When I was a teenager, I read about Copernicus, who was the first to say loudly that the earth revolves around the sun and not vice versa. Indeed, all the accumulated astronomical evidence at that time fit the old postulation. However, the simplest trajectory of the sun and planets became more and more complicated. “I guess something is wrong,” Copernicus said to himself. I have always admired Copernicus, one of the scientific lighthouses for swimming against the stream when scientific evidence tells you something is wrong.Seventy years ago Donald Hebb suggested that learning occurs in the brain through modification of synapse strength; synaptic plasticity. But what experimental evidence has been gathered so far to support this theory? As a physicist and neuroscientist, this question has often perplexed me. The evidence available is typically limited, and includes large fluctuations, inconsistencies, and timescale ranges between hundreds of milliseconds, hours, days and more. A rule is hardly, if ever, to be found.Moreover, synaptic plasticity was introduced to quantitatively explain learning capabilities, but the interplay is still ambiguous.“The assumption about brain learning and functionalities seems to be wrong,” I have told my students.This issue was the main motivation for a very unusual transition in my scientific career. After having been a full professor, dealing with theoretical research and teaching learning theory for almost two decades, six years ago I became an experimentalist and opened my own wet-lab.Before recently publishing our research, some leading neuroscientists reacted to my deliberations regarding brain learning and neuronal activity by saying, “Learning should occur and we must accept the existing premise we have at the moment,” as well as “welcome to neuroscience, which differs from physics, where data is noisy and is difficult to collect reliably.” When I raised my early passing thoughts that learning might occur at the dendrites and that current experimental evidence cannot rule it out, the answer was “who cares? Synapses and dendrites are connected in series.”Since I now wear two hats as a physicist and neuroscientist, I was not satisfied with these answers. Finding fundamental rules is the primary goal, and only later must more complex networks and tasks be investigated. In neuroscience, as demonstrated by Hebb’s theory becoming so deeply rooted in the scientific community, the logic seems to be reversed.More money, resources and manpower are not the solution.The proposed model of learning seems weaker than the existing one. There are only several dendrites per neuron, as opposed to many thousands of synapses, hence the capacity of neural networks is expected to be diminished. Nevertheless, the quality of learning is measured by other parameters beyond sheer capacity, including; rate, robustness to noise and sensitivity to interference among several events and their relative timing.Indeed, the number of synapses in our cortex is commonly assumed to exceed a billion times a million, whereas the number of seconds in our lifetimes is only several billion. This comparison indicates that we can memorize around one million events per second without reaching “disc full” status. It is obviously too much. Something is wrong in the current postulation.Our recent work calls upon the neuroscience community to conduct an immediate investigation to prove or disprove current fundamental assumptions in neuroscience. It requires much less than one percent of the available resources and is doable with the many existing and outstanding experimentalists and state-of-the-art facilities. The main new ingredient to be added is the extended stimulation of a stable neuron from several directions with controlled timings and strengths.The author, a professor in the Department of Physics and Gonda (Goldschmied) Multidisciplinary Brain Research Center at Bar-Ilan University, has just published a revolutionary new theory on brain learning that overturns the most common assumption in neuroscience put forth by Donald Hebb in 1949. Three months ago he and his team published an additional paper in which they overturned a century-old theory on how neurons in the brain work. The research appeared in the journal Scientific Reports.