A role for representations in inflexible behavior


Representationalists have routinely expressed skepticism about the idea that inflexible responses to stimuli (e.g. reflexive responses like the pupillary light reflex) are to be explained in representational terms. Representations are supposed to be more than just causal mediators in the chain of events stretching from stimulus to response, and it is difficult to see how the sensory states driving reflexes are doing more than playing the role of causal intermediaries. One popular strategy for distinguishing representations from mere causal mediators is to require that representations are decoupled from specific stimulus conditions. I believe this requirement on representation is mistaken and at odds with explanatory practices in sensory ecology. Even when sensory states have the job of coordinating a specific output with a specific input, we can still find them doing the work of representations, carrying information needed for organisms to respond successfully to environmental conditions. We can uncover information at work by intervening specifically on the information conveyed by sensory states, leaving their causal role undisturbed.

This is a preview of subscription content, log in to check access.


  1. 1.

    The canonical version of this approach is due to Skyrms (2010), who takes the informational content of a signal to be given by a list of the states of the world whose objective probability is changed by the signal and a vector indicating the extent and direction of the change in each case.

  2. 2.

    This issue is often referred to as the problem of error. The most popular strategy for addressing the problem is to understand error in terms of malfunction: error is possible only when information-carrying structures have the function of carrying information (e.g. Dretske 1986; Matthen 1988; Neander 1995).

  3. 3.

    Here I have in mind representationalists who endorse an information-theoretic approach to representation. Some representationalists, including Burge (2010) and Millikan (2004: ch. 5), reject the idea that representations are to be reductively analyzed in information-theoretic terms.

  4. 4.

    When I speak of receptors in what follows, I am talking specifically about exteroceptors.

  5. 5.

    My talk of representations being decoupled should be distinguished from Gärdenfors’s (1995) talk of detached representations. Detached representations in Gärdenfors’s sense “stand for objects or events that are neither present in the situation nor triggered by some recent situation.” They function “off-line,” so to speak. A representation can be decoupled without being detached.

  6. 6.

    For a recent defense of liberal representationalism, see Artiga (2016). For a response to Artiga, see Schulte (2019).

  7. 7.

    A common exception is habitual behavior: when a creature forms a habit, the conditioned response will continue without additional reinforcement.

  8. 8.

    For a recent critical discussion of Dretske’s views here, see Hofmann and Schulte (2014).

  9. 9.

    One might take issue with my characterization of the causal role of receptor signaling. I take receptor signaling to be an integral component of the ordered sequence running from the proximal stimulus to the response. One might think instead that receptor signaling figures in a larger causal chain extending from the ecologically significant feature of the environment to the reflexive response. Notice, though, that this alternative way of construing the causal role of receptor signaling is at odds with Dretske’s view about how signals succeed in carrying information. The signal does not need to be causally dependent on the ecologically significant feature of the world about which it carries information; the signal carries information about the signified in virtue of the fact that the signal changes the probability of the signified. In many cases a signal carries information about some ecologically significant feature of the world without the signal’s being causally dependent on the feature in question. The predator’s photoreceptors issue sensory signals carrying information about unprofitability, but these sensory states are not caused by unprofitability. They are caused by something that correlates with unprofitability, namely, bright coloration. Furthermore, the signal does not need to be causally dependent on the particular object or event about which it carries information. One event (signal) can carry correlational information about another (signified) when the correlation is due to a common cause. I discuss these points at length in forthcoming work.

  10. 10.

    I am assuming that selected-for inflexible behaviors are goal-directed. The relevant goal here is the biological goal of inclusive fitness. Even in a “reflex-like” organism such as a plant we should distinguish actions from mere reactions. When a plant’s DNA is damaged by exposure to sunlight, the plant is merely reacting to the stimulus. When a plant reorients its growth towards a light source (phototropism), the plant is engaged in a light-related activity. The ultimate goal of the activity is inclusive fitness.

  11. 11.

    Here I am drawing heavily on Shea’s (2018: 89–91) ‘evidential test’ for determining the role of information in successful performance of a task function.

  12. 12.

    Ramsey’s own positive suggestion about when receptor-issued signals function as representations is a slight tweak on Dretske’s view. Dretske asks when the information carried by a receptor representation is explanatorily relevant to the resulting behavior. His answer, of course, is that we find meaning or information at work when the representation is recruited to produce behavior through feedback-based learning. When a representation is recruited through feedback-based learning, the behavior is sensitive to the informational content of the representation. Following Dretske’s lead, Ramsey (2007: 27) suggests that a receptor-issued signal is functioning as a representation only when the information it carries is explanatorily relevant to the behavior it produces. The latter condition is met when the representation figures in learning or making inferences (2007: 141).

  13. 13.

    For further discussion of the idea that sensory systems are signaling systems, see Ganson (2018b).

  14. 14.

    While the distinction between innate and learned behaviors is relevant in some contexts, sensory ecologists tend not to worry about the ontogeny of behavior. They focus instead on the functional advantages of behavior and its underlying mechanisms (Stevens 2013: 8).

  15. 15.

    Skyrms (2010) attempts to capture the informational content of a signal via a list of the states of the world whose objective probability is changed by the signal and a vector indicating the extent and direction of the change in each case. Sensory ecologists are more often concerned with what Shea et al. (2018) call the functional content of a signal. In rough outline, we determine the functional content of a signal by identifying the state of the world causally responsible for the stabilization of the signaling system. For example, unprofitability is the state of the world that brings about the stabilization of aposematic signaling systems. So, the signals figuring in these systems have the function of carrying information about unprofitability. For further discussion see Ganson (2018b).

  16. 16.

    Burge (2010) does not endorse a de-coupleability requirement, so he does not fit the label “chauvinist” as we have defined it. However, he would agree with the chauvinist that the moth’s anti-predator behavior is not to be explained in representational terms. Burge concedes that the moth’s behavior is explicable in terms of functional registering of information, so he does not come into conflict with explanatory practices in sensory ecology. His point is just that we can account for the moth’s behavior without invoking distinctively representational notions like accuracy or veridicality conditions. An explanation in information-theoretic terms will suffice. For a compelling response to this line of argument, see Artiga (2016). Burge’s views on representation depend on the idea that perceptual psychology has a proprietary notion of representation. For criticisms of Burge’s claims about the role of representation in perceptual psychology, see Ganson et al. (2014). Burge also makes substantive assumptions about how explanations in perceptual psychology differ from explanations in biology. See Morgan (2018) for important criticisms of Burge on this issue.


  1. Artiga M (2014) Teleosemantics and pushmi-pullyu representations. Erkenntnis 79:545–566

    Article  Google Scholar 

  2. Artiga M (2016) Liberal representationalism: a deflationist defense. Dialectica 70:407–430

    Article  Google Scholar 

  3. Bateson P, Laland K (2013) Tinbergen’s four questions: an appreciation and an update. Trends Ecol Evol 28:712–718

    Article  Google Scholar 

  4. Bowdan E, Wyse G (1996) Sensory ecology: introduction. Biol Bull 191:122–123

    Article  Google Scholar 

  5. Burge T (2010) Origins of objectivity. Oxford University Press, Oxford

    Google Scholar 

  6. Cantwell-Smith B (1996) On the origin of objects. The MIT Press, Cambridge

    Google Scholar 

  7. Chemero A (2009) Radical embodied cognitive science. The MIT Press, Cambridge

    Google Scholar 

  8. Corcoran AJ, Barber JR, Conner WE (2009) Tiger moth jams bat sonar. Science 325(5938):325–327

    Article  Google Scholar 

  9. Cronin T, Johnsen S, Marshall N, Warrant E (2014) Visual ecology. Princeton University Press, Princeton

    Google Scholar 

  10. Dretske F (1986) Misrepresentation. In: Bogan R (ed) Belief: form, content, and function. Oxford University Press, Oxford

    Google Scholar 

  11. Dretske F (1988) Explaining behavior: reasons in a world of causes. The MIT Press, Cambridge

    Google Scholar 

  12. Dusenbery D (1992) Sensory ecology: how organisms acquire and respond to information. W. H. Freeman and Company, New York

    Google Scholar 

  13. Ganson T (2018a) Sensory malfunctions, limitations, and trade-offs. Synthese 195:1705–1713

    Article  Google Scholar 

  14. Ganson T (2018b) The senses as signalling systems. Australas J Philos 96:519–531

    Article  Google Scholar 

  15. Ganson T, Bronner B, Kerr A (2014) Burge’s defense of perceptual content. Philos Phenomenol Res 88:556–573

    Article  Google Scholar 

  16. Gärdenfors P (1995) Cued and detached representations in animal cognition. Behav Process 35:263–273

    Article  Google Scholar 

  17. Godfrey-Smith P (2013) Signals, icons, and beliefs. In: Ryder D, Kingsbury J, Williford K (eds) Millikan and her critics. Wiley-Blackwell, New York

    Google Scholar 

  18. Haugeland J (1998) Having thought: essays in the metaphysics of mind. Harvard University Press, Cambridge

    Google Scholar 

  19. Hofmann F, Schulte P (2014) The structuring causes of behavior: has Dretske saved mental causation? Acta Anal 29:267–284

    Article  Google Scholar 

  20. Martin G (2011) Understanding bird collisions with man-made objects: a sensory ecology approach. Ibis 153:239–254

    Article  Google Scholar 

  21. Martin G (2017) The sensory ecology of birds. Oxford University Press, Oxford

    Google Scholar 

  22. Martinez M (2013) Teleosemantics and indeterminacy. Dialectica 67:427–453

    Article  Google Scholar 

  23. Matthen M (1988) Biological functions and perceptual content. J Philos 85:5–27

    Article  Google Scholar 

  24. Millikan R (1989) Biosemantics. J Philos 86:281–297

    Article  Google Scholar 

  25. Millikan R (2004) Varieties of meaning. The MIT Press, Cambridge

    Google Scholar 

  26. Morgan A (2018) Mindless accuracy: on the ubiquity of content in nature. Synthese 195:5403–5429

    Article  Google Scholar 

  27. Neander K (1995) Misrepresentation and malfunction. Philos Stud 79:109–141

    Article  Google Scholar 

  28. Neander K (2013) Toward an informational teleosemantics. In: Ryder D, Kingsbury J, Williford K (eds) Millikan and her critics. Wiley-Blackwell, New York

    Google Scholar 

  29. Orlandi N (2014) The innocent eye: why vision is not a cognitive process. Oxford University Press, Oxford

    Google Scholar 

  30. Ramsey W (2007) Representation reconsidered. Cambridge University Press, Cambridge

    Google Scholar 

  31. Schulte P (2019) Challenging liberal representationalism: a reply to Artiga. Dialectica 73:331–348

    Article  Google Scholar 

  32. Shea N (2007) Consumers need information: supplementing teleosemantics with an input condition. Philos Phenomenol Res 75:404–435

    Article  Google Scholar 

  33. Shea N (2018) Representation in cognitive science. Oxford University Press, Oxford

    Google Scholar 

  34. Shea N, Godfrey-Smith P, Cao R (2018) Content in simple signalling systems. Br J Philos Sci 69:1009–1035

    Google Scholar 

  35. Skyrms B (2010) Signals: evolution, learning, and information. Oxford University Press, Oxford

    Google Scholar 

  36. Sterelny K (1995) Basic minds. Philos Perspect 9:251–270

    Article  Google Scholar 

  37. Sterelny K (2013) Thought in a hostile world. Blackwell, Oxford

    Google Scholar 

  38. Stevens M (2013) Sensory ecology, behavior, and evolution. Oxford University Press, Oxford

    Google Scholar 

  39. Wilkens L (2008) Primary inhibition by light: a unique property of bivalve photoreceptors. Am Malacol Bull 26:101–109

    Article  Google Scholar 

Download references


Many thanks to Peter Schulte and two anonymous reviewers for their helpful feedback.

Author information



Corresponding author

Correspondence to Todd Ganson.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ganson, T. A role for representations in inflexible behavior. Biol Philos 35, 37 (2020). https://doi.org/10.1007/s10539-020-09756-0

Download citation


  • Representation
  • Information
  • Dretske
  • Liberal representationalism
  • Reflexes
  • Sensory ecology
  • Signaling
  • Teleosemantics