Marcin Miłkowski

In this paper, I argue that embodied cognition, like many other research traditions in cognitive science, offers mostly fallible research heuristics rather than grand principles true of all cognitive processing. To illustrate this claim, I discuss Aizawa’s rebuttal of embodied and enactive accounts of vision. While Aizawa’s argument is sound against a strong reading of the enactive account, it does not undermine the way embodied cognition proceeds, because the claim he attacks is one of fallible heuristics. These heuristics may be helpful in developing models of cognition in an interdisciplinary fashion. I briefly discuss the issue of whether this fallibility actually makes embodied cognition vulnerable to charges of being untestable or non-scientific. I also stress that the historical approach to this research tradition suggests that embodied cognition is not poised to become a grand unified theory of cognition.

In this paper, we focus on the development of geometric cognition. We argue that to understand how geometric cognition has been constituted, one must appreciate not only individual cognitive factors, such as phylogenetically ancient and ontogenetically early core cognitive systems, but also the social history of the spread and use of cognitive artifacts. In particular, we show that the development of Greek mathematics, enshrined in Euclid’s Elements, was driven by the use of two tightly intertwined cognitive artifacts: the use of lettered diagrams; and the creation of linguistic formulae. Together, these artifacts formed the professional language of geometry. In this respect, the case of Greek geometry clearly shows that explanations of geometric reasoning have to go beyond the confines of methodological individualism to account for how the distributed practice of artifact use has stabilized over time. This practice, as we suggest, has also contributed heavily to the understanding of what mathematical proof is; classically, it has been assumed that proofs are not merely deductively correct but also remain invariant over various individuals sharing the same cognitive practice. Cognitive artifacts in Greek geometry constrained the repertoire of admissible inferential operations, which made these proofs inter-subjectively testable and compelling. By focusing on the cognitive operations on artifacts, we also stress that mental mechanisms that contribute to these operations are still poorly understood, in contrast to those mechanisms which drive symbolic logical inference.

Is the mathematical function being computed by a given physical system determined by the system’s dynamics? This question is at the heart of the indeterminacy of computation phenomenon (Fresco et al. [unpublished]). A paradigmatic example is a conventional electrical AND-gate that is often said to compute conjunction, but it can just as well be used to compute disjunction. Despite the pervasiveness of this phenomenon in physical computational systems, it has been discussed in the philosophical literature only indirectly, mostly with reference to the debate over realism about physical computation and computationalism. A welcome exception is Dewhurst’s ([2018]) recent analysis of computational individuation under the mechanistic framework. He rejects the idea of appealing to semantic properties for determining the computational identity of a physical system. But Dewhurst seems to be too quick to pay the price of giving up the notion of computational equivalence. We aim to show that the mechanist need not pay this price. The mechanistic framework can, in principle, preserve the idea of computational equivalence even between two different enough kinds of physical systems, say, electrical and hydraulic ones.

The purpose of this paper is to argue against the claim that morphological computation is substantially different from other kinds of physical computation. I show that some (but not all) purported cases of morphological computation do not count as specifically computational, and that those that do are solely physical computational systems. These latter cases are not, however, specific enough: all computational systems, not only morphological ones, may (and sometimes should) be studied in various ways, including their energy efficiency, cost, reliability, and durability. Second, I critically analyze the notion of “offloading” computation to the morphology of an agent or robot, by showing that, literally, computation is sometimes not offloaded but simply avoided. Third, I point out that while the morphology of any agent is indicative of the environment that it is adapted to, or informative about that environment, it does not follow that every agent has access to its morphology as the model of its environment.

Replicability and reproducibility of computational models has been somewhat understudied by “the replication movement.” In this paper, we draw on methodological studies into the replicability of psychological experiments and on the mechanistic account of explanation to analyze the functions of model replications and model reproductions in computational neuroscience. We contend that model replicability, or independent researchers' ability to obtain the same output using original code and data, and model reproducibility, or independent researchers' ability to recreate a model without original code, serve different functions and fail for different reasons. This means that measures designed to improve model replicability may not enhance (and, in some cases, may actually damage) model reproducibility. We claim that although both are undesirable, low model reproducibility poses more of a threat to long-term scientific progress than low model replicability. In our opinion, low model reproducibility stems mostly from authors' omitting to provide crucial information in scientific papers and we stress that sharing all computer code and data is not a solution. Reports of computational studies should remain selective and include all and only relevant bits of code.

In this paper, we argue that several recent ‘wide’ perspectives on cognition (embodied, embedded, extended, enactive, and distributed) are only partially relevant to the study of cognition. While these wide accounts override traditional methodological individualism, the study of cognition has already progressed beyond these proposed perspectives towards building integrated explanations of the mechanisms involved, including not only internal submechanisms but also interactions with others, groups, cognitive artifacts, and their environment. The claim is substantiated with reference to recent developments in the study of “mindreading” and debates on emotions. We claim that the current practice in cognitive (neuro)science has undergone, in effect, a silent mechanistic revolution, and has turned from initial binary oppositions and abstract proposals towards the integration of wide perspectives with the rest of the cognitive (neuro)sciences.

In this paper, I argue that computationalism is a progressive research tradition. Its metaphysical assumptions are that nervous systems are computational, and that information processing is necessary for cognition to occur. First, the primary reasons why information processing should explain cognition are reviewed. Then I argue that early formulations of these reasons are outdated. However, by relying on the mechanistic account of physical computation, they can be recast in a compelling way. Next, I contrast two computational models of working memory to show how modeling has progressed over the years. The methodological assumptions of new modeling work are best understood in the mechanistic framework, which is evidenced by the way in which models are empirically validated. Moreover, the methodological and theoretical progress in computational neuroscience vindicates the new mechanistic approach to explanation, which, at the same time, justifies the best practices of computational modeling. Overall, computational modeling is deservedly successful in cognitive science. Its successes are related to deep conceptual connections between cognition and computation. Computationalism is not only here to stay, it becomes stronger every year.

In this chapter, I sketch the history of mechanistic models of the mental, as related to the technological project of trying to build mechanical minds, and discuss the contemporary debates on psychological and cognitive explanations. In the first section, I introduce the Cartesian notion of mechanism, which has shaped the debate in the centuries to follow. Early mechanistic proposals are also connected with early attempts to formulate the computational account of thinking and reasoning, upheld notably by Hobbes and Leibniz. In the second section, associationist and behaviorist models of the mind are sketched, along with attempts to understand the neural system in terms of connections and associations. Early robotic models, built mostly by behaviorists and other students of animal behavior, are also introduced. In the third section, the focus is on early computational and cybernetic models of the mind. In the fourth section, I deal with computational models of mental mechanisms as proposed by students of artificial intelligence and cognitive science. Then, I sketch the contemporary debate over the question whether computational and psychological models are best understood as offering functional, or maybe mechanistic explanations.

Naturalism is currently the most vibrantly developing approach to philosophy, with naturalised methodologies being applied across all the philosophical disciplines. One of the areas naturalism has been focussing upon is the mind, traditionally viewed as a topic hard to reconcile with the naturalistic worldview. A number of questions have been pursued in this context. What is the place of the mind in the world? How should we study the mind as a natural phenomenon? What is the significance of cognitive science research for philosophical debates? In this book, philosophical questions about the mind are asked in the context of recent developments in cognitive science, evolutionary theory, psychology, and the project of the naturalisation. Much of the focus is upon what we have learned by studying natural mental mechanisms as well as designing artificial ones. In the case of natural mental mechanisms, this includes consideration of such issues as the significance of deficits in these mechanisms for psychiatry. The significance of the evolutionary context for mental mechanisms as well as questions regarding rationality and wisdom is also explored. Mechanistic and functional models of the mind are used to throw new light on discussions regarding issues of explanation, reduction and the realisation of mental phenomena. Finally, naturalistic approaches are used to look anew at such traditional philosophical issues as the correspondence of mind to world and presuppositions of scientific research.

In this paper, an account of theoretical integration in cognitive (neuro)science from the mechanistic perspective is defended. It is argued that mechanistic patterns of integration can be better understood in terms of constraints on representations of mechanisms, not just on the space of possible mechanisms, as previous accounts of integration had it. This way, integration can be analyzed in more detail with the help of constraintsatisfaction account of coherence between scientific representations. In particular, the account has resources to talk of idealizations and research heuristics employed by researchers to combine separate results and theoretical frameworks. The account is subsequently applied to an example of successful integration in the research on hippocampus and memory, and to a failure of integration in the research on mirror neurons as purportedly explanatory of sexual orientation.

In this paper, I show how semantic factors constrain the understanding of the computational phenomena to be explained so that they help build better mechanistic models. In particular, understanding what cognitive systems may refer to is important in building better models of cognitive processes. For that purpose, a recent study of some phenomena in rats that are capable of ‘entertaining’ future paths (Pfeiffer and Foster 2013) is analyzed. The case shows that the mechanistic account of physical computation may be complemented with semantic considerations, and in many cases, it actually should.

Recent work on skin-brain thesis suggests the possibility of empirical evidence that empiricism is false. It implies that early animals need no traditional sensory receptors to be engaged in cognitive activity. The neural structure required to coordinate extensive sheets of contractile tissue for motility provides the starting point for a new multicellular organized form of sensing. Moving a body by muscle contraction provides the basis for a multicellular organization that is sensitive to external surface structure at the scale of the animal body. In other words, the nervous system first evolved for action, not for receiving sensory input. Thus, sensory input is not required for minimal cognition; only action is. The whole body of an organism, in particular its highly specific animal sensorimotor organization, reflects the bodily and environmental spatiotemporal structure. The skin-brain thesis suggests that, in contrast to empiricist claims that cognition is constituted by sensory systems, cognition may be also constituted by action-oriented feedback mechanisms. Instead of positing the reflex arc as the elementary building block of nervous systems, it proposes that endogenous motor activity is crucial for cognitive processes. In the paper, I discuss the issue whether the skin-brain thesis and its supporting evidence can be really used to overthrow the main tenet of empiricism empirically, by pointing out to cognizing agents that fail to have any sensory apparatus.

The Computational Theory of Mind (CTM) holds that the mind is a computer and that cognition is the manipulation of representations. CTM is commonly viewed as the main hypothesis in cognitive science, with classical CTM (related to the Language of Thought Hypothesis) being the most popular variant. However, other computational accounts of the mind either reject LOTH or do not subscribe to RTM. CTM proponents argue that it clarifies how thought and content are causally relevant in the physical world, and they frame this argument in terms of the physical symbol hypothesis, formalism, or syntactic engines. This article focuses on specific problems with CTM, with four main sections covering the introduction of the three main variants of CTM, the most important conceptions of computational explanation in cognitive science, skeptical arguments against CTM raised by Hilary Putnam, and common objections to CTM.

Is there a field of social intelligence? Many various disciplines ap-proach the subject and it may only seem natural to suppose that different fields of study aim at explaining different phenomena; in other words, there is no spe-cial field of study of social intelligence. In this paper, I argue for an opposite claim. Namely, there is a way to integrate research on social intelligence, as long as one accepts the mechanistic account to explanation. Mechanistic inte-gration of different explanations, however, comes at a cost: mechanism requires explanatory models to be fairly complete and realistic, and this does not seem to be the case for many models concerning social intelligence, especially models of economical behavior. Such models need either be made more realistic, or they would not count as contributing to the same field. I stress that the focus on integration does not lead to ruthless reductionism; on the contrary, mechanistic explanations are best understood as explanatorily pluralistic.

W artykule przedstawiono argumenty, że konfirmacja tezy, iż istnieją moduły umysłowe wyjaśniające cechy umysłu, jest z kilku powodów kłopotliwa. Po pierwsze, istnieje kilka konkurencyjnych teorii modularności, które zresztą nie zawsze się wykluczają, przez co nie można między nimi rozstrzygać eksperymentalnie. Po drugie, tezy na temat modularności często oparte są na bezzasadnym założeniu, iż wyróżnianie specyficznych dziedzin (semantycznych lub składniowych) działania modułów nie jest problematyczne. Po trzecie, analizując znany z literatury moduł wykrywania oszustów, postulowany przez Cosmides w celu wyjaśnienia rzekomej irracjonalności objawiającej się w tzw. zadaniu Wasona, pokazuję, że wyjaśniane zjawisko nie zostało zdefiniowane dostatecznie dokładnie, a przez to nieostra jest funkcjonalna charakterystyka modułów je wyjaśniających. Co więcej, nie ma powodów sądzić, że zjawisko, które ten moduł miał wyjaśniać, w ogóle istnieje. Wskazuję też kilka problemów metodologicznych związanych ze zbieraniem eksperymentalnych świadectw na rzecz modularności, takich jak zaburzanie wyników eksperymentów przez uśrednianie i brak kontroli nad kluczowymi czynnikami wpływającymi na rezultaty uzyskiwane przez uczestników badania.

In this article, after presenting the basic idea of causal accounts of implementation and the problems they are supposed to solve, I sketch the model of computation preferred by Chalmers and argue that it is too limited to do full justice to computational theories in cognitive science. I also argue that it does not suffice to replace Chalmers’ favorite model with a better abstract model of computation; it is necessary to acknowledge the causal structure of physical computers that is not accommodated by the models used in computability theory. Additionally, an alternative mechanistic proposal is outlined.

The paper defends the claim that the mechanistic explanation of information processing is the fundamental kind of explanation in cognitive science. These mechanisms are complex organized systems whose functioning depends on the orchestrated interaction of their component parts and processes. A constitutive explanation of every mechanism must include both appeal to its environment and to the role it plays in it. This role has been traditionally dubbed competence. To fully explain how this role is played it is necessary to explain the information processing inside the mechanism embedded in the environment. The most usual explanation on this level has a form of a computational model, for example a software program or a trained artificial neural network. However, this is not the end of the explanatory chain. What is left to be explained is how the program is realized (or what processes are responsible for information processing in the artificial neural network). By using two dramatically different examples from the history of cognitive science I show the multi-level structure of explanations in cognitive science. These examples are (1) the explanation of human process solving as proposed by A. Newell & H. Simon; (2) the explanation of cricket phonotaxis via robotic models by B. Webb.

In Darwin’s Dangerous Idea, Daniel Dennett claims that evolution is algorithmic. On Dennett’s analysis, evolutionary processes are trivially algorithmic because he assumes that all natural processes are algorithmic. I will argue that there are more robust ways to understand algorithmic processes that make the claim that evolution is algorithmic empirical and not conceptual. While laws of nature can be seen as compression algorithms of information about the world, it does not follow logically that they are implemented as algorithms by physical processes. For that to be true, the processes have to be part of computational systems. The basic difference between mere simulation and real computing is having proper causal structure. I will show what kind of requirements this poses for natural evolutionary processes if they are to be computational.

Artificial models of cognition serve different purposes, and their use determines the way they should be evaluated. There are also models that do not represent any particular biological agents, and there is controversy as to how they should be assessed. At the same time, modelers do evaluate such models as better or worse. There is also a widespread tendency to call for publicly available standards of replicability and benchmarking for such models. In this paper, I argue that proper evaluation ofmodels does not depend on whether they target real biological agents or not; instead, the standards of evaluation depend on the use of models rather than on the reality of their targets. I discuss how models are validated depending on their use and argue that all-encompassing benchmarks for models may be well beyond reach