Why is the band Gamma so underestimated

Rhythmic oscillations are omnipresent in the brain and they reveal a lot about what the owner of the brain is doing: Even simple EEG measurements can tell with the naked eye whether someone is sleeping, daydreaming or solving a task in a highly concentrated manner. A generally accepted connection. However, it is unclear what the nature of this relationship is: Are the oscillating waves just a by-product of neuronal activity? Or do the rhythms actually fulfill a function in the processing of information in the brain?

"Especially the oscillations in the gamma band, between 30 and 80 Hertz, correlate so strongly with the increase and decrease in the activity of individual neurons - I can't even imagine that this should have no functional significance." says Nicolas Brunet, assistant professor at Millsaps College in Jackson, Mississippi and first author of a new study that for the first time shows a connection between visual perception under natural conditions and gamma activity in the human brain. The basic idea: if oscillations fulfill a function in perception, for example when seeing, they must always occur when there is something to see.

Oscillations: always to be seen when something can be seen?

For a long time it was not entirely clear whether this is really the case - what can be seen in the laboratory is not exactly representative of real life. Human as well as animal test subjects usually look at pictures of chessboards or striped patterns, the color and contrast of which can be precisely controlled. Numerous studies have shown that this simple, artificial stimulation is reliably accompanied by an increase in rhythmic activity in the gamma band. But what happens when realistic motifs are considered? Photos of landscapes, houses or people? A question that Nicolas Brunet got to the bottom of a few years ago in a study with macaques (a species of monkey), explains Pascal Fries, Director at ESI: “At first I was skeptical because I expected the natural and therefore uncontrollable ones Conditions could lead to unclear results. But our data then spoke a very clear language: Every single one of 65 natural images produced clear gamma oscillations. "

The two were all the more astonished when a working group from Stanford was unable to confirm their results in a study with a human test person. Due to severe epilepsy, the test subject had electrodes implanted that lay directly on the visual cortex. The American scientists showed the patient black and white photos with normal snapshot motifs such as houses, cars or faces. They found a significant increase in gamma band activity for only about half of the images viewed. The authors of the study concluded that the oscillations could not possibly be important for the perception of images.

An exchange of blows between different views is part of science, says Nicolas Brunet. “It's a bit like ping pong. You pass the ball until the one with the better empirical arguments prevails ”. He was far from ready to put the bat down. Together with Pascal Fries, he examined the data again himself.

Insightful second look

On closer inspection, they found that the strength of the oscillations depended on how much structure the picture contained. Images with little structure and larger uniform gray areas gave poor gamma responses; Images with strong structure across the photo also gave strong gamma responses. With the help of an algorithm that is often used in computer vision, Nicolas Brunet and Pascal Fries succeeded in precisely quantifying the correlation between image structure and rhythmic brain activity. In fact, the information content in the gamma band was so great that the researchers were able to tell with a 70% probability of being hit, using the gamma rhythm alone, which of two randomly drawn photos had been shown.

Convincing results, but with a downer. The data come from a single subject. Surgical interventions are only seen as the last of all solutions for patients with epilepsy. Only a few are actually operated on and only rarely is it necessary for the electrodes to be placed over the visual areas of the brain of all things. The significance of the results should not be underestimated, despite the character of the case study, says Nicolas Brunet: “We made a lot of observations on monkeys. Humans and apes are similar, but in the end apes are apes after all. " The results of the re-analysis suggest that the findings obtained through studies with macaques can actually be transferred to humans. Whether monkey or human: the more you can see, the clearer the vibrations. This supports the thesis that gamma oscillations play an important role not only for artificial stimuli and not only in animals, but also in the natural vision of humans.