The idea that the connections between neurons are involved in learning and memory is due to the anatomist Santiago Ramón y Cajal, who around 1890 had discovered those connections (synapses) in microscopy studies of brain preparations. Over half a century later, the psychologist Donald Hebb specified how neurons should rewire in order to store sensory information. Hebb postulated a simple mechanism for memory formation and associative learning that has been successfully tested in numerous computational models of neural networks. The common feature of these models is learning through correlations, a modern reformulation of Hebb’s hypothesis known as the Hebbian learning rule: pairs of neural units that are excited by the stimulus should increase their correlation after stimulation; pairs of neural units that are inhibited by the stimulus should also increase their correlation; pairs of neural units, where one is excited and the other inhibited should decrease their correlation; pairs of units where at least one does not respond to the stimulus should not change their correlated activity.
Recently, as part of a team work, I have provided evidence of this learning rule in a biological neural network, the antennal lobe, which is the analogue of the olfactory bulb in vertebrates, the brain area that encodes olfactory information. Details of this work are provided in my doctoral dissertation, as well as in R.F.Galán et al. (2006) Neural Computation 18(1), p.10-25. [PDF]. A succinct explanation of this phenomenon for a broader audience is found in the press release on EurekAlert!.
Figure: Upper-left pannel: Change of cross-correlations between neural units (glomeruli) predicted by the Hebbian learning rule. Lower-left pannel: Recorded change of cross-correlations between glomeruli showing remarkable similarity to the predicted matrix. Upper-right pannel: Odor-induced pattern of neural activity in the honeybee's antennal lobe. Lower-right pannel: Retrieved memory trace from the spontaneous activity recorded during 2 minutes after removal of the odor. This trace provides a unique signature that infers the odor-induced pattern above.
Back to research webpage