding the indeterminacy associated with orthodox forms of futurist physicalism. (shrink)

I challenge the dominant understanding of what it means to say two thoughts are associated. The two views that dominate the current literature treat association as a kind of mechanism that drives sequences of thought. The first, which I call reductive associationism, treats association as a kind of neural mechanism. The second treats association as a feature of the kind of psychological mechanism associative processing. Both of these views are inadequate. I argue that association should instead be seen as a (...) highly abstract filler term, standing in for causal relations between representational states in a system. Associations, so viewed, could be implemented by many different mechanisms. I outline the role that this view gives associative models as part of a top-down characterization of psychological processes of any kind and of any complexity. (shrink)

Life-science phenomena are often explained by specifying the mechanisms that bring them about. The new mechanistic philosophers have done much to substantiate this claim and to provide us with a better understanding of what mechanisms are and how they explain. Although there is disagreement among current mechanists on various issues, they share a common core position and a seeming commitment to some form of scientific realism. But is such a commitment necessary? Is it the best way to go about mechanistic (...) explanation? In this article, we propose an alternative antirealist account that also fits explanatory practice in the life sciences. We pay special attention to mechanistic models, i.e. scientific models that involve a mechanism, and to the role of coherence considerations in building such models. To illustrate our points, we consider the mechanism for the action potential. 1 Introduction2 Some Core Features of Mechanistic Explanation3 Scientific Realism and Mechanistic Explanation4 Antirealist Mechanistic Explanation: The Case of the Action Potential5 Some Outstanding Issues for the Antirealist Mechanist6 Two Problems for the Realist Mechanist7 Conclusions. (shrink)

In this paper, I examine Reinforcement Learning modelling practice in psychiatry, in the context of alcohol use disorders. I argue that the epistemic roles RL currently plays in the development of psychiatric classification and search for explanations of clinically relevant phenomena are best appreciated in terms of Chang’s account of epistemic iteration, and by distinguishing mechanistic and aetiological modes of computational explanation.

In a recent paper, Kaplan (Synthese 183:339–373, 2011) takes up the task of extending Craver’s (Explaining the brain, 2007) mechanistic account of explanation in neuroscience to the new territory of computational neuroscience. He presents the model to mechanism mapping (3M) criterion as a condition for a model’s explanatory adequacy. This mechanistic approach is intended to replace earlier accounts which posited a level of computational analysis conceived as distinct and autonomous from underlying mechanistic details. In this paper I discuss work in (...) computational neuroscience that creates difficulties for the mechanist project. Carandini and Heeger (Nat Rev Neurosci 13:51–62, 2012) propose that many neural response properties can be understood in terms of canonical neural computations. These are “standard computational modules that apply the same fundamental operations in a variety of contexts.” Importantly, these computations can have numerous biophysical realisations, and so straightforward examination of the mechanisms underlying these computations carries little explanatory weight. Through a comparison between this modelling approach and minimal models in other branches of science, I argue that computational neuroscience frequently employs a distinct explanatory style, namely, efficient coding explanation. Such explanations cannot be assimilated into the mechanistic framework but do bear interesting similarities with evolutionary and optimality explanations elsewhere in biology. (shrink)

This article examines three candidate cases of non-causal explanation in computational neuroscience. I argue that there are instances of efficient coding explanation that are strongly analogous to examples of non-causal explanation in physics and biology, as presented by Batterman, Woodward, and Lange. By integrating Lange’s and Woodward’s accounts, I offer a new way to elucidate the distinction between causal and non-causal explanation, and to address concerns about the explanatory sufficiency of non-mechanistic models in neuroscience. I also use this framework to (...) shed light on the dispute over the interpretation of dynamical models of the brain. _1_ Introduction _1.1_ Efficient coding explanation in computational neuroscience _1.2_ Defining non-causal explanation _2_ Case I: Hybrid Computation _3_ Case II: The Gabor Model Revisited _4_ Case III: A Dynamical Model of Prefrontal Cortex _4.1_ A new explanation of context-dependent computation _4.2_ Causal or non-causal? _5_ Causal and Non-causal: Does the Difference Matter? (shrink)

Executive functioning has been said to bear on a range of traditional philosophical topics, such as consciousness, thought, and action. Surprisingly, philosophers have not much engaged with the scientific literature on executive functioning. This lack of engagement may be due to several influential criticisms of that literature by Daniel Dennett, Alan Allport, and others. In this paper I argue that more recent research on executive functioning shows that these criticisms are no longer valid. The paper clears the way to a (...) more fruitful philosophical engagement with findings on the central executive system. (shrink)

How do individuals guide their activities towards some goal? Harry Frankfurt once identified the task of explaining guidance as the central problem in action theory. An explanation has proved to be elusive, however. In this paper, I show how we can marshal empirical research to make explanatory progress. I contend that human agents have a primitive capacity to guide visual attention, and that this capacity is actually constituted by a sub-individual psychological control-system: the executive system. I thus illustrate how we (...) can explain exercises of individual-level guidance by appeal to its sub-individual constitution. This opens up a new avenue for explicating agency. (shrink)

In paradigm exercises of agency, individuals guide their activities toward some goal. A central challenge for action theory is to explain how individuals guide. This challenge is an instance of the more general problem of how to accommodate individuals and their actions in the natural world, as explained by natural science. Two dominant traditions–primitivism and the causal theory–fail to address the challenge in a satisfying way. Causal theorists appeal to causation by an intention, through a feedback mechanism, in explaining guidance. (...) Primitivists postulate primitive agential capacities in their explanations. The latter neglect to explain how primitive capacities integrate with findings from natural science. The former do not explain why some feedback mechanism’s activity amounts to the agent’s guidance. In this paper I argue that both traditions should acknowledge a capacity to guide, as actually constituted by the executive system. I argue that appeals to this empirically discovered psychological system explain how individuals guide in a way that integrates with explanations from cognitive science. Individuals’ capacity to guide is embedded in the natural world through the activity of its constituent (mechanistic) components. (shrink)

The functionalist approach to kinds has suffered recently due to its association with law-based approaches to induction and explanation. Philosophers of science increasingly view nomological approaches as inappropriate for the special sciences like psychology and biology, which has led to a surge of interest in approaches to natural kinds that are more obviously compatible with mechanistic and model-based methods, especially homeostatic property cluster theory. But can the functionalist approach to kinds be weaned off its dependency on laws? Dan Weiskopf has (...) recently offered a reboot of the functionalist program by replacing its nomological commitments with a model-based approach more closely derived from practice in psychology. Roughly, Weiskopf holds that the natural kinds of psychology will be the functional properties that feature in many empirically successful cognitive models, and that those properties need not be localized to parts of an underlying mechanism. I here skeptically examine the three modeling practices that Weiskopf thinks introduce such non-localizable properties: fictionalization, reification, and functional abstraction. In each case, I argue that recognizing functional properties introduced by these practices as autonomous kinds comes at clear cost to those explanations’ counterfactual explanatory power. At each step, a tempting functionalist response is parochialism: to hold that the false or omitted counterfactuals fall outside the modeler’s explanatory aims, and so should not be counted against functional kinds. I conclude by noting the dangers this attitude poses to scientific disagreement, inviting functionalists to better articulate how the individuation conditions for functional kinds might outstrip the perspective of a single modeler. (shrink)

In artificial intelligence, recent research has demonstrated the remarkable potential of Deep Convolutional Neural Networks (DCNNs), which seem to exceed state-of-the-art performance in new domains weekly, especially on the sorts of very difficult perceptual discrimination tasks that skeptics thought would remain beyond the reach of artificial intelligence. However, it has proven difficult to explain why DCNNs perform so well. In philosophy of mind, empiricists have long suggested that complex cognition is based on information derived from sensory experience, often appealing to (...) a faculty of abstraction. Rationalists have frequently complained, however, that empiricists never adequately explained how this faculty of abstraction actually works. In this paper, I tie these two questions together, to the mutual benefit of both disciplines. I argue that the architectural features that distinguish DCNNs from earlier neural networks allow them to implement a form of hierarchical processing that I call “transformational abstraction”. Transformational abstraction iteratively converts sensory-based representations of category exemplars into new formats that are increasingly tolerant to “nuisance variation” in input. Reflecting upon the way that DCNNs leverage a combination of linear and non-linear processing to efficiently accomplish this feat allows us to understand how the brain is capable of bi-directional travel between exemplars and abstractions, addressing longstanding problems in empiricist philosophy of mind. I end by considering the prospects for future research on DCNNs, arguing that rather than simply implementing 80s connectionism with more brute-force computation, transformational abstraction counts as a qualitatively distinct form of processing ripe with philosophical and psychological significance, because it is significantly better suited to depict the generic mechanism responsible for this important kind of psychological processing in the brain. (shrink)

Deep learning is currently the most prominent and widely successful method in artificial intelligence. Despite having played an active role in earlier artificial intelligence and neural network research, philosophers have been largely silent on this technology so far. This is remarkable, given that deep learning neural networks have blown past predicted upper limits on artificial intelligence performance—recognizing complex objects in natural photographs and defeating world champions in strategy games as complex as Go and chess—yet there remains no universally accepted explanation (...) as to why they work so well. This article provides an introduction to these networks as well as an opinionated guidebook on the philosophical significance of their structure and achievements. It argues that deep learning neural networks differ importantly in their structure and mathematical properties from the shallower neural networks that were the subject of so much philosophical reflection in the 1980s and 1990s. The article then explores several different explanations for their success and ends by proposing three areas of inquiry that would benefit from future engagement by philosophers of mind and science. (shrink)

We outline a framework of multilevel neurocognitive mechanisms that incorporates representation and computation. We argue that paradigmatic explanations in cognitive neuroscience fit this framework and thus that cognitive neuroscience constitutes a revolutionary break from traditional cognitive science. Whereas traditional cognitive scientific explanations were supposed to be distinct and autonomous from mechanistic explanations, neurocognitive explanations aim to be mechanistic through and through. Neurocognitive explanations aim to integrate computational and representational functions and structures across multiple levels of organization in order to explain (...) cognition. To a large extent, practicing cognitive neuroscientists have already accepted this shift, but philosophical theory has not fully acknowledged and appreciated its significance. As a result, the explanatory framework underlying cognitive neuroscience has remained largely implicit. We explicate this framework and demonstrate its contrast with previous approaches. (shrink)

We provide an explicit taxonomy of legitimate kinds of abstraction within constitutive explanation. We argue that abstraction is an inherent aspect of adequate mechanistic explanation. Mechanistic explanations—even ideally complete ones—typically involve many kinds of abstraction and therefore do not require maximal detail. Some kinds of abstraction play the ontic role of identifying the specific complex components, subsets of causal powers, and organizational relations that produce a suitably general phenomenon. Therefore, abstract constitutive explanations are both legitimate and mechanistic.