A century-old hypothesis about neurons and brain activity refuted


According to the neural computation scheme, each neuron functions as an excitable element. Incoming electrical signals from connected neurons are accumulated and the neuron generates a short electrical pulse, a peak, when its threshold is crossed.

For over a hundred years, it has been speculated that every neuron is characterized by a unique short rest time of about two to three milliseconds after the peak, during which the neuron cannot regenerate a consecutive peak. This period of rest is followed by a longer period of neural stuttering responses until full responsiveness is achieved.

In an article published in the journal Physical examination E, a group of researchers led by Professor Kanter of Bar-Ilan University in Israel defies conventional wisdom by highlighting three new features they have discovered experimentally on neuronal refractory (resting) periods.

First, the length of the sleep time can exceed 20 milliseconds, almost 10 times longer than previously assumed. These long refractory periods are further increased depending on previous neuronal activity.

Second, the rest period is sensitive to the origin of the input signal. Stimulation of the neuron in different directions (eg, “left” and “right”) results in different refractory periods. When the neuron is stimulated from the left, for example, the duration of the refractory period is much longer than the stimulation from the “right”.

Third, the neuron is a very precise element with a sharp transition from the refractory period to full reactivity, with no intermediate stuttering phase where the same stimulation of the neuron only results in an evoked peak with a certain probability.

“We came to this conclusion using a new experimental setup, but in principle these results, along with other anisotropic properties of neurons, could have been discovered using technology that has been around since the 1980s.” , Kanter said. “The century-old belief that has been entrenched in the scientific world has caused this decades-long delay.”

These new findings call for a reexamination of neuronal functionality beyond the traditional framework and, in particular, an examination of the origin of degenerative diseases. Neurons unable to differentiate between “left” and “right,” similar to distortions throughout the human body, are characterized by spurious activities. “Our results could be a starting point to discover the origin of these diseases. This work also opens new horizons for advanced deep learning algorithms and artificial intelligence-based applications mimicking neuronal anisotropic properties, instead of the isotropic nodes used in current machine learning applications. »Kanter concludes.

– This press release was originally posted on the Bar-Ilan University website


Comments are closed.