Socioeconomic conclusions versus EEG sampling protocol error: A skeptical look at a recent bioRxiv preprint.

In the preprint below, the authors propose a new finding: the EEG alpha pattern correlates with educational level and familiarity with modern technologies. Yet (as also pointed out 4 months ago by Niko Bush in the comments on the paper on bioRxiv), EEG tracings were originally done in Germany in the 1920's and at MIT in the 1930's, and alpha was described well in persons in the US and in Germany during the early 20th century, a time when the technology felt relevant by the authors of the preprint was rare to nonexistent, and when most persons had only what the current authors would have seen as a low amount of education. Berger, writing in 1929, described alpha with average amplitudes of 15 to 20 micro-volts, using a primitive string galvanometer which might have attenuated voltages by a perhaps a third (see H. Berger, Uber das Elektrenkephalogramm des Menschen, Arch Psychiat Nervenkr 87:527-570, 1929).

So the low tech, low education contingent in the 21st century must differ in some other way. Is it due to low intelligence? But in these days of a commercial digital EEG of uncertain utility being commonly utilized by many psychologists, it's well documented that except in pathological cases of retardation the alpha is well seen in about 80% of normal EEG tracings, with about 20% showing lower amplitudes of less than 20 micro-volts and with about 4% showing only fast variant activity and no measurable alpha band (Niedermeyer and Da Silva, Elecroencephalography, Urban & Schwarzenberg Press, 1987).

Are the results reported in the current study related to technique of the EEG acquisition? Here, we may have found some clues. The study being considered used EEG hardware which was not set to acquire the actual amplitudes of the raw data. Rather, what was analyzed was a digital spectrum analysis of the data. This means that a normal alpha rhythm might have been unseen if sufficient muscle or movement artifact was present. Perhaps this explains the large degree to which the reported alpha "energy" calculation varied (10 to 1350, no units given). Of course, this would not yet explain the difference by educational and technlogical familiarity described in the study.

Further perusal of the Materials and Methods section does provide further clues. The equipment used was an Emotiv EPOC device which was used for EEG recording of duration 3 minutes per individual sample. The study states that "Participants answered a series of questions regarding their demographic, communication and mobility behavior in addition to having EEG recorded for three minutes while they were awake and seated with their eyes closed" which suggests that the sampling might have been set up and was finished within several minutes.

With standard clinical EEG acquisition, the EEG is set up over a period of 10 to 20 minutes, with the patient encouraged to relax and the patient often kept comfortably reclining or supine in a dimly lit room to promote sleep. The EEG alpha rhythm and amplitude is well known to be sensitive to anxiety, with about 3 times more subjects with anxiety showing low alpha voltage than in subjects who were not anxious (eg., Enoch, M.-A., White, K. V., Harris, C. R., Robin, R. W., Ross, J., Rohrbaugh, J. W. and Goldman, D. (1999), Association of Low-Voltage Alpha EEG With a Subtype of Alcohol Use Disorders. Alcoholism: Clinical and Experimental Research, 23: 1312–1319).

It seems quite likely to this writer that it is exactly that group of people who might be those of lower education, who were lower in familiarity with technology, who would also tend to be more anxious when multiple electrodes were first placed on their head. One wonders what the alpha power spectrum might have been if those individuals had been given more time to be less threatened by the EEG procedure, and the sampling done only after they were felt to be more relaxed as a result of taking that extra time.

ABSTRACT

----------------------------------

Modernization, Wealth And The Emergence Of Strong Alpha Oscillations In The Human EEG

Dhanya Parameshwaran, Tara C. Thiagarajan

doi: https://doi.org/10.1101/125898

Abstract

Oscillations in the alpha range (8-15 Hz) have been found to appear prominently in the EEG signal when people are awake with their eyes closed, and since their discovery have been considered a fundamental cerebral rhythm. While the mechanism of this oscillation continues to be debated, it has been shown to bear positive relation to memory capacity, attention and a host of other cognitive outcomes. Here we show that this feature is largely undetected in the EEG of adults without post-primary education and access to modern technologies. Furthermore, we show that the spatial extent and energy of the oscillation have wide variation, with energy ranging over a thousand fold across the breath of humanity with no centralizing mean. This represents a divergence in a fundamental functional characteristic of an organ demonstrating both that modernization has had a profound influence on brain dynamics and that a meaningful average human brain does not exist in a dynamical sense.

No comments:

Post a Comment

Risks for impaired post-stroke cognitive function

In a printed posted to the medRxiv preprint archive this month, I found a chart review of patients with stroke to determine factors (other t...