Tuesday, 12 February 2013

'Embodied Cognition Is Not What You Think It Is' - the paper!

Whoops, we did it again - a paper based on the blog! This time we are in press at Frontiers in Psychology, in a Research Topic on embodied cognition, with a paper we somehow got away with calling 'Embodied Cognition is Not What You Think It Is'. 

This paper
draws from a lot of posts on the blog on embodied cognition, perception-action and language. We have used this opportunity to tackle some key issues head on, and we like this paper a lot :) We cover all the important issues and we set up what we think is the way forwards for embodied cognitive science. In addition, it sets up the ground work that we want to build on with our own Research Topic on Radical Embodied Cognitive Neuroscience. We've laid out what we think is the task facing the brain; this is what the brain is engaging with, and so this is what we think neuroscience needs to work with in order to understand what the brain is doing.

It's the kind of paper that will either land with a splash or vanish without trace. We want it to make some serious waves, and we're hoping that we can encourage people to publish free Commentaries on it at Frontiers, to challenge us or pick up our challenges, and, most fun for all, to come work with us to take all this forwards! We want this to be the basis of an empirical research programme and we want you all to work with us on it :) At the very least, feel free to pepper us with questions; this paper is the start of something for us, not the end and we're interested in the response to this paper to frame the next step.

The paper

We begin the paper by introducing our argument that embodied cognition is unavoidably a more radical hypothesis than a lot of people think. As soon as you allow things other than the brain to be central players in cognition, it changes the game entirely, and cognitive science will have to change to match. Embodied cognition explanations will replace traditional cognitive ones.

We lay out what we think are the four key questions truly embodied cognition research must tackle:
  1. What is the task to be solved?
  2. What are the resources that the organism has access to in order to solve the task?
  3. How can these resources be assembled so as to solve the task?
  4. Does the organism, in fact, use these resources?
We then review research in the robotics and animal literature (based in part on some of the great examples in Barrett, 2011) that establishes that truly embodied solutions are viable alternatives to traditional cognitive explanations. After that we look at the outfielder problem and the dynamical system's model of the A-not-B error (Thelen et al, 2001) as examples of research that follows this research programme with great success. 

We finish this section with a contrast; we look at research we don't think is really about embodiment (specifically, Eerland et al, 2011 and Miles et al, 2010) and describe why we think this. It's exciting to have this critique published, as a counterpoint to the papers themselves.

We initially ended the paper there, but the reviewers pushed us to take this analysis further to make the paper more than just a review. So we went 'balls out' (to quote Reviewer 3 :) and included Sabrina's work on a non-representational account of language. We talk about
linguistic information and its parallels to perceptual information, and take our first swing at the key question, how does linguistic information get it's meaning? We round this out with a critique of two 'embodied' attempts to understand language comprehension, the action-sentence compatibility effect (e.g. Glenberg & Kaschak, 2002) and the sentence-picture verification task (e.g. Stanfield & Zwann, 2001).

We wrap up with a shoutout to Clever Hans and Oskar Pfungst which Sabrina is inordinately proud of :)

Thanks to Frontiers for helping us out with costs, and thanks to our reviewers who gave us a fair ride and pushed us into adding the language section; it really makes this paper kick a lot of ass. Thanks also to the editors, Dermot Lynott, Louise Connell & Judith Holler for hosting the topic and accepting our feisty abstract. And thanks to all our readers on the blog who have asked us annoying questions and guided us in interesting directions as we attempt to pull all these threads together. 

Now go download the open access paper
, tweet about it, tell everyone you know - at the very least our alt metrics will look exciting!

ResearchBlogging.orgWilson, A., & Golonka, S. (2013). Embodied Cognition is Not What you Think it is Frontiers in Psychology, 4 DOI: 10.3389/fpsyg.2013.00058


  1. 'balls out', 'kick a lot of ass'? I'm all in favour of good science communication but if this is the future ...

  2. Cool! How long until the Broadway musical?

  3. Nice summary of your arguments and position. I suspect I will be using it with my students in my grad seminar. It seems to me that your two principal examples provide an obvious "shout-out" to Powers' perceptual control theory and homeostatic mechanisms (a la Cziko).

    1. Funny you should mention perceptual control theory; I've only encountered it recently and I'm not that into it. There's clear overlap with ecological psychology, but there was a lot about it that didn't quite hang together.

      (This is where I got into it a bit with one of the perceptual control people; I haven't thought much more about it, but I know Eric Charles has.)

    2. CONGRATS!

      I suspect you would warm up to PCT. Though I don't think you'll ever find it revolutionary, you would like the methods and results. (I suspect aspects were revolutionary when it started, but I'm not sure if any has a "wow" factor at this point.) The language difference do present a problem, but that can be overcome.

      At any rate, you can see it in a wider context by checking out the June issue of The Review of General Psychology... it will be pretty damn awesome.

  4. I appreciate the clear style of the paper. While I have problems with some of the posts loaded with terms like affordance, I enjoyed reading the Frontiers paper. Thank you.

    Would you, Andrew or Sabina, respond to may question under your preceding post? Note that Jackendoff in Foundations of Language refuted the term representation, though from different reason than you did. However, he uses it again, maybe to be understood.

    1. I have that question saved for me to go back to when I have time! Next few weeks, hopefully. I've never read the paper, I'll ask Sabrina; I'll add it to our pile though.

  5. You guys should abandon all that dull motor and language stuff and become sex researchers. Think about it. The task to be solved. The resources to be accessed. How are they assembled. Seriously.

  6. thanks for sharing.