The Life & Mind Seminar Network

Tokyo Life and Mind seminar today

Posted in Seminars by Nathaniel Virgo on July 2, 2013

Today we will have a Life and Mind Seminar by Christopher Buckley at the Tokyo University’s Komaba campus. All welcome!

1pm-2pm room 18-829, Komaba campus, Tokyo University

Christopher L. Buckley

The importance of sensorimotor feedback for behaviour has led many philosophers and cyberneticists to suggest that mind itself does not reside within the head alone but emerges from the interaction of brain/body and environment. Today the modern proponents of these ideas have begun to mount a systematic challenge to some of the dominating concepts in the cognitive sciences. Variously called the sensorimotor approach or enactivism by practitioners these ideas suggest that perception and action must combine to allow conscious experience.
Despite some notable successes in invertebrate systems the migration of these ideas from the cognitive sciences to mainstream, vertebrate dominated, neuroscience has been slow. For example even some of the first experiments in neuroscience demonstrated that brain state (a pattern brain activity and responses) is different whether a subject is passively viewing or actively engaged with an environment. Yet neuroscientists have largely appealed to internal mechanisms to explain these changes and have largely ignored the role of sensorimotor feedback. However I would argue this does not amount to conceptual resistance to the role sensorimotor feedback but, at least in part, reflects experimental practices which are dominated by heavily restrained, or anaesthetized animals, where body/environment feedback is minimised.
Recent experimental innovations mean that this state of affairs is beginning to radically change. Closed-loop experimental paradigms that utilise virtual reality in mice and fish, well circumscribed sensory-motor systems, and advances in brain machine brain interfaces (BMBI) are becoming more widespread. Consequently, in vivo electrophysiology and optogenetics of behaving animals is quickly becoming an achievable gold standard. This work places the sensorimotor loop at the heart of neural processing and promises to give enactivist ideas renewed relevance for mainstream neuroscience.
In this talk I will describe theoretical and experimental work we have been conducting in collaboration with several laboratories to examine the role of sensorimotor feedback in neural processing. I describe data analysis and modelling work of zebrafish larvae in swim simulators, the whisker system of mice, and cortical BMBI’s. I present evidence that sensorimotor feedback via the body/environment directly modulates the activity and responses of neural populations. Specifically I will describe evidence that the qualitative changes in brain dynamics observed when an animal engages the world constitutes an overall stabilization of neural dynamics by negative body/environmental feedback. Further I will show how the interruption of this feedback produces large responses in brain activity which we argue comprises an overlooked sensory mechanism that can enchance signal-to-noise ratios. Lastly I describe our new experimental work that explores the use of artificial feedback to control cortical stability and enhance sensory perception.

11 Responses

Subscribe to comments with RSS.

  1. Tom Froese said, on July 2, 2013 at 1:14 am

    I am intrigued by the following statement:

    Specifically I will describe evidence that the qualitative changes in brain dynamics observed when an animal engages the world constitutes an overall stabilization of neural dynamics by negative body/environmental feedback. Further I will show how the interruption of this feedback produces large responses in brain activity which we argue comprises an overlooked sensory mechanism that can enchance signal-to-noise ratios.

    This dichotomy, in terms of brain activity, between normal perception (based on operational sensorimotor loops) and abnormal “perception “(caused by interruption of sensorimotor feedback) reminds me of the predictions of the disjunctive theory of perception. This theory holds that “the veridical perceptions and hallucinations in question have a different nature: the former have mind-independent objects as constituents, and the latter do not.” Is it possible that the constitutive role of environmental feedback could explain this difference?

    • Nathaniel Virgo said, on July 2, 2013 at 1:33 am

      I’ve been struggling with something related to this for a while. From the enactive/sensorimotor/extended point of view, the objects we experience are a fundamental part of our experience. I believe that first-person experience cannot be explained without taking the world into account as part of the thing that’s doing the experiencing. (i.e. it’s not all in the brain, or even all in the brain-body.) But we do have experiences when we dream. Are these fundamentally the same thing as the experiences we have when we’re actively engaged in the world, or are they fundamentally different?

      The obvious answer (from an enactive point of view) is that they are fundamentally different. But then again, I personally dream (when I remember my dreams) in vivid Technicolor, and the redness of dream-red is every bit as vivid as the redness of reality-red. The two seem to feel very much the same to me. Dreams are inconsistent, of course, and differ from the experience of reality in many other ways, but it seems to me that the most basic “qualia” level of the experience is the same, or at least very similar. So from a first-person point of view it seems that there is *something* that’s fundamentally the same about the two types of experience, and I’m wondering how the enactive approach can explain this.

      In other words, if interaction with the world is fundamentally constitutive of conscious experience, then how can we explain dream-consciousness? I don’t see this as a challenge to enactivism but rather as an interesting question that could lead to new insights if we can figure out an answer to it. Do you know of any work on this question from within the enactive paradigm?

      • Tom Froese said, on July 3, 2013 at 9:59 pm

        I am happy to see that there is a general convergence in this community on the view that our experience is partially constituted by the world itself, which of course naturally poses dreams (and hallucinations, etc.) as interesting research topics.

        It does indeed follow that different mechanisms must be involved. But perhaps they do not need to be fundamentally different. I can remember some dreams where the real world did partially constitute my dream experience. For example, I once had a dream where I was in a house and there was this really strange noise everywhere and I couldn’t figure out where it came from, until I slowly realized that I was actually in bed and that my alarm clock was ringing! So here we have a more or less smooth transition from non-veridical to veridical experience.

        Accordingly, maybe we are dealing with a continuum of experience, where sense-making during dreaming tends to be much less constituted by the world than sense-making during waking, but where experience is never absolutely world-independent or absolutely world-dependent. The notion of sense-making is quite fitting to describe this middle way, because it jointly implies sensing the world and making the world, and it seems that the balance between these two processes may depend on the situation.

        I am curious to see how Mike’s paper will address this fascinating topic!

    • christopherlbuckley said, on July 4, 2013 at 1:35 am

      Hi Tom,

      Thanks for the interest. Sorry for not replying earlier, I thought I was on the list but actually I wasn’t.

      Im not sure how this relates to hallucinations Tom but I would like to be a bit more specific about what I mean by the interruption of feedback.

      I think I need to distinguish between two forms of feedback with the world, behavioural and sensory feedback. We have studied both.

      Sensory feedback arises in active sensing modalities such as visual saccading, whisking in rodents, sniffing. In this case we have argued that this feedback stabilises the brains dynamics. Furthermore this kind of feedback can be interrupted by relevant events in the world which produces large response in the brain (giving high SNR). We argue that animals set up active feedback with world so they are acutely sensitive to its interruption. Heres a link to rough pdf that explains the basic neuroscience of this for whisking.

      Behavioral feedback we would argue is slightly different. This is the feedback you set up during a behaviour, when cycling or playing tennis for example. The feedback is there to sustain the behaviour. The interruption of this feedback is problematic, falling of a bike, a meteor hits when your playing tennis.

      To look at this we’ve been studying flow stabilisation behaviour in zebrafish larvea. It seem that even this feedback corresponds to a stabilisation of brain dynamics. Specifically the onset swimming behavior in zebrafish larvea (they modulate swim velocity to keep the same position even when water flow is changing) stabilises brain dynamics. We think this is because sensorimotor feedback from the behaviour is overall negative is sign. In this case I’m not exactly sure what the role of interrupting feedback plays. Although you could suggest that when the behaviour is interrupted, say by a rapid chaotic eddy, the destabilisation in the brain constitutes a large noisy exploratory response that correspond to the fish seeking to regain smooth feedback with the world.

      • Jose A. Fernandez said, on July 4, 2013 at 2:09 pm

        Hi All, it is nice to read again all comments from this community!

        I think Chris’ talk is really interesting! I am curious about two points …

        Firstly, the consensus in visual neurophysiology is that cells in visual cortex should increase their responses when stimulated in their receptive fields. However, several lines of evidence report deviations to the general dogma…. In any case, we can assume that most regions of the visual sensory brain are at a steady state (let us say, the ongoing activity); then populations of neurons that are responsive to those sensory stimuli loose stability.

        Thus my question is: are those qualitative changes that Chris refers to regulated by attention? Perhaps, the further you attend a sensory event, the more you will enhance any response to that event and suppress others. Furthermore, adaptation at synaptic level can also recover local stability in the order of minutes in those responsive populations of neurons. I am struggling then with this idea: Is such a feedback (Chris’ one) the one that produces stability or the one that triggers some neuronal mechanisms (e.g. synaptic plasticity, homeostatic regulations, etc.) of the brains dynamics? In any case, I find fascinating the challenge that Chris has ahead: how the current dogma will explain the functional significance of your observations.

        Finally, how your observations of brain stability relate to Tononi’s homeostatic regulation ? According to Tononi, population of neurons (if not the whole brain) lose stability when the organism is awake. Then, homeostatic regulations of ‘normal’ brain activities take place during sleep. This idea is associated to his studies on sleep deprivation and local sleep in rats (i.e. a local phenomenon in which group of neurons shout down their activities simultaneously). Importantly, local sleep is further evidenced during sleep, and less evidenced during awake stage. So, I am very curious about whether your observation is general, or just related to sensory systems. In other words, can ‘offline’ stabilization of neural dynamics be explained by learned negative body/environmental feedback during awake state if, in principle, sensory systems are ‘disconnected’ during sleep? Maybe this is an unconnected question with your main stream of discussion but I’m curious about your answer.

        All the best,

      • Tom Froese said, on July 6, 2013 at 1:01 am

        Hi Chris,

        Many thanks for the detailed response. I am not sure I fully understand the distinction between the two kinds of feedback, since both require both perception and action. But I don’t think that this distinction is as crucial as another distinction that I am trying to make. I just had a nice chat with Takashi, and I explained my point as something like this.

        Feedback can be conceptualized in two ways: 1) as providing external input for internal processing, 2) as constituting a wider dynamical system for extended processing that includes body and world.

        When you talk about the animal (brain?) setting up active feedback to be sensitive to its interruption, and that the large response in the brain is to improve SNR, it sounds to me that your approaching the phenomenon from conception 1), not 2). Am I wrong?

        From the perspective of 2) things appear slightly different: the animal is already in the world to begin with and so is sensitive to interruption by default, and the large response in the brain is understandable because after interruption cognition and perception can no longer be “outsourced” into body and world and must be compensated for by additional internal activity.

        What would be interesting is if we could find an experimental way to differentiate between these two points of view!

        All the best,

  2. mjsbeaton said, on July 3, 2013 at 8:44 am

    I should probably have weighed in earlier to say: I really like the sound of your abstract, Chris, and it’s really heartening to hear real scientists starting to describe their work in this way.

    It makes me think that there will be less and less for philosophers to do, in the area! That said, Nathaniel, I have just finished a paper about exactly this. It is in final proofing stage, I was going to send a post about it shortly, when the rest of the special issue is finalised and the page numbers fixed, etc.

    To expand just a little, on the background to these issues, as I currently see it. There are disjunctivists and direct realists lurking within the heart of analytic philosophy, who I believe have said a lot that is exactly right about this problem. (It should be noted that these people are fundamentally and explicitly anti-representationalist; therefore, at least in that way, not at all what most enactivists think most analytic philosophers are like.) Also, for what it is worth, I also think there has been – and still is – a very strong current of internalism, about *experience itself*, in enactivism. So, I see a big part of my current project as trying to show how (and that!) these two quite separate streams of philosophy can work together.

  3. Ezequiel said, on July 3, 2013 at 12:33 pm

    Apologies for a very quick reply to issues that deserve a more extensive discussion (must finish off 2 papers by the end of the week!). But regarding Nathaniel’s question about dreaming, I expect something like an enactivist view on the subject is going to be the central topic of Evan Thompson’s forthcoming book on “Waking, Dreaming, Being”

    Evan has also written on mental imagery and this may be relevant too.

    And I like the idea that the agent-environment coupling stabilises neural (or generally internal) dynamics (would like to know more; is there anything written by Chris on this?). Miguel Aguilera (a PhD student in Zaragoza) has recently worked on a minimal model where he reaches a similar conclusion (as far as I remember):

    Aguilera, M, Bedia, M.G., Santos, B.A., Barandiaran, X.E. (forthcoming). The Situated HKB Model: how sensorimotor spatial coupling can alter oscillatory brain dynamics

    Not yet published but surely to be available from his site:

  4. mjsbeaton said, on July 16, 2013 at 8:25 am

    Well, everything about this is controversial, not to mention confusing. But anyway, here is my paper arguing that the world that we are experiencing really does (constitutively) form part of our experience. I am not sure whether Chris would agree… but I hope so!

    • Nathaniel Virgo said, on July 16, 2013 at 9:46 am

      They want me to enter my email address in order to receive the ‘free’ pdf, whereupon I will receive a non-opt-out-able email every time there’s a new issue. I don’t want to give the information or to receive the spam – do you have a link to a preprint version?

      • mjsbeaton said, on July 16, 2013 at 8:18 pm

        I retain copyright in my own paper. So I’ll email it to you, and put it on my page shortly, too.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: