Skip to main content

Mental Representation


"Mental representations" are the coin of contemporary cognitive psychology, which proposes to explain the etiology of subjects' behavior in terms of the possession and use of such representations. "How does a subject manage to move through her darkened bedroom without stumbling over the furniture? She has an accurate mental representation of the room's layout, knows her initial position in the room, and is able to use this representation, in roughly the way a mariner uses a chart, to navigate through the room." "How does a sighted subject manage to recover information, available in the retinal image, about 'what's where' in her environment? She computes a series of representations, using information present in the retinal image, that eventuates in a three-dimensional representation of the distal objects present in the subject's visual field." "Why do native speakers of English have difficulty recognizing the grammaticality of so-called garden-path sentences such as 'The horse raced past the barn fell'? In recovering the meaning of a sentence, a speaker first constructs a representation of the syntactic structure of the sentence. In the case of garden-path sentences, the parsing processes that construct this representation mistakenly take the sentence's subject noun phrase to be a complete sentence, thus concluding that the entire sentence is ungrammatical." Cognitive ethologists offer similar explanations of many animal behaviors: Foraging red ants are said to practice a form of dead reckoning to maintain a representation of their current location relative to their nest, which they use to find their way back; migratory birds are said to navigate using representations of various sorts (celestial, magnetometric, topographic, etc.) that are either innate or learned as juveniles.

If, as these explanations apparently assume, mental representations are real entities that play a causal role in the production of a subject's behavior, then presumably it makes sense to ask about the form in which the information contained in these representations is encoded. This question has been the focus of considerable debate, especially with respect to mental imagery. Descriptionalists argue that, subjective impressions to the contrary notwithstanding, all mental representation, including mental imagery, is descriptional in form; mental representations are said to represent in a way similar to the ways linguistic descriptions represent. Descriptionalists subscribe to a language of thought hypothesis, according to which all human cognition is conducted in a quasi-linguistic medium. Pictorialists, by contrast, argue at least some mental representations, notably those involved in mental imagery, represent in ways similar to the ways pictures represent. The issues in dispute here are not straightforwardly empirical. Neither party believes that we literally have descriptions or pictures in our heads; rather, their claims are about similarities to the respective ways that pictures and descriptions represent. But it is precisely these similarity claims that render this debate obscure. What are the respective ways that pictures and descriptions represent, and what are the salient similarities such that if they hold they would justify characterizing mental representations as being of one form rather than the other? It is not obvious that there is a definitive answer to either of these questions.

To describe the representations to which psychological and ethological explanations appeal as mental is not to imply that their possessors are conscious of them; typically the representations are nonconscious or subconscious. Nor is it to imply that these representations are nonphysical; there is no commitment here to dualism. Psychologists and ethologists presume that the representations to which their explanations appeal are neurologically realized, physical structures. The point of describing the representations as mental is simply to emphasize the particular explanatory role that these representations play in these explanations. The explanations undertake to explain a kind of purposive behavior on the part of a subject, in which the particular behavior exhibited by the subject is typically modulated in a characteristic fashion, not only by the goal or purpose of the behavior, but also by the environment in which the behavior is exhibited. Thus, for example, our subject's movement through her darkened bedroom is modulated by her knowledge of the current layout of the room. The mental representations that figure in these explanations serve two distinct explanatory roles: (1) They explain why a subject behaves in one way rather than anothershe behaves as she does because she currently has this particular representation rather than another, and this representation is causally efficacious in the etiology of her behaviorand (2) they explain how the subject's behavior manages to be modulated (in characteristic ways) by her environment. Mental representations are able to play this dual explanatory role by virtue of possessing both physico-formal and semantic (intentional) properties that are linked in such a way as to ensure that a subject's environment can modulate her behavior. Basically, the cognitive processes that make use of mental representations are causally sensitive to the physico-formal properties of these representations that encode their semantic properties in much the way that sound-reproduction processes are sensitive to the physico-formal properties of records, tapes, and CDs.

Commonsense psychological explanations of behavior standardly appeal to beliefs, desires, intentions, and other so-called propositional attitudes (e.g., "Jones went to the refrigerator because he wanted a beer and believed there to be one there"). Behaviorists and eliminativists have challenged the legitimacy of these explanations, arguing that propositional attitudes either do not exist or do not figure in the etiology of behavior. Impressed with the prominent explanatory role of mental representations in cognitive psychological and ethological explanations, many philosophers of mind, notably Jerry Fodor, have proposed establishing the materialistic respectability of these explanations by appeal to the notion of mental representation. Their strategy is to explicate propositional attitudes in terms of mental representations. They defend a doctrine called the representational theory of mind (RTM), which holds that possessing a propositional attitude (e.g., believing that it is sunny today) is a matter of having a mental representation that (1) expresses the propositional content of that attitude (viz., that it is sunny today) and (2) plays a causal-functional role in the subject's mental life and behavior characteristic of the attitude in question (viz., the characteristic role of beliefs in modulating goal-satisfying behavior). More formally, for any organism O, any attitude A toward the proposition P, there is a mental representation MR such that MR means that (expresses the proposition that) P and a relation R (which specifies the characteristic causal-functional role of the MRs that are associated with a given A); and O bears attitude A to P if and only if O stands in relation R to MR. So formulated, RTM is silent as to the form of the mental representations that express the propositional contents of attitudes; proponents of RTM, however, invariably assume that these representations are syntactically structured entities, composed of atomic constituents (concepts) that refer to or denote things and properties in the world. More colorfully, these representations are sentences in the language of thought. The structure and meaning of these sentential representations purportedly explain the particular semantic and causal properties that propositional attitudes exhibit.

RTM is clearly realist in its construal of propositional attitudes: It purports to explain, not only what they are, but also how they could have both the causal and semantic properties that common sense attributes to them (viz., of being causally efficacious in the production of other thoughts and of behavior, and of being semantically evaluable, as, e.g., true or false). RTM is equally realist in its construal of mental processes, which, it holds, are causal sequences of the tokenings of mental representation. These sequences are said to be proof-theoretic in character, with the sequential states in a thought process functioning like premises in an argument. Thought processes are, like arguments, generally truth preserving.

Proponents of RTM claim to find strong empirical support for the doctrine in the apparent explanatory (and predictive) successes of cognitive science, whose theories are heavily committed to the existence of mental representations. Critics tend to dismiss this claimed support, arguing that what is at issue is not whether there are mental representations but whether there are mental representations with the particular properties demanded by RTM. Critics argue that propositional-attitude contents cannot always be paired with mental representations in the way that RTM requires: A subject may bear a certain attitude to a proposition but lack, among the many mental representations that cognitive scientific theories attribute to her, any mental representation of that particular proposition. Thus, for example, more than one critic has pointed out that, while David Marr's computational theory of early vision (see his Vision [1982]) attributes to the visual system the assumption that objects in the visual field are rigid in translation, the theory does not attribute to the visual system an explicit representation of that assumption; rather, the assumption is implicit in the operation of visual processes. Proponents, for their part, have tended to dismiss such counterexamples as "derivative" cases, arguing that RTM nonetheless holds for what they term the "core" cases of propositional attitudes. Such a response presumes that there is a nonquestion-begging characterization of the class of core cases. It also presumes that the class so characterized includes those propositional attitudes that figure in the commonsense psychological explanations that RTM is intended to vindicate. It remains an open question whether either of these presumptions can be met.

Other critics of RTM have challenged the doctrine's apparent commitment to "classical" cognitive architectures that presume a principled distinction between mental representations, on the one hand, and the computational processes that are defined over these representations, on the other. These critics point out that connectionist computational models of cognition do not preserve such a distinction, so that, if, as these critics presume, cognitive architecture is connectionist rather than classical, then RTM is untenable. Not surprisingly, proponents of RTM have been in the forefront of efforts to demonstrate that cognitive architecture is not connectionist.

Still other critics of RTM have focused on the semantics of the postulated mental representations, arguing that, if RTM is to provide a materialistic vindication of explanations that appeal to propositional attitudes, it must be possible to provide a "naturalistic" semantics, a theory of content, for these representations. By such a semantics these critics understand a materialistic account, invoking no intentional or semantic notions, of how it is possible for mental representations to have the semantic properties that they do (of being about things in the world, of being truth valued, etc.). There is general agreement among critics and proponents alike that none of the proposed naturalistic semantics is adequate, but, where critics see in these failures the symptoms of RTM's untenability, proponents see the beginnings of a difficult but eventually successful research project. There is disagreement among critics as to the import for cognitive science itself of there possibly being no naturalistic semantics for mental representations. Some argue that it would impugn the claimed explanatory role of mental representations; others argue that it would not. Whatever the upshot of these arguments, the untenability of RTM would not in and of itself impugn the explanatory role of mental representations in cognitive science, since that commitment to mental representations does not entail RTM. One can perfectly well be a representationalist in the way that most cognitive scientists are without also being a proponent of RTM.

See also Cognitive Science; Connectionism; Eliminative Materialism, Eliminativism; Imagery, Mental; Language of Thought; Mental Causation; Philosophy of Mind.


Block, N., ed. Imagery. Cambridge, MA: MIT Press, 1981. A collection focusing on debate between descriptionalists and pictorialists regarding mental imagery.

Crane, T. Mechanical Mind: A Philosophical Introduction to Minds, Machines, and Mental Representation. London: Routledge, 2003.

Devitt, M. Coming to Our Senses: A Naturalistic Program for Semantic Localism. Cambridge, U.K.: Cambridge University Press, 1996.

Dretske, F. Naturalizing the Mind. Cambridge, MA: MIT Press, 1995.

Dretske, F. Perception, Knowledge and Belief. Cambridge, U.K.: Cambridge University Press, 2000.

Field, H. "Mental Representation." Erkenntnis 13 (1978): 961.

Fodor, J. A. Concepts: Where Cognitive Science Went Wrong. Oxford: Oxford University Press, 1998.

Fodor, J. A. The Elm and the Expert. Cambridge, MA: MIT Press, 1994.

Fodor, J. A. "Fodor's Guide to Mental Representation." Mind 94 (1985): 5597.

Fodor, J. A. "The Persistence of the Attitudes." In Psychosemantics. Cambridge, MA: MIT Press, 1987.

Gunther, Y., ed. Essays on Nonconceptual Content. Cambridge, MA: MIT Press, 2003.

Jackendoff, R. Languages of the Mind: Essays on Mental Representation. London: Bradford, 1995.

Marr, D. Vision. San Francisco: Freeman, 1982.

Matthews, R. "Is There Vindication through Representationalism?" In Meaning in Mind: Fodor and His Critics, edited by B. Loewer and G. Rey. Oxford: Blackwell, 1991.

Matthews, R. "Troubles with Representationalism." Social Research 51 (1984): 10651097.

McDowell, J. Mind and World. Cambridge, MA: Harvard University Press, 1994.

Pylyshyn, Z. "The Explanatory Role of Representations." In Computation and Cognition. Cambridge, MA: MIT Press, 1984.

Pylyshyn, Z. Seeing and Visualizing: It's Not What You Think. Cambridge, MA: MIT Press, 2003.

Sterelny, K. The Representational Theory of Mind: An Introduction. Oxford: Blackwell, 1991.

Stich, S. Deconstructing the Mind. Oxford: Oxford University Press, 1996.

Stich, S., and T. Warfield, eds. Mental Representation: A Reader. Oxford: Blackwell, 1994. An excellent collection of papers on theories of content for mental representation; includes Field (1978) and Fodor (1985).

Strawson, G. Mental Reality. Cambridge, MA: MIT Press, 1994.

Thau, M. Consciousness and Cognition. Oxford: Oxford University Press, 2002.

Tye, M. Consciousness, Color, and Content. Cambridge, MA: MIT Press, 2000.

Robert J. Matthews (1996)

Bibliography updated by Alyssa Ney (2005)

Cite this article
Pick a style below, and copy the text for your bibliography.

  • MLA
  • Chicago
  • APA

"Mental Representation." Encyclopedia of Philosophy. . 14 Aug. 2018 <>.

"Mental Representation." Encyclopedia of Philosophy. . (August 14, 2018).

"Mental Representation." Encyclopedia of Philosophy. . Retrieved August 14, 2018 from

Learn more about citation styles

Citation styles gives you the ability to cite reference entries and articles according to common styles from the Modern Language Association (MLA), The Chicago Manual of Style, and the American Psychological Association (APA).

Within the “Cite this article” tool, pick a style to see how all available information looks when formatted according to that style. Then, copy and paste the text into your bibliography or works cited list.

Because each style has its own formatting nuances that evolve over time and not all information is available for every reference entry or article, cannot guarantee each citation it generates. Therefore, it’s best to use citations as a starting point before checking the style against your school or publication’s requirements and the most-recent information available at these sites:

Modern Language Association

The Chicago Manual of Style

American Psychological Association

  • Most online reference entries and articles do not have page numbers. Therefore, that information is unavailable for most content. However, the date of retrieval is often important. Refer to each style’s convention regarding the best way to format page numbers and retrieval dates.
  • In addition to the MLA, Chicago, and APA styles, your school, university, publication, or institution may have its own requirements for citations. Therefore, be sure to refer to those guidelines when editing your bibliography or works cited list.