An Introduction to the Physiology of Ordinary Consciousness

The Homonculus

Daniel Dennett, a philosopher from Tufts University in Boston and author of Consciousness Explained, discussed the issue of whether there is a particular organ of the brain in which the sensory data coming in through the eyes to the primary visual cortex is converted further in order to make it what we see and report about, i.e. to make it "qualia". This is really the idea that there should be some particular organ of the self in the brain.

In looking at the connectivity in the visual system Dennett dispenses with the idea of the homunculus or a little man inside the brain which does the consciousness's work. As Dennett says, it is an empirical fact that there is not an homunculus in the brain, and even if there were we would still have to go through the same investigation of the homunculus' consciousness, and so on ad infinitum. For Dennett:

"The work done by the homunculus in the Cartesian Theatre ... must be distributed in both space and time within the brain." [Dennett, from his presentation to Tucson II]

It seems pretty unlikely that there would be a particular organ of the brain which takes the incoming, by-now-integrated (necessarily so, if this model were to work), data flow from one's body and the world and converts it (puts it up on the screen of the minds' eye) into the qualia laden subjective world that we report.

Each layer of processing is the subject of each pattern of data from the input sensors in its stream, transforming the pattern into salient features appropriate for the next layer. A multitude of processing subsystems take part in all the activities of the every-day. They interweave their connectivity throughout the brain, supplying divergent and convergent information pathways wherever necessary.

The N.C.C.

So the problem for those neurophysiologists who are interested in consciousness is in discovering the neurophysiology of mental processes. If there is no one organ of consciousness: what is actually going on in the brain and what could the distributed systems of the brain be that, at least, underlie consciousness? Put another way: What are the neural correlates of consciousness (which phrase is often reduced to the N.C.C.)? What should the search for the NCC provide us? and will it provide us with an explanation of consciousness? For example, the neural correlates of seeing are the primary visual pathway and its divergences. But vision is not the be-all and end-all of consciousness (the blind are after all perfectly conscious). Drawing on David Chalmers presentation to Tucson II: Being conscious involves having, and being able to report having, information which is in some way 'globally available' to an organism. If a neural mechanism of 'global availability' were found empirically, then such neural mechanism could be an NCC. Chalmers calls this a "bridging principle" between consciousness (the phenomenal) and neurophysiology.

Bridging Principle: - Consciousness ~ Global availability Empirical work: - Global availability ~ Neural process N Conclusion: - Neural process N ~ Consciousness [slide from Chalmers presentation at Tucson II]

There are several candidates for a neural mechanism of global availability, including Llinas' 40Hz oscillations and Baars' "Global Workspace".

A full story about the NCC should simultaneously: - 1. explain availability - 2. underlie consciousness (Though it probably won't explain consciousness!) [slide from Chalmers presentation at Tucson II]

Chalmers also, doesn't think that there will be a single NCC. It is more likely that consciousness is a distributed 'activity' and involves many different processing systems handling data in many modalities which becomes available within many forms of representation and description.

We are searching for the embodiments of subjectivity, and we must ask whether or not that is all we need to explain what it is that brings about consciousness.

Distributivity

The neurophysiologist Joseph Bogen argues that consciousness is subjectivity, and he comments that you can't see subjectivity; it's like looking for the wind, you can only see its effects. Bogen suggests that we look for a center (a nucleus) that has distributivity (i.e. widespread inward and outward connectivity) as a site that produces subjectivity as consciousness. He is referring to the Intra-laminar Nuclei which is a subpart of the Thalamus. We will return to Bogen and the thalamus later, once we have looked a little at the anatomy of the brain and the physiology of the nerves.

Neurophysiology

So what are the roles of neuroanatomy and neurophysiology? To quote from one of the great neurobiologists, David Hubel, who with Torsten Wiesel did most important work in elucidating the anatomy and physiology of the visual system:

"Anatomy seeks to describe the various elements of the brain and how they are put together; physiology asks how the parts function and how they work together." [Hubel, 1979, p40]

"The number of nerve cells, or neurons, that make up one's three pounds or so of brain is in the order of 10(to the power of 11) (a hundred billion) give or take a factor of 10. The neurons are surrounded, supported and nourished by glial cells, whose number is also large. A typical neuron consists of a cell body, ranging from about five to 100 micrometers (thousandths of a millimeter) in diameter, from which emanate one major fiber, the axon, and a number of fibrous branches, the dendrites. The axon may give off branches near its beginning and it often branches extensively near its end. In general terms the dendrites and the cell body receive incoming signals; the cell body combines and integrates them (roughly speaking, it averages them) and emits outgoing signals and it also serves for the general upkeep of the cell; the axon transports the outgoing signals to the axon terminals, which distribute the information to a new set of neurons." [Hubel, 1979, p39]

Physiology is impossible without the data of anatomy, but then we must ask what the structures anatomy describes are for?

Well, let's do a quick précis of neuroanatomy.

Sagittal view of Brain

Crank up the bandwidth across your Corpus Callosum The brain is generally divided into three layers of architecture which house the processing subsystems for its variety of functions:

The Hindbrain (or old brain) at the base of the brain provides the connection into the spinal cord. The medulla which controls some of the vital functions such as the heart and respiration. And the cerebellum, which is involved with the co-ordination of movement with sensory and somatic data from the rest of the brain and the body. It might be called the integrator of bodily control.

The Midbrain Contains the tectum which is involved with visual reflexes and detection of movement and the reticular formation which is an array of structures concerned largely with basic arousal, vital survival reflexes and the thalamus which distributes sensory and motor information to the cortex, and receives control information from the cortex.

The Forebrain or the cortex (also known as the cerebrum): overlaying the rest of the brain, greatly enfolded and larger by far than the older parts of the brain. Describing its functions working from the front of the brain to the rear: In the forward regions of the cortex, the frontal lobes are involved in planning and control of movements. Behind this is the primary motor cortex which is involved in the control of movement. Then the somato-sensory cortex which receives information about the body senses. Behind that are the association areas involved in memory and interpretation of sensory data. The primary visual cortex is at the rear of the brain and the primary auditory cortex is at the temporal lobes of the cortex.

Cortex from the left side

The spinal cord is the main nerve trunk, bringing all the sensory data of the body up to the brain and carrying the control signals back to the organs of the body. The medulla and the pons interconnect the brain and the spinal cord and cranial nerves (which carry information and control to and from the face and the surface of the head).

The midbrain ties together the visual and auditory systems with the nerves which control movement. The thalamus is the main relay station for the major sensory systems that project to the cerebral cortex, and as we shall see in the talk by Bernie Baars and Jim Newman the cortex projects back onto the thalamus to control what it receives from the sensory systems. The hypothalamus acts together with the pituitary gland as a master control system, releasing hormones which control the activities of other glands. The hormones then released by the endocrine glands feedback on the pituitary gland and hypothalamus regulating their activity. The hormones also probably have a great deal to contribute in the regulation of neurotransmitters and the overall modulation of brain activity. The limbic system seems to be mainly concerned with the complex analysis of odours and the appropriate responses to those data. The reticular system is what keeps us awake and aware of ther world, it is usually known as the reticular activating system.

Overlaying this whole complex is the cerebral cortex which

"...makes humans what they are. Within the vast human cortex lies a critical part of the secret of human consciousness, our superb sensory capacities and sensitivities to the external world, our motor skills, our aptitudes for reasoning and imagining and above all our unique language abilities." [Thompson, 1985, p22].

A neuron is a single cell with quite specific architecture. Incoming information is transferred to the neuron through the synapses, small chemical detector bulbs on the end of a tree-like array of fibers, extending from the main cell body, called dendrites. The neuron is activated by impulses transferred to it, via the synapses, from the axons of many other neurons, and fires its own impulse response, through its own axon, when enough of the appropriate inputs have been activated. The details of nerve structure; myelin sheathing, membrane potential and transmission, etc.

Roger Penrose, in Shadows of the MInd has a nice description of the nerve transmission picture:

...the biological picture is of classical nerve signals traveling out from the central bulb (soma) of the neuron, along the very long fiber called an axon, this axon bifurcating into separate strands at various places. Each strand finally terminates at a synapse - the junction at which the signal is transferred, usually to a subsequent neuron, across a synaptic cleft. It is at this stage that the neurotransmitter chemicals carry the message that the previous neuron has fired, by moving from one cell (neuron) to the next. This synaptic junction would often occur at the treelike dendrite of the next neuron, or else on its soma. [Penrose, 1994, p353]

Information processing takes place in the neurons and is a function of the numbers of connections of axons to dendrites via the synapses and whether or not the threshold for firing of that nerve is reached. The ease with which a synapse can operate is determined by the availability of neurotransmitter in the pre-synaptic bulb and the levels of neuromodulator molecules in the synaptic cleft as well as the excitatory or inhibitory nature of the synapse. All the excitatory and inhibitory synaptic junctions add to and subtract from the neuron's threshold trigger potential to determine whether it fires at any particular moment. There is some evidence to suggest that neuronal decision making takes place in the dendrites as well as in the averaging or thresholding of all synapses onto the neuron.

The Thalamus

Bernie Baars and Jim Newman have developed a concept which for Baars is expressed in psychological terms as the Global Workspace or the working memory, and for Newman is expressed physiologically in their proposed neural corrrelate of working memory the extended reticulo-thalamic activating system (eRTAS) with its massive interconnections to and from the cortex.
Baars and Newman talk about the "Global Workspace"

I think that for me this was the most significant idea of the conference: this demonstration of the psychological: the working memory; being very closely coupled with the physiological: the eRTAS; as a really functional solution to the NCC problem. It is the extraordinary amount of interconnectivity between the thalamus and the cortex that allows the cortex, particularly the forebrain, to regulate the flow of information to itself: so that it doesn't get swamped, so that it can focus on what it needs to know about from moment to moment. This is a huge, self-regulatory feedback circuit which binds the processes of the brain, or at least those important to consciousness, together over time, within the time scale of our normal moment-to-moment operating frame.

[See the e-seminars led by James Newman for a series of papers and commentaries on the "Thalamo-cortical Foundations of Conscious Experience"]

Now to return to Joseph Bogen. Bogen is on a search for the neural basis of "subjectivity". He suggests that any candidate for a neural correlate of consciousness must have widespread ditributivity, i.e. inward and outward connectivity. He refers to a subsection of the thalamus: the intralaminar nuclei as a site which satisfies his criteria as a primary site for subjectivity and he gives as primary evidence for this the fact that although quite widespread lesions in most parts of the brain do not cause loss of consciousness, merely modification of it, a very small lesion in the intralaminar nuclei will cause irreversible loss of consciousness.

Rodolfo Llinas provided some detail on the physical shape of the interconnections through the thalamus as mentioned by Jim Newman. He argues that "the form and the way things are connected in space are very important" [Llinas' presentation to Tucson II] and makes the wonderful statement that:

Llinas suggests that the view of the brain as a whole may give us an indication of how the brain works as a whole. He (as do Baars and Newman) describes the brain as being like the hub of a wheel. There is an array of radial projections carrying information patterns from all the afferent sensory systems upto the thalamus and projecting all the sensory pathways on up into the cortex. There are also a massive array of neural projections back from the cortex onto the thalamus. These project onto the array of gatelets of the intralaminar nuclei (see Bogen) and give the cortex control over what information it is getting.

Llinas has been using magneto-encephalography, (a vastly more sensitive brain activity sensing technology than electro-encephalography) to watch the brain functioning and has discovered bursts of 40Hz oscillation. These bursts are re-set when a sensory event occurs in an experimental situation and travel deep into the brain in a dialogue with the cortex. These continue while you are asleep. It is probable that dreams are the free-running of these oscillations through the thalamo-cortical system triggered only by the inherent noise of a biological system temporarily lacking the external sensory stimulus of being awake.

Susan Greenfield proposes a model of Neuronal Assemblies in the cortex which shrink and swell more or less on demand as the overall conscious situation requires.
See Susan Greenfield on "Neural Assemblies"

Alan Hobson has also developed a model of the ways in which consciousness changes over time. First he defines consciousness as

"a graded integration of multiple cognitive functions yielding a unified representation of the world, our bodies and ourselves".

This integrated system and its unified representation also goes through a continuum of states as we go through our days and, in longer term, our years. So this model that Hobson has developed shows that:

the level of consciousness changes as a function of activation;
the focus of consciousness changes as a function of input/output gating; and
the form (or perhaps the state) of consciousness changes as a function of modulatory neurotransmitter ratios.

He concludes that

"Consciousness is the forebrain's representation of the world, our bodies and ourselves. It is always a construction whose level, focus and form depends upon the brain stem."

Given that all processes take time to occur the feedback proportion of the output of the thalamo-cortical system will have been delayed with respect to the input information. In other words we have a short term memory, the previous state will be partly overlaid over the current state.

Also, given that we are in a condition of continuity (of continuous stimulus), even between waking and sleeping, there will always be new input being overlayed by previous input and so we live in that sense of continuity that we know as our temporal unity (one aspect of the unity of our consciousness). But this is a resonating system, it free-runs, it is an oscillating system. Llinas' 40Hz oscillations are one representation of the liveliness of this resonating system. These are the most complex and the most organized resonance's in one sub-system and will presumably be similar in other complexly organized layers of the brain. Like the visual system which keeps us in a generally seamless world.

Newman, James (1997) "Putting the Puzzle Together: Part I: Toward a General Theory of the Neural Correlates of Consciousness." in Journal of Consciousness Studies, 4, No.1, 1997, pp47-66.

Penrose, R. (1994) Shadows of the Mind. Oxford.

Thompson, R F. (1985) The Brain. An Introduction to Neuroscience Freeman.

Nauta, W. & Feirtag, M. "The Organisation of the Brain" Scientific American, Sept. 1979, pp78-105

[See Poggio & Koch, SCI AM, May 1987]