Zoned-out ratThe New Scientist suggests that zoned-out rats may give us a clue to consciousness.

It’s all to do with the Default Mode Network, or DMN. You might think that when we stop concentrating on a particular task and sit back for a few quiet minutes the level of activity in our brains would fall, but it turns out this isn’t really so: instead, more or less the same level of activity appears to continue, but it switches to a different set of areas – in particular, a linked set of areas in the cortex and elsewhere. This is the DMN, but what is it doing?

A completely honest answer, I think, would be that we don’t exactly know except that it’s something other than concentrating on a task.  In human subjects the DMN seems to be associated with daydreaming, but also with other detached modes of thought.  Why would this help explain consciousness? It seems that in patients with locked-in syndrome, where consciousness is fully retained but the patient is unable to move, the DMN is functioning normally, whereas in persistent vegetative syndrome, where consciousness is absent, it is disrupted.

I can think of a further reason to think that this might shed light on consciousness. It’s not much of a stretch to see DMN activity as being the kind of thinking that isn’t directly related to inputs and outputs. When we’re working on a task those are crucial, but one plausible account of the role of consciousness is exactly that it lets us escape from giving instant responses to our surroundings and lets us develop longer-term plans, deeper understanding, and more complex behaviour. If the DMN represents useful mental activity detached from inputs and outputs it is exactly the thing whose existence the behaviourists denied, which is pretty much the same as one conception of consciousness.

The New Scientist and others speak of the DMN as associated with introspection, but I can’t see the evidence for that. To be daydreaming or thinking in general terms about stuff that is or might be going on is not introspection. I think there’s some confusion going on here between thinking internally and thinking about what’s going on internally: and perhaps a further suggestion that introspection= self-awareness = consciousness: those are tenable but debatable equations which don’t seem to be vindicated or disproved by the mere existence of the DMN. So perhaps the excitement is premature.

The rats are not that reassuring either. The New Scientist reports that analogues of the human DMN have been found in monkeys, and now even in rats. That’s interesting, but unless we rate the consciousness of rats unusually highly it seems to show that the DMN  cannot explain any uniquely human level of consciousness. Fair enough: I don’t disdain rat consciousness altogether: but it’s worse than that because, as I understand it, the evidence currently suggests that younger human children don’t have an identifiable DMN. It would be somewhat weird to attribute to rats a level of reflective consciousness which is absent in human infants – wouldn’t it?  If more were needed to put us off, it is not quite 100% agreed that the DMN is in fact a functional entity in itself; it could yet turn out to be more like the mere absence of the TPN, the Task Positive Network which is its opposite (or complement) – the similar set of areas which appear to work together when we’re engaged in a specific task. Perhaps the level of neuronal activity in the brain stays high, not because the DMN is really processing anything, but because the brain just uses a lot of energy to tick over?

Still, if the DMN doesn’t explain what consciousness is, it’s hard to resist the view that it’s telling us something about how it works. Problems with the DMN have been put forward as possible causes of Alzheimer’s, autism, and schizophrenia (I think everything has been put forward as a possible cause of schizophrenia). The range of problems is perhaps an indication of the vagueness of the theories. There is some good evidence of a correlation between Alzheimer’s and disrupted DMN: but then the DMN includes quite a siginficant sampling of some important areas of the brain, so that may not mean all that much. It could be that when consciousness is disrupted the DMN tends naturally to get disrupted too, without that implying that the DMN actually runs or constitutes even the less-focused forms of consciousness.

At the end of the day what we’re left with is that our brains – and even rat brains – don’t use the same circuits for task-related and non-task related activity, but go through a fairly large-scale switch of resources.  Even if we’re idly daydreaming about driving into town already, it seems we bring in a different set of neurons to do it with. There has to be some good reason for this, but what…?

9 Comments

  1. 1. Arnold Trehub says:

    Peter: “It’s not much of a stretch to see DMN activity as being the kind of thinking that isn’t directly related to inputs and outputs. When we’re working on a task those are crucial, but one plausible account of the role of consciousness is exactly that it lets us escape from giving instant responses to our surroundings and lets us develop longer-term plans, deeper understanding, and more complex behaviour.”

    I think you are on the right track, Peter. In the DMN reflective mode we can recall and relate many different images in our memories of the phenomenal world without the need to respond to the demands of a specific task. As I think about it, I’m reminded of what happens in the brain, according to the retinoid model, when we are faced with the task of searching and responding to a particular object among many different objects. In this case the brain must be concerned with a particular kind of input in order to give the right output. It turns out that the DMN finding of a change in the distribution, but not the level of activity. would be predicted by the the retinoid model. In *The Cognitive Brain*, Ch. 12, “Self-Directed Learning in a Complex Environment”, a computer simulation tested the retinoid-synaptic matrix system when the task was to search and find a particular object in a complex environment. (See here: http://www.people.umass.edu/trehub/thecognitivebrain/chapter12.pdf).

    According to the model, when searching for objects (pp. 210-211), two different neuronal processes were automatically triggered — all cells in the mosaic arrays of the synaptic matrices (SM) received an increment of inhibitory input, while those particular SM class-cells that corresponded to the searched-for object received a sustained increment of excitation which caused a priming of their associated images in the SM mosaic cells. The net effect is that the shift from the reflective (DMN) mode to the search task induces a shift in the distribution pattern of neuronal activity from a broad range of mosaic cells in the imaging matrices to just those mosaic cells that match the pattern of the object to be found. When the system stops searching (back to the DMN mode), all imaging matrices are disinhibited and the DMN areas of the brain become active again. This seems similar to what happens in the DMN experiments.

  2. 2. Bill says:

    Hi Peter,

    Why assume that DMN is “reflective” and introspective and thus (perhaps) self-reflective and therefore “conscious”? Could it not also be possible that the DMN is a more basal, active, but autonomous brain state that is less reflective and not even invloving the self conscious “I”?

    We have all had the experience of driving one place when we were supposed to drive somewhere else, eg. driving half way to work instead of to school to drop off the kids. These periods of time, “lapses” if you will, can involve the performance of highly complex tasks, but a mind that is elsewhere, and frequently the individual has no memory for the lapse period, just “coming to one’s senses” and remembering what task was actually intended by the “conscious self” before slipping into automaton (DMN?) mode.

    I would be very interested to hear from Arnold about how this every day experience fits into the retinoid theory. My assumption is that the retinoid system is disengaged during these lapses but that many sub system networks are very active until the retinoid center re-engages and we “come to our senses”.

    Thanks Peter for this important blog that seems to be the most informed and interesting clearinghouse for consciousness study and dialogue in existence.

  3. 3. Bill says:

    edit to above:

    These periods of time, “lapses” if you will, can involve the performance of highly complex tasks, but a mind that is elsewhere,

  4. 4. Bill says:

    I’ll try again

    edit to above:

    These periods of time, “lapses” if you will, can involve the performance of highly complex tasks, but a mind that is elsewhere, or “nowhere” as the absence of memory for the antecedents and events of the lapse would seem to indicate,….

  5. 5. Dan Lurie says:

    Bill,

    The reasoning behind the DMN being involved specifically in reflective cognition is in part due to the fact that the brian areas making up the DMN, when doing other tasks, are the kind of areas involved in complex cognition.

    I think what you’re referring to has a good point, but more to do with meta-awareness than the DMN specifically. I’d recommend checking out a paper by Christoff et al. in PNAS from 2009 that looks at DMN activity, mind wandering, and meta-awareness.

    http://www.pnas.org/content/early/2009/05/11/0900234106

  6. 6. Arnold Trehub says:

    Bill: “We have all had the experience of driving one place when we were supposed to drive somewhere else, eg. driving half way to work instead of to school to drop off the kids.”

    According to my model of the cognitive brain, this kind of experience results from our starting off on “autopilot” appropriate to a very familiar travel routine. In my own experience, it usually happens when the first part of the trip is the same as the more familiar route and I start thinking about something unrelated to my driving. Then I’m likely to miss the required branch off point in my travel routine and I end up in the wrong place. (For an idea of how our “autopilot” routines are encoded in the brain see The Cognitive Brain, Ch. 8 , pp. 142-145, “Encoding Plans”.) In this kind of mind wandering, the retinoid system is active but is out of the loop of sensory-motor control.

  7. 7. Bill says:

    Thanks Dan,
    I’ve skimmed and am re-reading the paper you cited. Very helpful in understanding the current understanding of the “DMN”.

    Thanks Arnold,
    My new copy of “The Cognitive Brain” arrived today. I look forward to delving in to it.

    Bill

  8. 8. Eric Thomson says:

    FWIW, poster this Tuesday at Society for Neuroscience meeting:
    ‘Is the default mode network a key network in disorders of consciousness?’ By Fernandez-Espejo et al.

    Abstract:
    The default mode network (DMN) has been defined as a group of brain regions which are known to be more active during resting state conditions. Recent studies have suggested that functional connectivity within this network may be related to the level of consciousness in severely brain-damaged patients. It is commonly assumed that functional connectivity reflects underlying structural anatomy. However, functional connectivity is also observed between regions where there is little or no structural connectivity. There are currently no studies that have assessed the structural connectivity within the default mode network in patients with disorders of consciousness. We acquired diffusion tensor imaging (DTI) data in a sample of 66 minimally conscious state and unresponsive wakefulness syndrome (also known as vegetative state) patients as well as in 18 healthy volunteers. DTI images were processed using FMRIB Software Library (FSL, v.4.1) following a standard protocol. Diffusion modelling and probabilistic tractography were carried out using FDT between spherical ROIs located in the posterior cingulate/precuneus, medial frontal cortex and temporoparietal junctions. Fractional anisotropy (FA) mean values were obtained from the resulting tracts. Group comparisons between patients and healthy volunteers were performed with those mean FA values obtained, using the global white matter FA as a covariate to control for the global damage. There were no statistically significant differences between patients and healthy volunteers in any of the tracts analyze. This result suggests that the structural connectivity within the DMN is not specifically impaired in patients with disorders of consciousness, but its impairment is part of more widespread damage to the global white matter. Therefore, it cannot be considered an essential network in the explanation of the anatomy of consciousness and its disorders. Further multimodal studies with larger cohorts, combining fMRI and DTI are needed to assess the relationship between functional and structural connectivity in this population as well as to explore their possible potential for explaining their clinical profiles.

  9. 9. Vicente says:

    Might be that consciousness results in part from DMN activity, but this works suggests that the quality of it is spoiled by the same DMN, and it could be convenient to shut it down as much as possible.

    Meditation experience is associated with differences in default mode network activity and connectivity

    Judson A. Brewera,1, Patrick D. Worhunskya, Jeremy R. Grayb, Yi-Yuan Tangc, Jochen Weberd, and Hedy Kobera (Yale Univ.)

    PNAS 2011 ; published ahead of print November 23, 2011, doi:10.1073/pnas.1112029108

Leave a Reply