Proponents of mechanistic explanation all acknowledge the importance of organization. But they have also tended to emphasize specificity with respect to parts and operations in mechanisms. We argue that in understanding one important mode of organization—patterns of causal connectivity—a successful explanatory strategy abstracts from the specifics of the mechanism and invokes tools such as those of graph theory to explain how mechanisms with a particular mode of connectivity will behave. We discuss the connection between organization, abstraction, and mechanistic explanation and (...) illustrate our claims by looking at an example from recent research on so-called network motifs. (shrink)

Mechanistic explanation has an impressive track record of advancing our understanding of complex, hierarchically organized physical systems, particularly biological and neural systems. But not every complex system can be understood mechanistically. Psychological capacities are often understood by providing cognitive models of the systems that underlie them. I argue that these models, while superficially similar to mechanistic models, in fact have a substantially more complex relation to the real underlying system. They are typically constructed using a range of techniques for abstracting (...) the functional properties of the system, which may not coincide with its mechanistic organization. I describe these techniques and show that despite being non-mechanistic, these cognitive models can satisfy the normative constraints on good explanations. (shrink)

Structural realism is considered by many realists and antirealists alike as the most defensible form of scientific realism. There are now many forms of structural realism and an extensive literature about them. There are interesting connections with debates in metaphysics, philosophy of physics and philosophy of mathematics. This entry is intended to be a comprehensive survey of the field.

The concept of mechanism is analyzed in terms of entities and activities, organized such that they are productive of regular changes. Examples show how mechanisms work in neurobiology and molecular biology. Thinking in terms of mechanisms provides a new framework for addressing many traditional philosophical issues: causality, laws, explanation, reduction, and scientific change.

David Marr's theory of vision has been widely cited by philosophers and psychologists. I have three projects in this paper. First, I try to offer a perspicuous characterization of Marr's theory. Next, I consider the implications of Marr's work for some currently popular philosophies of psychology, specifically, the "hegemony of neurophysiology view", the theories of Jerry Fodor, Daniel Dennett, and Stephen Stich, and the view that perception is permeated by belief. In the last section, I consider what the phenomenon of (...) vision must be like for Marr's project to succeed. (shrink)

Fictionalists propose that some apparently fact-stating discourses do not aim to convey factual information about the world, but rather allow us to engage in a fiction or pretense without incurring ontological commitments. Some philosophers have suggested that using mathematical, modal, or moral discourse, for example, need not commit us to the existence of mathematical objects, possible worlds, or moral facts. The mental fictionalist applies this reasoning to our mental discourse, suggesting that we can use ‘belief’ and ‘desire’ talk without committing (...) to the existence of beliefs and desires as mental entities. Most arguments for mental fictionalism are based on two key suppositions: first, that there are ontological concerns about mental entities; and second, that these ontological concerns justify a fictionalist interpretation of mental discourse. This paper challenges both suppositions and argues that the standard arguments for mental fictionalism are substantially weaker than arguments for other forms of fictionalism in the philosophical literature. (shrink)

Models occupy a central role in the scientific endeavour. Among the many purposes they serve, representation is of great importance. Many models are representations of something else; they stand for, depict, or imitate a selected part of the external world (often referred to as target system, parent system, original, or prototype). Well-known examples include the model of the solar system, the billiard ball model of a gas, the Bohr model of the atom, the Gaussian-chain model of a polymer, the MIT (...) bag model of quark confinement, the Lorenz model of the atmosphere, the Lotka-Volterra model of the predator-prey interaction, or the hydraulic model of an economy, to mention just a few. All these models represent their target systems (or selected parts of them) in one way or another. (shrink)

A Bayesian mind is, at its core, a rational mind. Bayesianism is thus well-suited to predict and explain mental processes that best exemplify our ability to be rational. However, evidence from belief acquisition and change appears to show that we do not acquire and update information in a Bayesian way. Instead, the principles of belief acquisition and updating seem grounded in maintaining a psychological immune system rather than in approximating a Bayesian processor.

The success of the Bayesian approach to perception suggests probabilistic perceptual representations. But if perceptual representation is probabilistic, why doesn't normal conscious perception reflect the full probability distributions that the probabilistic point of view endorses? For example, neurons in MT/V5 that respond to the direction of motion are broadly tuned: a patch of cortex that is tuned to vertical motion also responds to horizontal motion, but when we see vertical motion, foveally, in good conditions, it does not look at all (...) horizontal. This article argues that the best Bayesian approach to this problem does not require probabilistic representation. (shrink)

Explanations are very important to us in many contexts: in science, mathematics, philosophy, and also in everyday and juridical contexts. But what is an explanation? In the philosophical study of explanation, there is long-standing, influential tradition that links explanation intimately to causation: we often explain by providing accurate information about the causes of the phenomenon to be explained. Such causal accounts have been the received view of the nature of explanation, particularly in philosophy of science, since the 1980s. However, philosophers (...) have recently begun to break with this causal tradition by shifting their focus to kinds of explanation that do not turn on causal information. The increasing recognition of the importance of such non-causal explanations in the sciences and elsewhere raises pressing questions for philosophers of explanation. What is the nature of non-causal explanations – and which theory best captures it? How do non-causal explanations relate to causal ones? How are non-causal explanations in the sciences related to those in mathematics and metaphysics? This volume of new essays explores answers to these and other questions at the heart of contemporary philosophy of explanation. The essays address these questions from a variety of perspectives, including general accounts of non-causal and causal explanations, as well as a wide range of detailed case studies of non-causal explanations from the sciences, mathematics and metaphysics. (shrink)

While Bayesian models have been applied to an impressive range of cognitive phenomena, methodological challenges have been leveled concerning their role in the program of rational analysis. The focus of the current article is on computational impediments to probabilistic inference and related puzzles about empirical confirmation of these models. The proposal is to rethink the role of Bayesian methods in rational analysis, to adopt an independently motivated notion of rationality appropriate for computationally bounded agents, and to explore broad conditions under (...) which Bayesian agents would be rational. The proposal is illustrated with a characterization of costs inspired by thermodynamics. (shrink)

Rational analysis is an influential but contested account of how probabilistic modeling can be used to construct non-mechanistic but self-standing explanatory models of the mind. In this paper, I disentangle and assess several possible explanatory contributions which could be attributed to rational analysis. Although existing models suffer from evidential problems that question their explanatory power, I argue that rational analysis modeling can complement mechanistic theorizing by providing models of environmental affordances.

Not all scientific explanations work by describing causal connections between events or the world's overall causal structure. In addition, mathematicians regard some proofs as explaining why the theorems being proved do in fact hold. This book proposes new philosophical accounts of many kinds of non-causal explanations in science and mathematics.

This article reviews the recent literature on idealization, specifically idealization in the course of scientific modeling. We argue that idealization is not a unified concept and that there are three different types of idealization: Galilean, minimalist, and multiple models, each with its own justification. We explore the extent to which idealization is a permanent feature of scientific representation and discuss its implications for debates about scientific realism.

According to a growing trend in theoretical neuroscience, the human perceptual system is akin to a Bayesian machine. The aim of this article is to clearly articulate the claims that perception can be considered Bayesian inference and that the brain can be considered a Bayesian machine, some of the epistemological challenges to these claims; and some of the implications of these claims. We address two questions: (i) How are Bayesian models used in theoretical neuroscience? (ii) From the use of Bayesian (...) models in theoretical neuroscience, have we learned or can we hope to learn that perception is Bayesian inference or that the brain is a Bayesian machine? From actual practice in theoretical neuroscience, we argue for three claims. First, currently Bayesian models do not provide mechanistic explanations; instead they are useful devices for predicting and systematizing observational statements about people's performances in a variety of perceptual tasks. That is, currently we should have an instrumentalist attitude towards Bayesian models in neuroscience. Second, the inference typically drawn from Bayesian behavioural performance in a variety of perceptual tasks to underlying Bayesian mechanisms should be understood within the three-level framework laid out by David Marr ( [1982] ). Third, we can hope to learn that perception is Bayesian inference or that the brain is a Bayesian machine to the extent that Bayesian models will prove successful in yielding secure and informative predictions of both subjects' perceptual performance and features of the underlying neural mechanisms. (shrink)

Are people rational? This question was central to Greek thought and has been at the heart of psychology and philosophy for millennia. This book provides a radical and controversial reappraisal of conventional wisdom in the psychology of reasoning, proposing that the Western conception of the mind as a logical system is flawed at the very outset. It argues that cognition should be understood in terms of probability theory, the calculus of uncertain reasoning, rather than in terms of logic, the calculus (...) of certain reasoning. (shrink)

When we think of scientific realism, there seem to be to ways to conceive of what it is about. The first is to see it as a view about scientific theories; the second is to see it as a view about the world. Some philosophers, most typically from Australia, think that the second way is the correct way. Scientific realism, they argue, is a metaphysical thesis: it asserts the reality of some types of entity, most typically, unobservable entities. I agree (...) that scientific realism has a metaphysical dimension, but I have insisted that it has other dimensions too. In my (1999), I took scientific realism to consist of three theses (or stances). (shrink)

underpinning of the cognitive sciences. I argue, however, that it often fails to provide adequate explanations, in particular in conjunction with competence theories. This failure originates in the idealizations in competence descriptions, which either ?block? the cascade, or produce a successful cascade which fails to explain cognition.

Debates about scientific realism are closely connected to almost everything else in the philosophy of science, for they concern the very nature of scientific knowledge. Scientific realism is a positive epistemic attitude toward the content of our best theories and models, recommending belief in both observable and unobservable aspects of the world described by the sciences. This epistemic attitude has important metaphysical and semantic dimensions, and these various commitments are contested by a number of rival epistemologies of science, known collectively (...) as forms of scientific antirealism. This article explains what scientific realism is, outlines its main variants, considers the most common arguments for and against the position, and contrasts it with its most important antirealist counterparts. (shrink)

A natural way to think of models is as abstract entities. If theories employ models to represent the world, theories traffic in abstract entities much more widely than is often assumed. This kind of thought seems to create a problem for a scientific realist approach to theories. Scientific realists claim theories should be understood literally. Do they then imply the reality of abstract entities? Or are theories simply—and incurably—false? Or has the very idea of literal understanding to be abandoned? Is (...) then fictionalism towards scientific theories inevitable? This paper argues that scientific realism can happily co-exist with models qua abstracta. (shrink)

Bayes' Theorem is a simple mathematical formula used for calculating conditional probabilities. It figures prominently in subjectivist or Bayesian approaches to epistemology, statistics, and inductive logic. Subjectivists, who maintain that rational belief is governed by the laws of probability, lean heavily on conditional probabilities in their theories of evidence and their models of empirical learning. Bayes' Theorem is central to these enterprises both because it simplifies the calculation of conditional probabilities and because it clarifies significant features of subjectivist position. Indeed, (...) the Theorem's central insight — that a hypothesis is confirmed by any body of data that its truth renders probable — is the cornerstone of all subjectivist methodology. (shrink)

Many phenomena in the natural world are complex, so scientists study them through simplified and idealised models. Philosophers of science have sought to explain how these models relate to the world. On most accounts, models do not represent the world directly, but through target systems. However, our knowledge of target systems is incomplete. First, what is the process by which target systems come about? Second, what types of entity are they? I argue that the basic conception of target systems, on (...) which other conceptions depend, is as parts of the world. I outline the process of target system specification and show that it is a crucial step in modelling. I also develop an account of target system evaluation, based on aptness. Paying close attention to target system specification and evaluation can help scientists minimise the frequency and extent of mistakes, when they are using models to investigate phenomena in complex real-world systems. (shrink)

This article attempts to clarify the commitments of a predictive coding approach to perception. After summarizing predictive coding theory, the article addresses two questions. Is a predictive coding perceptual system also a Bayesian system? Is it a Kantian system? The article shows that the answer to these questions is negative.

Scientific discourse is rife with passages that appear to be ordinary descriptions of systems of interest in a particular discipline. Equally, the pages of textbooks and journals are filled with discussions of the properties and the behavior of those systems. Students of mechanics investigate at length the dynamical properties of a system consisting of two or three spinning spheres with homogenous mass distributions gravitationally interacting only with each other. Population biologists study the evolution of one species procreating at a constant (...) rate in an isolated ecosystem. And when studying the exchange of goods, economists consider a situation in which there are only two goods, two perfectly rational agents, no restrictions on available information, no transaction costs, no money, and dealings are done immediately. Their surface structure notwithstanding, no competent scientist would mistake descriptions of such systems as descriptions of an actual system: we know very well that there are no such systems. These descriptions are descriptions of a model-system, and scientists use model-systems to represent parts or aspects of the world they are interested in. Following common practice, I refer to those parts or aspects as target-systems. What are we to make of this? Is discourse about such models merely a picturesque and ultimately dispensable façon de parler? This was the view of some early twentieth century philosophers. Duhem (1906) famously guarded against confusing model building with scientific theorizing and argued that model building has no real place in science, beyond a minor heuristic role. The aim of science was, instead, to construct theories, with theories understood as classificatory or representative structures systematically presented and formulated in precise symbolic.. (shrink)

Are all three of Marr's levels needed? Should they be kept distinct? We argue for the distinct contributions and methodologies of each level of analysis. It is important to maintain them because they provide three different perspectives required to understand mechanisms, especially information-processing mechanisms. The computational perspective provides an understanding of how a mechanism functions in broader environments that determines the computations it needs to perform. The representation and algorithmic perspective offers an understanding of how information about the environment is (...) encoded within the mechanism and what are the patterns of organization that enable the parts of the mechanism to produce the phenomenon. The implementation perspective yields an understanding of the neural details of the mechanism and how they constrain function and algorithms. Once we adequately characterize the distinct role of each level of analysis, it is fairly straightforward to see how they relate. (shrink)

The prominence of Bayesian modeling of cognition has increased recently largely because of mathematical advances in specifying and deriving predictions from complex probabilistic models. Much of this research aims to demonstrate that cognitive behavior can be explained from rational principles alone, without recourse to psychological or neurological processes and representations. We note commonalities between this rational approach and other movements in psychology – namely, Behaviorism and evolutionary psychology – that set aside mechanistic explanations or make use of optimality assumptions. Through (...) these comparisons, we identify a number of challenges that limit the rational program's potential contribution to psychological theory. Specifically, rational Bayesian models are significantly unconstrained, both because they are uninformed by a wide range of process-level data and because their assumptions about the environment are generally not grounded in empirical measurement. The psychological implications of most Bayesian models are also unclear. Bayesian inference itself is conceptually trivial, but strong assumptions are often embedded in the hypothesis sets and the approximation algorithms used to derive model predictions, without a clear delineation between psychological commitments and implementational details. Comparing multiple Bayesian models of the same task is rare, as is the realization that many Bayesian models recapitulate existing (mechanistic level) theories. Despite the expressive power of current Bayesian models, we argue they must be developed in conjunction with mechanistic considerations to offer substantive explanations of cognition. We lay out several means for such an integration, which take into account the representations on which Bayesian inference operates, as well as the algorithms and heuristics that carry it out. We argue this unification will better facilitate lasting contributions to psychological theory, avoiding the pitfalls that have plagued previous theoretical movements. (shrink)

What counts as a computation and how it relates to cognitive function are important questions for scientists interested in understanding how the mind thinks. This paper argues that pragmatic aspects of explanation ultimately determine how we answer those questions by examining what is needed to make rigorous the notion of computation used in the (cognitive) sciences. It (1) outlines the connection between the Church-Turing Thesis and computational theories of physical systems, (2) differentiates merely satisfying a computational function from true computation, (...) and finally (3) relates how we determine a true computation to the functional methodology in cognitive science. All of the discussion will be directed toward showing that the only way to connect formal notions of computation to empirical theory will be in virtue of the pragmatic aspects of explanation. (shrink)

Sensorimotor psychology studies the mental processes that control goal-directed bodily motion. Recently, sensorimotor psychologists have provided empirically successful Bayesian models of motor control. These models describe how the motor system uses sensory input to select motor commands that promote goals set by high-level cognition. I highlight the impressive explanatory benefits offered by Bayesian models of motor control. I argue that our current best models assign explanatory centrality to a robust notion of mental representation. I deploy my analysis to defend intentional (...) realism, to rebut eliminativism and instrumentalism regarding mental representation, and to explore the relation between intentionality and normativity. (shrink)

This paper explores a novel form of Mental Fictionalism: Fictionalism about talk of neural representations in cognitive science. This type of Fictionalism promises to (i) avoid the hard problem of naturalising representations, without (ii) incurring the high costs of eliminating useful representation talk. In this paper, I motivate and articulate this form of Fictionalism, and show that, despite its apparent advantages, it faces two serious objections. These objections are: (1) Fictionalism about talk of neural representations ultimately does not avoid the (...) problem of naturalising representations; (2) Fictional representations cannot play the explanatory role required by cognitive science. (shrink)

Recently philosophers of science have begun to pay more attention to the use of highly idealized mathematical models in scientific theorizing. An important example of this kind of highly idealized modeling is the widespread use of optimality models within evolutionary biology. One way to understand the explanations provided by these models is as a censored causal explanation: an explanation that omits certain causal factors in order to focus on a modular subset of the causal processes that led to the explanandum. (...) In this paper, I first argue that the censored causal model approach fails to establish a permanent explanatory role for optimality models in biology and mischaracterizes the explanatory virtues of biological optimality modeling. In addition, I argue that many biological optimality explanations cannot be characterized as censored causal explanations. In response, I propose an alternative approach that analyzes optimality models’ reliance on synchronically representing a system’s constraints and tradeoffs as well as their employment of various kinds of idealization in order to provide equilibrium explanations. (shrink)

Not all models are explanatory. Some models are data summaries. Some models sketch explanations but leave crucial details unspecified or hidden behind filler terms. Some models are used to conjecture a how-possibly explanation without regard to whether it is a how-actually explanation. I use the Hodgkin and Huxley model of the action potential to illustrate these ways that models can be useful without explaining. I then use the subsequent development of the explanation of the action potential to show what is (...) required of an adequate mechanistic model. Mechanistic models are explanatory. (shrink)

This paper takes up a recent challenge to mechanistic approaches to computational implementation, the view that computational implementation is best explicated within a mechanistic framework. The challenge, what has been labelled “the abstraction problem”, claims that one of MAC’s central pillars – medium independence – is deeply confused when applied to the question of computational implementation. The concern is that while it makes sense to say that computational processes are abstract (i.e. medium-independent), it makes considerably less sense to say that (...) they are also concrete processes of a mechanism. After outlining the problem and its effect on MAC, I examine a recent response from Kuokkanen and Rusanen [2018. “Making Too Many Enemies: Hutto and Myin’s Attack on Computationalism.” Philosophical Explorations 21 (2): 282–294. doi:10.1080/13869795.2018.1477980]. I argue that Kuokkanen and Rusanen’s response comes up short insofar as it makes problematic trade-offs among various desiderata we have for a theory of implementation. This leads to a general dilemma for MAC: either give up being an objective theory of implementation or concede the abstraction problem and so reintroduce triviality concerns. In response, I argue that conceiving of computations as abstracta rather than illata provides a way to avoid the proposed dilemma and articulate a notion of medium independence that addresses the abstraction problem. (shrink)

There is a certain excitement in vision science concerning the idea of applying the tools of bayesian decision theory to explain our perceptual capacities. Bayesian models are thought to be needed to explain how the inverse problem of perception is solved, and to rescue a certain constructivist and Kantian way of understanding the perceptual process. Anticlimactically, I argue both that bayesian outlooks do not constitute good solutions to the inverse problem, and that they are not constructivist in nature. In explaining (...) how visual systems derive a single percept from underdetermined stimulation, orthodox versions of bayesian accounts encounter a problem. The problem shows that such accounts need to be grounded in Natural Scene Statistics (NSS), an approach that takes seriously the Gibsonian insight that studying perception involves studying the statistical regularities of the environment in which we are situated. Additionally, I argue that bayesian frameworks postulate structures that hardly rescue a constructivist way of understanding perception. Except for percepts, the posits of bayesian theory are not representational in nature. bayesian perceptual inferences are not genuine inferences. They are biased processes that operate over nonrepresentational states. (shrink)

Can the output of human cognition be predicted from the assumption that it is an optimal response to the information-processing demands of the environment? A methodology called rational analysis is described for deriving predictions about cognitive phenomena using optimization assumptions. The predictions flow from the statistical structure of the environment and not the assumed structure of the mind. Bayesian inference is used, assuming that people start with a weak prior model of the world which they integrate with experience to develop (...) stronger models of specific aspects of the world. Cognitive performance maximizes the difference between the expected gain and cost of mental effort. Memory performance can be predicted on the assumption that retrieval seeks a maximal trade-off between the probability of finding the relevant memories and the effort required to do so; in categorization performance there is a similar trade-off between accuracy in predicting object features and the cost of hypothesis formation; in casual inference the trade-off is between accuracy in predicting future events and the cost of hypothesis formation; and in problem solving it is between the probability of achieving goals and the cost of both external and mental problem-solving search. The implemention of these rational prescriptions in neurally plausible architecture is also discussed. (shrink)

This article focuses on a case that expert practitioners count as an explanation: a mathematical account of Plateau’s laws for soap films. I argue that this example falls into a class of explanations that I call abstract explanations.explanations involve an appeal to a more abstract entity than the state of affairs being explained. I show that the abstract entity need not be causally relevant to the explanandum for its features to be explanatorily relevant. However, it remains unclear how to unify (...) abstract and causal explanations as instances of a single sort of thing. I conclude by examining the implications of the claim that explanations require objective dependence relations. If this claim is accepted, then there are several kinds of objective dependence relations. 1 Introduction2 A Case3 Abstract and Causal Explanations4 Recent Work on Mathematical Explanation5 Explanation and Dependence6 Conclusion. (shrink)