Tag Archives: brain waves

The John paper 2


This is the second part of a look at an old paper: E. Roy John; The neurophysics of consciousness; Brain Research Reviews, 39, 2002 pp 1-28. It is about brain waves.

Some background is needed to understand his model. EEG measures the voltage potentials on the scalp and from these are inferred the voltage potentials on/in the the cortex outer layers. From the changes in these potentials it is possible to look at the underlying waves. The trace is mathematically separated into component waves and these are graphed in separate bands of frequency. The amount of energy in each band is calculated. This is the power spectrum. The conventional bands are: gamma (25-50 cycles per second or Hz), beta (12.5-20), alpha (7.5-12.5), theta (3.5-7.5), delta (1.5-3.5). These fluctuations in potential affect the activity of neurons. The potential across the neuron’s membrane has a threshold which allows initiation of an electrical signal to be propagated along the neuron cell and axon. The potentials imposed by the electrical waves bring the neuron membrane closer or further from the threshold and therefore make activity easier or harder. (John does not deal with any contribution glia cells make to this system – perhaps too early a paper for that.) The imposition of a wave will tend to synchronize the activity of neurons because they will tend to reach threshold at nearly the same time and then have similar periods of recovery when they cannot signal. Each cycle of the wave will bring more neurons into the synchrony. The waves arise in pacemaker cells that naturally oscillate at a particular frequency (like heart pace maker cells).

John gives the following picture of the actions of brain waves:

The observed predictability of the EEG power spectrum arises from regulation by anatomically complex homeostatic systems in the brain. Brainstem, limbic, thalamic and cortical processes involving large neuronal populations mediate this regulation, utilizing all the major neurotransmitters. Pacemaker neurons distributed throughout the thalamus normally oscillate synchronously in the alpha (7.5–12.5 Hz) frequency range. Efferent globally distributed thalamo-cortex projections produce the rhythmic electrical activity known as the alpha rhythm, which dominates the EEG of an alert healthy person at rest.”

Note: there are a number of ‘nuclei reticularis’ or ‘reticular nuclei’ in the brain and John does not say which he is referring to. The problem is not serious. There is an extention of the spinal cord that runs through the brain stem, midbrain and ends in the thalamus, called the reticular formation. Somewhere between the brain stem and the thalamus, very probably at the thalamus end is the nucleus he is referring to. The reticular system includes the ascending reticular activating system and it must be active for the brain to function normally.

Nucleus reticularis can hyperpolarize the cell membranes of thalamic neurons by gamma-amino-butyric acid (GABA) release, slowing the dominant alpha rhythm into the lower theta range (3.5–7.5 Hz), and diminishing sensory throughput to the cortex. Theta activity can also be generated in the limbic system, possibly by theta pacemaker cells in the septal nuclei which can be inhibited by entorhinal and hippocampal influences. Slow delta activity (1.5–3.5 Hz) is believed to originate in oscillator neurons in deep cortical layers and in the thalamus, normally inhibited by input from the ascending reticular activating system in the midbrain. Delta activity may reflect hyperpolarization of cortical neurons resulting in dedifferentiation of neural activity. Activity in the beta band (12.5–20 Hz) is believed to reflect cortico-cortical and thalamo-cortical transactions related to specific information processing. Activity in the gamma bands (25– 50 Hz) may reflect cortico-thalamo-cortical reverberatory circuits, as well as back-propagation of axonal discharges to the dendrites of cortical pyramidal cells, which may play an important role in perception as proposed in this paper.”

Note: Although the cortex does affect these rhythms, it is not the source of the original pacemakers.

John works through an example: “…Assume that a subject is asleep, with diminished activity in the ascending reticular activating system, an EEG dominated by slow delta and theta waves reflecting inhibition of the thalamus by nucleus reticularis and consequent diminution of sensory input to the cortex…a sudden increase of stimuli in the environment results in inhibition of nucleus reticularis releasing the thalamic cells from inhibition by n. reticularis. The dominant activity of the EEG power spectrum becomes more rapid, with return of alpha activity. Increased flow of information through the thalamus to the cortex is facilitated, resulting in cortico-cortical interactions reflected by increased beta activity. Coincidence detection by pyramidal cells comparing this exogenous input with readout of endogenous activity activates cortical-thalamic loops generating gamma activity and mediating perception of the sensory information. Collaterals (side branches) go to n. reticularis from corticothalamic axons. The cortex can activate n. reticularis by these axons indirectly en passage or directly by glutamatergic path-ways, to suppress the arrival of information to the cortical level. Indirectly, as an alternative result of cortical influences, dopaminergic striatal projections can inhibit the (reticular formation). Such inhibition enables inhibition of thalamic neurons by n. reticularis, blocking transmission from the thalamus to the cortex. The dominant activity of the power spectrum slows toward or into the theta range. The cortex can thus modulate its own information input. The potential role of this mechanism in awareness and the focusing of attention should be apparent.”

Examination of the momentary voltage fields (LFPs local field potentials) on the scalp reveals a kaleidoscope with positive hills and negative valleys on a landscape, or ‘microstate’, which changes continuously. Computerized classification of microstates observed in EEGs of 400 normal subjects, aged 6–80 years, yielded the same small number of basic topographic patterns in every individual, with approximately equal prevalence. The topographies of these instantaneous brain voltage fields closely resemble the computed modes of factor loadings obtained in SPC (calculated spatial principle component analysis) studies. This correspondence suggests that the SPC loadings are not a computational artifact, but may reflect biologically meaningful processes.

The mean microstate duration slowly decreases during childhood, stabilizing for healthy young adults at ~82+/-4 ms. Although the field strength waxes and wanes, the stable landscapes persist with this duration. … The transition probabilities from microstate to microstate are apparently altered during cognitive tasks. Different microstates seem to correlate with distinctive modes of ideation. The stability of the microstate topographies and their mean duration across much of the human life span again supports the suggestion of genetic regulation.

Note: John does not mention modes such as the default mode but this seems like a description of mode changes – again too early for that.

Perceptual time is regulated, parsed into discontinuous intervals. Although subjective time is experienced as continuous, brain time is discontinuous, parsed by some neuro-physiological process into epochs of ~80 ms which define a ‘traveling moment of perception’. Sequential stimuli that occur within this brief time interval will be perceived as simultaneous, while events separated by a longer time are perceived as sequential. Other evidence has led to similar proposals that consciousness is discontinuous and is parsed into sequential episodes by synchronous thalamo-cortical activity. Multimodal asynchronous sensory information may thereby be integrated into a global instant of conscious experience. The correspondence between the experimentally obtained durations of each subjective episode and the mean duration of microstates suggest that a microstate may correspond to a ‘perceptual frame’. The phenomenon of ‘backward masking’ or metacontrast, consisting of the ability of a later sensory input to block perception of an event earlier in time, suggests that perhaps two separate events within a single frame are required for conscious perception. These two events might represent independent inputs to a comparator. (This seems to mean the a stimulus must be stable over a good part of a frame to be saved.)

The exact time at which conscious perception occurs following sensory input is unclear. Certainly, it is delayed beyond 50–100 ms since stimuli are particularly susceptible to masking by a competing stimulus during this period. Psychophysical evidence shows that the perceptual frame closes at ~80–100 ms after occurrence of a specific event. Although it is clear that time for the brain is discontinuous, the frame duration may differ in the various sensory modalities. A mechanism may be required to synchronize sensory elements sampled at different rates in disparate modalities. Based on train duration studies, Libet has suggested that perception may occur as late as 300–500 ms post stimulus. Extending train duration of repetitive direct cortical stimuli up to but not beyond 300–500 ms lowered perceptual threshold. These train duration effects have been reproduced for stimuli applied to the cerebral cortex via intracerebral electrodes. Similar duration effects have been shown using repetitive transcranial magnetic or direct electrical stimulation of the cortex and sensory deficit or neglect in healthy volunteers.”

In order to achieve the stable persistence of LFP topography revealed by microstate analysis, while displaying such duration effects and susceptibility to disruption by masking stimuli, some reentrant or reverberatory brain process must sustain cortical transactions as a steady state, independent of the activity of individual neurons.”

Note: John seems to be thinking in terms of standing waves here.

Such a process, called the ‘hyperneuron’, has been postulated and described in some detail. This persistent electrical field, produced by reverberating loops, may correspond to a neural correlate of the ‘dynamic core’ postulated by Tononi and Edelman. According to this concept, there must exist a set of spatially distributed and meta-stable thalamo-cortical elements that sustains continuity of awareness in spite of constantly changing composition of the neurons within that set.”

More in a later post.