Tobias Schlicht

Mit dem Niedergang des Behaviorismus in der Psychologie startete der Siegeszug der Kognitionswissenschaft als in‐ terdisziplinärer, vor allem empirischer Untersuchung mentaler Phänomene. Ein Kernbegriff in dieser multidisziplinären Forschung ist der einer mentalen Repräsentation, mit dem alle mentalen Phänomene erklärt werden sollen. Kognitive Prozesse sollen als Informationsverarbeitungen in dem Sinne verstanden werden, dass es sich um Berechnungen bzw. computationale Vorgänge von mentalen Repräsentationen im Gehirn handeln soll. Herausforderungen sind Fragen wie die folgenden: Wie kann dieses Explanans einer Repräsentation als Postulat gerechtfertigt und in ein naturalistisches Weltbild integriert werden? Wie kann der Inhalt solcher Repräsentationen zur Erklärung von Verhaltensweisen bestimmt werden? Eine weitere Herausforderung bieten alternative Erklärungsansätze kognitiver Phänomene; denn in den letzten Jahrzehnten entwickelte sich eine Opposition gegen mentale Repräsentationen unter dem Sammelnamen „Enaktivismus“ bzw. „Situierter Kognition“, die die Verkörperung und situative Einbettung kognitiver Phänomene und Akte in die Umwelt hervorheben. Dieser Artikel liefert eine knappe Einführung in zeitgenössische Debatten zur Natur mentaler Repräsentationen.

Alvin Goldman considers simulation-theory as a successor of Theodor Lipps’ theory of empathy but does not consider the alternative phenomenological conceptions of empathy proposed by Max Scheler, Edmund Husserl and Edith Stein in critical evaluation of Lipps’ account. In this chapter, we contrast these accounts of empathy and evaluate them briefly with respect to the contemporary debate in philosophy and cognitive science. We can only sketch these complex debates here but emphasize that they introduce a new dimension of social cognition next to mindreading which involves consciousness as an important ingredient.

We started our discussion with several considerations about the wide range of phenomena that go under the heading of social cognition. We have looked at approaches to how mindreading works, the attribution of mental states for the purpose of explaining and predicting behaviour of others. In this regard, Chaps. 2, 3 and 4 elaborated and critically discussed the theory-theory, modularity-theory and simulation-theory as different philosophical approaches aiming to capture how such attribution works. Discussion of theories of empathy and enactive approaches brought further facets and situational differences to light. Consideration of the ambitious predictive processing framework put forward in cognitive science and philosophy completed the picture. In this concluding chapter, these discussions are briefly summarized, before some new developments and future challenges are outlined. These challenges, to do with the development of new technologies that engender new forms of social interaction, involve interactions via digital media and with humanoid robots and other AI systems.

For a long time, the theory-theory, modularity-theory and simulation-theory have dominated the debate on social cognition.. In this chapter, we look at new conceptions of cognition and alternative suggestions about how we understand other people’s mindsunder the heading of “enactivism”. Shaun Gallagher (2005), Thomas Fuchs (2018), Hanne de Jaegher and Ezequiel Di Paolo (2007) and others appeal to traditional phenomenological accounts of empathy to criticize central theoretical assumptions of representational accounts and argue for the primacy of interaction to observation and of perception to more sophisticated cognitive strategies to understand others. We scrutinize their arguments for the constitutive role of interaction for understanding and for the directness of perception and evaluate their claim that an inactive account can replace existing theory- and simulation-theories.

One day, the young wizard Harry Potter is required to see his teacher, Professor Severus Snape, who announces that he is going to teach him in the magical art of occlumency, the capacity to shield one’s own mind from the magical access and manipulation from the outside. Harry’s devious opponent in J.K. Rowling’s bestselling novels, Lord Voldemort, is an excellent master in legilimency, the opposite “ability to extract feelings and memories from another person’s mind”. – “He can read minds?”, Harry enquires quickly, simply to be corrected by Snape. “You do not understand fine distinctions… Only muggles talk of ‘mind-reading’. The mind is not a book, to be opened at will and examined at leisure. Thoughts are not etched on the inside of skulls, to be perused by any invader. The mind is a complex and many-layered thing, Potter – or at least, most minds are.” Professor Snape smirks and continues: “It is true, however, that those who have mastered legilimency are able, under certain conditions, to delve into the minds of their victims and to interpret their findings correctly” (Rowling, 2003, p. 468–9).

In this chapter, the nativist alternative to theory-theory is introduced and discussed. So-called modularity-theories appeal to innate domain-specific mechanisms, sometimes called modules, to explain cognitive development in certain specific domains. These modules contrast with domain-general learning mechanisms also endorsed by empiricist theory-theorists. We discuss this group of theories by introducing two representative accounts. First, Susan Carey and Elisabeth Spelke distinguish several basic mechanisms of core knowledge which process information about physical objects, numbers, agents and other domains. Second, Simon Baron-Cohen uses empirical evidence to support his distinction of several specific mechanisms supporting social cognition, each of them producing and processing representations of diverging complexity, for example, information about eye gaze, or about triadic relations between two agents and an object, culminating in a theory-of-mind module specialized for mindreading. Such nativists sometimes appeal to empirical evidence about early implicit false-belief understanding and about selected impairments in social understanding in autistic subjects.

Proponents of theory-theory support their hypothesis of a radical theory change in the cognitive development of young children by appealing to empirical data allegedly demonstrating that children start to understand that other people can have false beliefs at around the age of four years. In the chapter, we present this ground-breaking set of empirical data, its various interpretations and the almost 40-year history of the various forms of false-belief tasks which raise interesting questions about implicit and explicit understanding and yield further theoretical proposals to capture the experimental data.

Towards the end of the 1980s, theory-theory and modularity-theory were accompanied by a rival approach, the simulation-theory, put forward by philosophers Robert Gordon (1986), Jane Heal (1986) and Alvin Goldman (1989) and by developmental psychologist Paul Harris (1992). They reject the idea of a rich body of information needed to understand others. In this chapter, we discuss Alvin Goldman's detailed proposal, according to which we use our own subjective experience as a model to understand what others think and feel. We analyze the notion of mental simulation in some detail and evaluate its application to two distinct levels of mindreading, both of which employ the imagination as a core capacity. Finally, we evaluate theories of embodied simulation which appeal to the mirror neuron system on the implementation level in support of a version of simulation-theory and discuss the contribution of mirror neurons to social understanding more generally.

So far, we have introduced and discussed three prominent approaches to mindreading – theory-theory, modularity-theory, and simulation-theory – put forward as competing descriptions of the representational and computational processes on what Marr (1982) called the algorithmic level of description. As we saw in Chaps. 5 and 6, theories of empathy and enactive accounts of direct perceptual experience, though invariably put forward as competitors to these views, really describe a different phenomenon under the heading of social cognition, namely, empathy. ‑ In this chapter, we introduce the recently popular predictive processing framework of perception, cognition, and action and evaluate its explanatory strength in the context of social cognition. It has been praised by philosophers and cognitive scientists alike for its unificatory power as a theory of cognition in general, and recently some researchers have also claimed that it can provide a unifying view of social understanding. Since the literature on this topic is still very sparse, we will keep the discussion brief and relate the theory to the theories we have discussed so far.

The popular predictive processing (PP) framework posits prediction error minimization (PEM) as the sole mechanism in the brain that can account for all mental phenomena, including consciousness. I first highlight three ambitions associated with major presentations of PP: (1) Completeness (PP aims for a comprehensive account of mental phenomena), (2) Bayesian realism (PP claims that PEM is implemented in the brain rather than providing only a model), and (3) Naturalism (PP is typically presented as yielding a naturalistic view of the mind). Then I demonstrate that many proponents of PP also endorse a form of Kantian transcendental idealism (TI), based on a characterization of experiential content as the brain's currently best hypothesis about the world. I argue that endorsing this claim (4), that is, that we only experience the world as it appears, but not the world itself, sabotages achieving the three ambitions. The argument proceeds by discussing the prospects of each ambition in turn, drawing on discussions in the philosophy of science about realism and its alternatives, about the motivation and features of computational models, and about the foundational role of consciousness for science.

The scientific study of consciousness with its cornerstone of searching the neural correlates of consciousness constitutes an exciting and lively area of research. Yet, the empirical study of consciousness is surrounded by a host of philosophical challenges some of which are not new, while others have arisen and been elaborated over the last decades. The terminology can sometimes be confusing, causing misunderstandings how the challenges relate to each other. This paper provides an overview of several much-discussed philosophical problems of consciousness, offering a way of arranging them and some thoughts about how they relate. Most labels are taken from the literature so that there is a progression from the “easy” via the “hard” to the “harder” problem(s), and some labels are made up, like the “hardest” problem. For reasons of space, there is not detailed discussion of the possible solutions to these problems even though some problems have been formulated in response to others. For example, the “real” problem approach and the “illusion” or “meta” problems can be read partly as responses to the “hard” problem.

Predictive processing is often presented as a unifying framework for perception, action, and cognition, being able to explain most mental phenomena : with regard to perception, the brain harbours a generative model issuing top-down expectations that are matched against bottom-up sensory feedback. Mismatches lead to error messages and model updates until the brain is 'getting it right'. The core notion of prediction error minimization commits the framework to a specification of accuracy conditions. We therefore turn to issues related to the determination of satisfaction conditions discussed in the philosophy of perception. In particular, we rely on work by Recanati who shows that the accuracy conditions of perceptual content are partly determined by the intentional mode and the perceiver. It is important to notice that the self can enter the specification of accuracy conditions in two ways, namely as subject or as object. Even if we do not perceive ourselves explicitly as an object, we always implicitly represent ourselves as subject. We discuss whether and how these two ways of self-representation can be respected in the predictive processing framework.

The papers in this special issue make important contributions to a longstanding debate about how we should conceive of and explain mental phenomena. In other words, they make a case about the best philosophical paradigm for cognitive science. The two main competing approaches, hotly debated for several decades, are representationalism and enactivism. However, recent developments in disciplines such as machine learning and computational neuroscience have fostered a proliferation of intermediate approaches, leading to the emergence of completely new positions, in particular the Predictive Processing approach. Here, we will consider the different approaches discussed in this volume.

Mental representation is one of core theoretical constructs within cognitive science and, together with the introduction of the computer as a model for the mind, is responsible for enabling the ‘cognitive turn’ in psychology and associated fields. Conceiving of cognitive processes, such as perception, motor control, and reasoning, as processes that consist in the manipulation of contentful vehicles representing the world has allowed us to refine our explanations of behavior and has led to tremendous empirical advancements. Despite the central role that the concept plays in cognitive science, there is no unanimously accepted characterization of mental representation. Technological and methodological progress in the cognitive sciences has produced numerous computational models of the brain and mind, many of which have introduced mutually incompatible notions of mental representation. This proliferation has led some philosophers to question the metaphysical status and explanatory usefulness of the notion. This book contains state-of-the-art chapters on the topic of mental representation, assembling some of the leading experts in the field and allowing them to engage in meaningful exchanges over some of the most contentious questions. The collection gathers both proponents and critics of the concept of mental representation, allowing them to engage with topics such as the ontological status of representations, the possibility of formulating a general account of mental representation which would fit our best explanatory practices, and the possibility of delivering such an account in fully naturalistic terms.

We discuss whether affordances can be reasons, against the background of two interlocked considerations: (1) While the problematic degree of idealization in accounts of reasons that treat them as mental states speaks in favor of the alternative view which treats them as facts, a cognitive consideration relationship is still required to account for the motivating role of reasons. (2) While recent enactive accounts of cognition hold promise to avoid over-intellectualization of acting for reasons, these are so far either underdeveloped or treat reasons as mental states after all. Considering affordances as reasons promises to strengthen the enactivist project. We first motivate factualism about reasons, then introduce enactivism and finally discuss whether affordances can play the three roles of explanatory, justifying and motivating reasons. Since we do not take this discussion to be exhaustive but rather as outlining a research program, we point to desiderata for further work.

In the debate about the nature of social cognition we see a shift towards theories that explain social understanding through interaction. This paper discusses autopoietic enactivism and the we-mode approach in the light of such developments. We argue that a problem seems to arise for these theories: an interactionist account of social cognition makes the capacity of shared intentionality a presupposition of social understanding, while the capacity of engaging in scenes of shared intentionality in turn presupposes exactly the kind of social understanding that it is intended to explain. The social capacity in question that is presupposed by these accounts is then analyzed in the second section via a discussion and further development of Searle’s ‘sense of us’ and ‘sense of the other’ as a precondition for social cognition and joint action. After a critical discussion of Schmid’s recent proposal to analyze this in terms of plural pre-reflective selfawareness, we develop an alternative account. Starting from the idea that infants distinguish in perception between physical objects and other agents we distinguish between affordances and social affordances and cash out the notion of a social affordance in terms of “interaction-oriented representations”, parallel to the analysis of object affordances in terms of “action-oriented representations”. By characterizing their respective features we demonstrate how this approach can solve the problem formulated in the first part.

In this response we address additions to as well as criticisms and possible misinterpretations of our proposal for a second-person neuroscience. We map out the most crucial aspects of our approach by (1) acknowledging that second-person engaged interaction is not the only way to understand others, although we claim that it is ontogenetically prior; (2) claiming that spectatorial paradigms need to be complemented in order to enable a full understanding of social interactions; and (3) restating that our theoretical proposal not only questions the mechanism by which a cognitive process comes into being, but asks whether it is at all meaningful to speak of a mechanism and a cognitive process when it is confined to intra-agent space. We address theoretical criticisms of our approach by pointing out that while a second-person social understanding may not be the only mechanism, alternative approaches cannot hold their ground without resorting to second-person concepts, if not in the expression, certainly in the development of social understanding. In this context, we also address issues of agency and intentionality, theoretical alternatives, and clinical implications of our approach

This introductory textbook provides a comprehensive and up-to-date overview of the main issues in contemporary philosophy of social cognition. It explains and critically discusses each of the key philosophical answers to the captivating question of how we understand the mental life of other sentient creatures. Key Features: · Clearly and fully describes the major theoretical approaches to the understanding of other people’s minds. · Suggests the major advantages and limitations of each approach, indicating how they differ as well as the ideas they have in common. · Tests each philosophical theory against the best available empirical data from psychology, neuroscience and psychopathology. · Includes suggestions for additional reading and practice study questions at the end of each chapter. Philosophy of Social Cognition is essential reading for all undergraduate and graduate students taking introductory courses on social cognition. It is also ideal for courses on cognitive neuroscience, social psychology and sociological theory.

This paper reviews several ways in which Kant’s approach to cognition has been influential and relevant for the development of various paradigms in cognitive science, such as functionalism, enactivism, and the predictive processing model of the mind. In the second part, it discusses philosophical issues arising from recent developments in artificial intelligence in relation to Kant’s conception of cognition and understanding. More precisely, it investigates questions about perception, cognition, learning, understanding, and about the age-old debate between empiricists and rationalists in the context of so-called deep neural network architectures as well as the relevance of Kant’s conception of cognition and understanding for these issues.

In spite of the remarkable progress made in the burgeoning field of social neuroscience, the neural mechanisms that underlie social encounters are only beginning to be studied and could —paradoxically— be seen as representing the ‘dark matter’ of social neuroscience. Recent conceptual and empirical developments consistently indicate the need for investigations, which allow the study of real-time social encounters in a truly interactive manner. This suggestion is based on the premise that social cognition is fundamentally different when we are in interaction with others rather than merely observing them. In this article, we outline the theoretical conception of a second-person approach to other minds and review evidence from neuroimaging, psychophysiological studies and related fields to argue for the development of a second-person neuroscience, which will help neuroscience to really go social; this may also be relevant for our understanding of psychiatric disorders construed as disorders of social cognition.

This paper presents arguments from empirical research and from philosophical considerations to the effect that phenomenality and content are two distinct and independent features of mental representations, which are both relational. Thus, it is argued, classical arguments that infer phenomenality from content have to be rejected. Likewise, theories that try to explain the phenomenal character of experiences by appeal to specific types of content cannot succeed. Instead, a dynamic view of consciousness has to be adopted that seeks to explain consciousness by certain ways of processing representations. Therefore, only empirical methods that are able to investigate the dynamics of the mind can be used for the “quest for consciousness” proper. Moreover, the central intuitions about consciousness are best explained when phenomenality and content are clearly distinguished.

In this paper, we put forward an interdisciplinary framework describing different levels of self-representations, namely non-conceptual, conceptual and propositional self-representations. We argue that these different levels of self-representation are differently affected by cultural upbringing: while propositional self-representations rely on “theoretical” concepts and are thus strongly influenced by cultural upbringing, non-conceptual self-representations are uniform across cultures and thus universal. This differentiation offers a theoretical specification of the distinction between an independent and interdependent self-construal put forward in cross-cultural psychology. Hence, this does not only allow for a deeper understanding of different self-conceptions, but also for a formulation of new hypotheses regarding the cultural influence on self-representations. As one example, we will highlight the role of the proposed levels of self-representation for emotional experience and formulate some major implications of our interdisciplinary framework for future empirical research

Dieser Beitrag nimmt als Ausganspunkt die gegenwärtige kontroverse Debatte um den adäquaten Erklärungsansatz in der Kognitionswissenschaft, und zwar zwischen Enaktivisten einerseits und Repräsentationalisten andererseits. Beispielhaft wurde aufgezeigt, dass in Bezug auf die soziale Kognition der enaktivistische Ansatz eine Voraussetzung machen muss, die er mit seinen eigenen Mitteln nicht einzufangen in der Lage ist. Denn der Versuch, das Verstehen des Anderen durch gemeinsame Sinnstiftung der Akteure innerhalb ihrer sozialen Interaktion zu erklären, gerät in einen Zirkel, da nicht erklärt wird, wie die Kopplung und damit der Prozess der sozialen Interaktion allererst zustande kommen kann. Als Lösung dieses Problems wird eine repräsentationalistische Analyse der vorausgesetzten Fähigkeit, andere als potentielle Kooperationspartner anzusehen, entwickelt, und zwar mit den Mitteln von Millikans Teleosemantik. Diese Theorie aus dem Kontext der Debatte um die Naturalisierung mentaler Repräsentationen liefert mit den PP Repräsentationen den adäquaten Begriff, um sowohl verkörperte, handlungsorientierte Repräsentationen zur Analyse von Affordanzen, als auch verkörperte interaktionsorientierte Repräsentationen zur Analyse sozialer Affordanzen zu entwickeln. Dies erlaubt die Annahme einer zentralen Funktion für die soziale Wahrnehmung, die den blinden Fleck des enaktivistischen Ansatzes ausfüllen kann. Zugleich demonstriert diese Analyse der sozialen Wahrnehmung die explanatorische Überlegenheit (minimal) repräsentationalistischer Erklärungen gegenüber rein enaktivistischen Alternativen.