Opinion
Modularity and cognition

https://doi.org/10.1016/S1364-6613(99)01289-9Get rights and content

Abstract

Modularity is a concept central to cognitive science, and Fodor’s analysis of cognitive modularity in his book The Modularity Of Mind has been widely influential – but also widely misunderstood. It is often claimed that the possession of some or other system-property is a necessary condition for that system to be modular in Fodor’s sense, but Fodor made it clear that he was not proposing a definition of modularity, nor proposing any necessary conditions for the applicability of the term. He was simply suggesting a number of system properties that are typical of modular systems. I argue that it is nevertheless possible to derive a useful definition of modularity from the kinds of arguments put forward by Fodor: A cognitive system is modular when and only when it is domain-specific. Given any such proposed module, the other features of modularity discussed by Fodor should be dealt with as empirical issues: for each feature (innateness, for example), it is an empirical question whether or not the proposed module has that feature.

Section snippets

Necessary conditions for modularity?

It is clear that many authors have understood Fodor to be proposing that the features of modularity listed in Box 1 are necessary conditions for the applicability of the term – that is, a cognitive system is not a module unless it possesses all of these features.

For example, Wojciulik, Kanwisher and Driver3 found in an fMRI study that neural activity in the fusiform gyrus in response to faces was modulated by attention, and took this result to be surprising because that region is supposed to be

What is information encapsulation?

Bishop5 and Hulme and Snowling6 both argue that Fodorian modularity is inconsistent with top-down processing, because such processing constitutes a violation of information encapsulation. They then go on to argue that, because top-down processing is characteristic in children when they are dealing with language, the concept of modularity is inappropriate in a developmental context. Both concede, however, that this is not a problem as far as the skilled language user is concerned because skilled

Could modules be ‘assembled’?

In the preliminary discussion in his book, Fodor poses the following question. If a computational system is a module, could it be

‘…“assembled” (in the sense of having been put together from some stock of more elementary subprocesses) or does its virtual architecture map relatively directly onto its neural implementation?’ (Fodor, p. 37.)

His answer was that modules are not assembled. It is interesting to note here not only that he gives no reason for making this choice, but also that, when he

A neo-Fodorian account of modularity

First, it is helpful here to distinguish between ‘knowledge module’ and ‘processing module’. A knowledge module is a body of knowledge that is autonomous: independent of other bodies of knowledge; for example, a linguist might say ‘syntax is a module’. A processing module is a mental information-processing system; for example, a psycholinguist might say ‘during language comprehension, sentences are parsed by a syntactic processing module’. I am concerned here with processing modules, not

Acknowledgements

I thank Robyn Langdon, Anna Woollams, Greg Currie and Hugh Clapin for their comments on earlier drafts of this article, and participants at the ARC Workshop on Cognitive Simulation, Theory of Mind and Modularity, particularly Martin Davies, Jay Garfield, Philip Gerrans, Philip Pettit, Kim Sterelny and Daniel Stoljar, for valuable discussion and criticism.

References (15)

  • R.I. Rumiati et al.

    Visual object agnosia without alexia or prosopagnosia

    Vis. Cognit.

    (1997)
  • D. Marr

    Vision

    (1982)
  • J.A. Fodor

    The Modularity of Mind

    (1983)
  • E. Wojciulik et al.

    Covert visual attention modulates face-specific activity in the human fusiform gyrus: an fMRI study

    J. Neurophysiol.

    (1998)
  • Attention to faces in the fusiform gyrus

    Trends Cognit. Sci.

    (1998)
  • D.V.M. Bishop

    Cognitive neuropsychology and developmental disorders: uncomfortable bedfellows

    Q. J. Exp. Psychol.

    (1997)
  • C. Hulme et al.

    Deficits in output phonology: an explanation of reading failure?

    Cognit. Neuropsychol.

    (1992)
There are more references available in the full text version of this article.

Cited by (233)

  • Resting state EEG network modularity predicts literacy skills in L1 Chinese but not in L2 English

    2021, Brain and Language
    Citation Excerpt :

    Although the EEG data were collected at the scalp level, the phase coherence estimates of phase relationships should be partially invariant to volume conduction (Chennu et al., 2017; Fraga González et al., 2018); hence, the modularity index computed based on the phase coherence matrix should be a valid indicator for brain network modularity. A neural network with high modularity has many connections within modules which allow for faster processing (Coltheart, 1999) and reduce network wiring costs (Clune, Mouret, & Lipson, 2013). Meanwhile, a high modularity neural network also has few connections between modules which results in a neural network with relatively independent modules and so increases the flexibility in learning (Ellefsen, Mouret, & Clune, 2015).

  • Delusions and theories of belief

    2020, Consciousness and Cognition
  • 3 x Phonology

    2022, Canadian Journal of Linguistics
View all citing articles on Scopus
View full text