This abstract, written for SFN, introduces the main result of my PLoS Computational Biology paper and makes a remark about neural tuning (SFN2012). I wrote this abstract while the NIPS and the PloS paper were under review. The poster (SFN2012poster) shows that a single saturating  dendritic subunit enables a neuron to compute linearly inseparable functions, but I presented also another result. This work also showed that a neuron computing such function is broadly tuned;  it is a prequel to what I am currently working on, i.e. neural tuning and coding. It demonstrates that a broad neural tuning is compatible with a positive impact of non-linear summation in dendrites.

I submitted this abstract to the Society For Neuroscience conference 2010 in San Diego (SFN2010). It presents a pseudo experimental protocol to test whether supralinear dendrites have an impact on the spiking property of a neuron. This work looks like a study from Polsky et al titled “Computational subunits in thin dendrites of pyramidal cells”. In this study, however, the neuron is silent, Polsky and colleagues focus on the subthreshold behavior of the neuron whereas in this poster the emphasis is on the spiking behavior of the neuron (SFN2010poster). Sadly, no experimentalist has yet tried to demonstrate this particular spiking behavior.

This abstract, submitted to FENS in 2010, deals with intrinsic plasticity in dendrites (FENS2010). Despite being one of my favorite, this topic is absent from my thesis manuscript. Also, the two main results of this study are absent from the abstract but present in the poster (Fens2010poster). Before talking about the result this poster defines the notion of storage capacity as the number of functions a neuron can compute. The two results are: first the storage capacity increases with the number of subunits; second the necessary range of synaptic weights to reach this capacity diminishes. The second result is by far the most interesting, one expects that the information storage capacity is boosted by the number of dendritic subunits, but it is a bit more surprising that dendritic subunits increase the efficiency of this storage. The intuition of this result is that as the maximal capacity can be reached with only binary weights with an unbounded number of dendritic subunits.

After my thesis manuscript. I am uploading my first abstract (Cosyne2010A). I wrote it with the help of Mark Humphries and Boris Gutkin. This abstract partly deals with the feature binding problem, even if we didn’t know its name at the time. Interestingly, the same year at COSYNE, another abstract also dealt with “binding problems”: Legenstein and Maass had wrote this abstract and they published the following year an article titled “Branch-specific plasticity enables self-organization of nonlinear computation in single neurons” where they presented the feature binding problem at the single neuron level. The poster is a bit messy and contains too much information (Cosyne2010Aposter) but I had a lot of useful feedbacks in this conference.