A Deflationary Account of Mental Representation

Among the cognitive capacities of evolved creatures is the capacity to represent. Theories in cognitive neuroscience typically explain our manifest representational capacities by positing internal representations, but there is little agreement about how these representations function, especially with the relatively recent proliferation of connectionist, dynamical, embodied, enactive, and Bayesian approaches to cognition. This paper sketches an account of the nature and function of representation in cognitive neuroscience that couples a realist construal of representational vehicles with a pragmatic account of representational content. The resulting package is called a deflationary account of mental representation, and the chapter argues that it avoids the problems that afflict competing accounts.

Error Detection and Representational Mechanisms

This chapter defends a mechanistic and teleosemantic view of naturalized intentionality that underlies the role of error detection via coherence checking. Representational mechanisms serve the biological functions of representing, which are related to the semantic value of representation: its truth or falsity, its being vacuous or satisfied, or its accuracy. If representational mechanisms contain (or interact with) error-detection mechanisms, the semantic value of representation is causally relevant. As long as semantic value is causally relevant in cognitive explanations, the content of representation is arguably causally relevant, which vindicates the notion of mental representation in contemporary scientific research. Error detection is understood mechanistically in terms of coherence checking, which is purely computational and does not presuppose any semantic function. This chapter analyzes this conceptually and demonstrates that this account is descriptively adequate by citing a recent experiment on zebra finches, even though discrepancy detection is not always related to intentionality.

Functionalist Interrelations among Human Psychological States Inter Se, Ditto for Martians

Functionalism is designed to allow that psychological states can be multiply realized. Mark Sprevak has argued that, for a functionalist account of psychological states to apply to creatures that are organized in a very different way from humans (call them Martians), the way a psychological state is functionally individuated has to be relatively coarse-grained (Sprevak 2009). The argument for coarse-grained individuation fails if we distinguish functionalism about what it takes to be a psychological state in general from functionalism about a particular state type such as belief. Theorists are not precluded from including functional relations to consciousness or deliberate judgment in their account of (human) belief, consistent with allowing that Martians would have their own collection of functionally interrelated psychological states. Sprevak’s coarse-grained functionalism implies an implausibly liberal form of extended cognition. The point about functional interrelations allows us to avoid that conclusion without jettisoning functionalism (as Sprevak suggests we should): records in a human notebook may not enter into the right interrelations with other human psychological states to count as beliefs; nor do they enter into any interrelations with Martian psychological states.

Situated Mental Representations

If we want to account for mental representations (MRs) as being used in scientific explanations and realized by neural correlates in biological systems, then we have to give up the traditional Fodorian view of rigid symbolic MRs. However, we do not have to throw out the baby with the bathwater and accept anti-representationalism. Instead this chapter offers a new account of mental representations as real, nonstable, use-dependent, and situated. As such, they can be pivotal constituents of scientific explanations. The chapter demonstrates the empirical adequacy of this account by discussing cases of birds and rats relying on what-where-when memory. It argues that we need to involve nonlinguistic MRs to adequately account for their abilities. In this way, the alternative theory presented here provides a detailed description of situated mental representations: it combines a functionalist account of MR with a relational dimension that can vary with the situation type and that allows for nonstatic constructions of MRs in specific contexts.

Nonnatural Mental Representation

This chapter distinguishes between two types of representation, natural and nonnatural. It argues that nonnatural representation is necessary to explain intentionality. It also argues that traditional accounts of the semantic content of mental representations are insufficient to explain nonnatural representation and, therefore, intentionality. To remedy this, the chapter sketches an account of nonnatural representation in terms of natural representation plus offline simulation of nonactual environments plus tracking the ways in which a simulation departs from the actual environment. To represent nonnaturally, a system must be able to decouple internal simulations from sensory information by activating representational resources offline. The system must be able to represent things that are not in the actual environment and to track that it’s doing so; i.e., there must be an internal signal or state that can indicate whether what is represented departs from the actual environment. In addition, the system must be able to manipulate a representation independently of what happens in the actual environment and keep track that it’s doing so. In short, nonnatural representations are offline simulations whose departure from the actual environment the system has the function to keep track of. This is a step toward a naturalistic, mechanistic, neurocomputational account of intentionality.

Introduction

In the introduction we overview the main debates about mental representations. In the first part we focus on three questions. First, what explanatory role do mental representations play in different paradigms of cognitive science, such as classicism, connectionism, dynamical theories, and predictive processing? Second, what criteria do we need to introduce mental representations, and how can we distinguish between non-representational and representational cognitive states? And finally, how can intentionality be naturalized and what are the main challenges for naturalistic theories of intentionality? In the second part of the introduction, we present the individual chapters in the volume, and situate them within the context of broader debates.

Reifying Representations

The representational theory of mind (RTM) holds that the mind is stocked with mental representations: mental items that represent. They can be stored in memory, manipulated during mental activity, and combined to form complex representations. RTM is widely presupposed within cognitive science, which offers many successful theories that cite mental representations. Nevertheless, mental representations are still viewed warily in some scientific and philosophical circles. This chapter develops a novel version of RTM: the capacities-based representational theory of mind (C-RTM). According to C-RTM, a mental representation is an abstract type that marks the exercise of a representational capacity. Talk about mental representations embodies an ontologically loaded way of classifying mental states through representational capacities that the states deploy. Complex mental representations mark the appropriate joint exercise of multiple representational capacities. The chapter supports C-RTM with examples drawn from cognitive science, including perceptual representations and cognitive maps, and applies C-RTM to long-standing debates over the existence, nature, individuation, structure, and explanatory role of mental representations.

Representing as Coordinating with Absence

This chapter offers an account of mental representation that spells out the conditions for representational status. I argue that we should draw a distinction between issues concerning content and issues concerning representational role, and that the challenge of understanding representational role is at least as pressing as the more familiar challenge of giving an account of content. I then argue that we should understand representational role in terms of serving as a stand-in, and use the idea of structural isomorphism, with examples from animal cognition and from the literature on analog representation, to capture standing-in.

Defending Representation Realism

The representations that are invoked by theorists and researchers in cognitive science allow for a variety of different ontological interpretations. Along with both straightforward realist and eliminativist positions, there are various forms of deflationism. Deflationist accounts deny that the explanatory value or even accuracy of representational theories depends upon the existence of objectively real structures or states that play a representational role in the brain. Alternatively, many deny the existence of any sort of representational content that is objectively real and independent of our explanatory goals or interpretative activities. This chapter argues that this sort of representational deflationism doesn’t really work. After spelling out what a robust sort of realism does or does not entail, the chapter offers some general reasons for thinking realism is preferable to deflationism. Then it looks at three versions of deflationism and argues that all three either fail to capture our scientific practice, or collapse into a more straightforward sort of realism or eliminativism.

Deflating Deflationism about Mental Representation

The radically enactive, embodied view of cognition (REC) holds that cognition is not always and everywhere grounded in the manipulation of contentful representations. Arguments for REC have assumed that its opponents defend a substantive notion of representation—a notion that entails the existence of content-carrying mental states. This paper considers the prospects of representationalism of a different stripe—one that prefers deflated notions representation. For example, deflationists hold that talk of mental representations might just be a kind of convenient labeling that does not commit theorists to any substantive claims about the explanatory work done by psychosemantic properties. Taking the deflationary option thus undercuts the crucial motivation for positing mental representations in the first place. This chapter argues that, should the deflationist arguments prove warranted, they provide reason to hold that some forms of cognition are contentless, à la REC.