Superior Index Go to the next: Chapter 8
Print Files: A4 Size.
Common sense notions of how the world works suggest that a conscious decision to move a finger can be the efficient cause of finger movement. If we accept this, then we must accept that mental events can cause the firing of action potentials in the brain (and therefore in peripheral nerves and finally muscles). In other words, unless we are prepared to relegate our own consciousness to the status of a secondary epiphenomenon, which merely fools us into thinking that it affects our actions but which actually has neither causal powers nor any use at all, then we must accept that an individual mind does have the ability to move matter in its own brain. Philosophers may play with zombies and neurophysiologists may operate professionally on the assumption that consciousness is merely an epiphenomenon, but few of us are on a day-to-day level willing to give up the notion that we (i.e. our conscious minds) can exert an influence on what we do. Therefore one major requirement of a satisfying theory of consciousness must be that it should be obvious how whatever the theory proposes as being identical with consciousness can influence the physiological function of the brain.
This fairly basic requirement generally defeats dualist theories of consciousness, which argue that mind and matter are fundamentally different and thus that consciousness is non-physical. Quantum mechanical speculations notwithstanding, it is not at all obvious from our present understanding of physics how a purely non-physical entity or event could directly cause the rearrangement of matter that underlies the firing of even a single neural action potential. The requirement also causes serious problems for purely materialist accounts of consciousness such as the psychoneural identity theory. On this idea, consciousness per se can not have any causal properties, because consciousness is just what-the-brain-does and the brain, being a material instrument, simply carries out its appointed tasks in a purely mechanical way according to the deterministic principles of cause and effect. The "self" (which must be simply another brain process) then either does or does not become aware of the outcome of the other deterministic processes that produce behavior. In other words, according to the neural identity theory, our view of ourselves as having "free will" or being able to make conscious choices must be purely an illusion. While it is not impossible that an idea so counterintuitive could be correct, it would certainly be more comfortable to encounter a theory which did allow for conscious decisions to have some causal efficacy.
The theory being put forward here does allow for consciousness to have causal efficacy. While not frankly dualist, it is not materialist in the usually accepted sense either. To reiterate, the present theory is that consciousness is identical with certain spatiotemporal patterns in the electromagnetic field. Although they are not matter exactly, these electromagnetic patterns are physical in the sense that they are readily able to exert a back-influence on the physiology of neurons in the brain which generated them. In fact the evidence that such electromagnetic patterns do influence neuronal physiology is clear and present and does no violence at all to currently accepted notions of how the physical world works. So on the electromagnetic field theory of consciousness, there is no in-principle objection to the idea that consciousness per se can have a direct effect on the behavior of the organism.
This chapter starts by describing the fluctuating history of the idea that brain-generated electromagnetic fields are important in the normal function of the brain and then details more recent evidence that the sorts of electromagnetic patterns we are proposing as being identical with consciousness do routinely influence brain physiology on a millisecond to millisecond basis. For a full appreciation of this evidence, the reader should be familiar with the material presented in Appendix B, on the currently accepted basics of cellular neurophysiology and also that in Appendix C, on how these cellular mechanisms operate in the generation of what are today called field potentials.
The idea that endogenous electric fields are important in normal brain function probably first emerged in the Gestalt psychology of Wolfgang Kohler in the 1920s (Kohler, 1920 [143]; Kohler, 1929 [144]). Kohler's theory, much of which was originally published in German, is summarized in a review by Karl Lashley (Lashley & Semmes, 1951 [159]). To use Lashley's words, Kohler's field theory was based on the following assumptions:
This theory of Kohler's was remarkably prescient in many respects and was quite influential among psychologists of the day - until it ran into Lashley. Lashley performed what he saw as an experimental test of Kohler's ideas, as follows. In one monkey he placed four thin strips of gold foil under the dura along the surface of the visual cortex. In one more monkey he inserted a number of gold pins into the visual cortex at right angles to the surface. The aim of these manipulations was to short-circuit electric current flow in the cortex, thus distorting the postulated "figure-currents" and therefore (if the theory was correct) distorting visual perception. However Lashley made no attempt to determine whether there was in fact any disruption of current flow induced by his gold strips or inserts.
To examine the effects of the gold inserts on visual perception, Lashley tested the monkeys once, as soon as they recovered from the anesthetic. He used four visual discrimination tasks, on which the monkeys had been trained before the operation. The tasks were to choose one of two metal plates covering food dishes: in the first task they had to discriminate between a red and a green plate, in the second between a striped plate and a plate with a diamond on it, in the third between an "S" and a cross and in the fourth between dots in a diamond shape and random dots. The monkeys performed about as well after the operation as they had before, both on these tests and on tests like locating and retrieving bits of food that were in plain view (except that it was noted that the one with inserted pins "occasionally failed to see a small bit of food in the cup"). On the basis of this, Lashley felt justified in concluding that
"the action of electric currents, as postulated by field theory, is not an important factor in cerebral integration" (Lashley & Semmes, 1951 [159]).
By today's standards, it is probably fair to say that the paper reporting these experiments would never have made it past the referees of any reputable journal. Lashley reports no attempts to measure whether the pins or strips did in fact produce any of the "significant distortions of the electrical field" which they were assumed to have produced. The monkeys were tested only once. He reports no attempts to follow up the small behavioral change, suggestive of at least a minor distortion of vision, that was noted in the monkey with inserted pins. In fact the quality of these experiments is reminiscent of what must have been the quality of experimentation that led Lashley to state unequivocally in 1948 that "Uncomplicated destruction of major portions of the prestriate region ... has not been found to produce any disturbances in sensory or perceptual organization (Lashley as quoted in (Mishkin, Ungerleider, & Macko, 1983 [195])). In this latter instance it is difficult to know whether it was a preconceived idea that the neural tissue of the brain is functionally homogeneous that led to inadequate experimentation, or badly done experiments that produced the idea that the brain is functionally homogeneous, but Lashley was certainly convinced of this idea (Lashley, 1950 [158]). The notion that brain tissue is functionally homogeneous has of course been shown subsequently to be completely wrong, as a cursory glance at any basic neuroscience text written in the last thirty years will show. Lashley's conclusion that electric field effects are not important in the brain is similarly wrong, as shown by work quoted later in this chapter.
Whatever the quality of his experimentation however, Lashley was (and to some extent still is) an extremely influential investigator. Also, his conclusions about the irrelevance of electric field effects in brain function fitted in general with the independent ideas concurrently being arrived at by Jack Eccles, who was just as influential among physiologists as Lashley was among psychologists.
During the 1930s an ongoing debate raged between pharmacologists (led by Sir Henry Dale) and physiologists (led by Sir John Eccles) about whether neurons communicate by means of chemical or electrical transmission. At the time, the only verified example of chemical transmission known was vagal inhibition of the heart. This inhibition seemed to operate so slowly that physiologists in general and Eccles in particular preferred the idea that the faster communication known to occur at other synapses must be mediated directly by the electrical currents associated with action potentials. However when Eccles' own pioneering intracellular recordings in the central nervous system clearly showed that at least some central synaptic transmission actually is chemical (and when Sir Karl Popper had sat Sir John Eccles down and convinced him that in science it is not only acceptable but actually honorable to disprove one's own hypotheses), this strong preference for electrical transmission was swiftly and properly abandoned (Brock, Coombs, & Eccles, 1952 [39]). Unfortunately the pendulum of the zeitgeist then swung completely against any importance of intercellular electrical effects in the brain at all. Not only direct neuron-to-neuron electrical transmission (which we now know actually does sometimes occur) but also the idea that electric fields operating at larger distances might be important were forgotten in the new enthusiasm for chemical transmission. In the minds of most neuroscientists, the baby was, at least temporarily, thrown out with the bathwater.
After these two body blows, the idea that electric fields are important in brain function lay dormant for some years. But you can't keep a good idea down and experimental results soon started quietly accumulating to support the notion.
In the 1960s and 70s, work on the Mauthner cell in the medulla of the goldfish (Faber & Korn, 1973 [83]; Furshpan & Furukawa, 1962 [105]; Furukawa & Furshpan, 1963 [106]; Korn & Faber, 1975 [150]) showed that under normal conditions, a positive extracellular current generated by a group of interneurons produces a functional inhibition of the M-cell, hyperpolarizing its axon hillock and initial segment and thus preventing it from generating action potentials. One special condition favoring this particular example of an electric field effect in the central nervous system is that the extracellular resistance in this region is about five times higher than that of adjacent tissue (see Appendix C for an explanation of why this increases the size and effectiveness of extracellular field potentials). This high resistance is due to the fact that the axon hillock, initial axon segment and portions of the soma of the M-cell are surrounded by a so-called axon cap, which is penetrated by the terminals of the interneurons. The terminals of the interneurons do also make inhibitory chemical synapses with the M-cell (which confirms the suggestion that single neurons may mediate both electrical and chemical inhibition of their target cell), but these terminals do not have any electrotonic contacts (gap junctions) with the M-cell. This is an important point. Field effects involve coupling through the extracellular space, not via specialised low-resistance junctions.
In the case of the Mauthner cell, a reciprocal field effect of the M-cell on the interneurons has also been demonstrated (Korn & Faber, 1973 [149]). When the M-cell is active, it also effects a field effect inhibition of the interneurons. As well as the high extracellular resistance in the area, a second condition favoring the occurrence of this field effect is that the processes of the interneurons lie in parallel with the lines of extracellular current flow associated with the M-cell action potential (Faber & Korn, 1989 [84]). This arrangement favors the intracellular channeling of current and thus the occurrence of field effects.
Another structure in the central nervous system with the same encapsulated geometry as the M-cell is the mammalian cerebellar cortex. Here the axon terminals of basket cells converge on and surround the initial segment of the Purkinje cells (Palay & Chan-Palay, 1974 [215]). The orderly arrangement of Purkinje cells also establishes them as likely targets for field effects, particularly in the presence of radially directed extracellular currents. Thus it is not too surprising that electrical inhibition of the Purkinje cells has been demonstrated (Korn & Axelrad, 1980 [151]). Again as with the M-cell, the inhibitory basket cells mediate a two-component inhibition of their target Purkinje cells, with an early electrical phase being succeeded by a later classic chemical ipsp (inhibitory postsynaptic potential).
Another structure where relatively high extracellular resistances occur is the hippocampus. In rat hippocampus, resistivity measurements show an extracellular impedance 1.5-3 times higher around the pyramidal and granule cell bodies than in intermediate regions, for reasons that are not clear (Jefferys, 1984 [129]). Again pyramidal and granule cells are lined up in a conspicuously orderly fashion. On the basis of these two conditions, it might be expected that extracellular fields would influence pyramidal and granule cell physiology in the hippocampus and in fact such a finding has now been demonstrated by a number of studies. Extracellular electric fields generated by the activity of nearby neurons have been shown to influence pyramidal cell excitability and synchronization under normal conditions in the dentate gyrus of the rat hippocampus in vitro (Snow & Dudek, 1986 [264]) and also in rat hippocampal regions CA1 and CA3 both in vitro (Richardson, Turner, & Miller, 1984 [237]; Snow & Dudek, 1986 [264]; Turner, Richardson, & Miller, 1984 [297]) and in situ (Dalkara, Krnjevic, Ropert, & Yim, 1986 [69]; Taylor, Krnjevic, & Ropert, 1984 [285]; Yim, Krnjevic, & Dalkara, 1986 [319]). These electric field effects are associated with population spikes3 in neighboring cell populations and act by inducing subthreshold passive depolarizations of the cell bodies of nearby pyramidal cells, which can be demonstrated in the complete absence of chemical synaptic transmission. These subthreshold depolarizations have the effect of exciting and synchronizing the population of pyramidal cells.
A number of regions of the cerebral cortex obviously fulfil at least one of the requirements for the effectiveness of electric field effects, in that neurons there are arranged in an orderly, lined-up fashion. However no studies have been reported on extracellular resistivity measurements in various regions of the cerebral cortex and no studies comparable to those done on the hippocampus have been done to see whether field effects are significant under normal conditions in the cerebral cortex.
Most of the actions we perform in everyday life, and even many of the experiences that affect our actions, can happen quite well on a sub-conscious level. Emotions can be subconscious - people may be said to be "not in touch with their feelings". Desires are quite often kept out of consciousness, especially if they are socially unacceptable ones. Over-learned actions require very little in the way of consciousness - we are capable of performing extremely complicated acts like typing our thoughts onto a computer or driving home from work without being at all conscious of the details of what we are doing. In an emergency we often make quite complex decisions and act on them very fast, on a completely sub-conscious basis - we "act first and think later". We sometimes even make more leisurely decisions about what to do in a novel situation by withdrawing the conscious mind and allowing the right answer to simply "come" - all the work involved in weighing up pros and cons, assigning weights to various factors according to current overall goals and figuring out the best solution is then done on a sub-conscious level and the decision simply presents itself as a fait accompli to the conscious mind. In fact creativity, which we feel to be one of the major features that distinguishes us humans in all our conscious glory from present-day computers, actually seems to operate best on an unconscious level. The "aha!" experience is a well-documented feature of the mental life of unusually creative persons. In this kind of creativity, the most effective strategy seems to be feed in all the relevant information, chew it over consciously for a bit and then forget about the problem. The creative solution then simply pops into mind some time later, when the subject is doing something completely different, or even sleeping. The point is, while consciousness may be involved to a greater or lesser extent in all these features of life, it does not seem to be actually necessary for any of them. So what, one might wonder, is the good of it?
In fact, consciousness seems actually to be required in only two circumstances: (1) when we are in the learning phase of acquiring new information or a new skill (which is generally accepted as requiring access to the hippocampus or cerebellum) and (2) to access the language system.
As we have seen in Chapter 1, while the concepts of consciousness and attention are theoretically distinct, consciousness is at the very least intimately associated attention. One can not be fully conscious of something without paying attention to it and conversely one can not pay attention to something without its entering consciousness (except possibly in the anomalous case of brain-damaged "blindsight" patients, where attention to a particular location in space apparently enhances detection of stimuli at that location in the absence of consciousness of these stimuli (Kentridge, Heywood, & Weiskrantz, 1999 [142])). Similarly, learning is only readily accomplished if one pays attention to (and therefore is conscious of) the thing that is to be learned4, 5. The biological importance of this requirement to pay attention when learning is obvious: if brains stored a representation of absolutely everything that happened to the organism and not just the important things to which attention was paid, the system would soon saturate.
So in almost all circumstances, consciousness appears to be necessary (though unfortunately, as any harassed student in exam week will tell you, not sufficient) for learning. In this context it is interesting that the major structures in the brain that are known to be involved in learning are the hippocampus and the cerebellum (see Chapter 1). As shown in the earlier sections of this chapter, the hippocampus and the cerebellum are exactly the places where the importance of electric field effects in normal function has been demonstrated. It is not currently known exactly how the hippocampus operates in processing sensory information so that it can be stored in long-term memory, but what can be said is that (a) consciousness is necessary for learning and thus arguably for accessing the hippocampus (b) spatiotemporal patterns in the electromagnetic field are somehow involved in normal hippocampal function (c) the hypothesis put forward in this book is that spatiotemporal patterns in the electromagnetic field are identical with consciousness. Clearly this is an area of some interest, which would repay further experimental study.