Can divergent connectivity generate reliable sparse activity patterns?
The connectivity in the brain can be highly convergent and highly divergent. In the olfactory system of insects, for example, the connections from olfactory receptor neurons to projection (PNs) and local neurons (LNs) in the antennal lobe, the primary center for olfactory information processing, are highly convergent. The connections from the PNs to the Kenyon cells (KCs) in the secondary center of processing, the mushroom bodies, are, on the other hand, highly divergent. While the role of convergence and its properties in improving signal to noise ratios are if not fully understood at least generally accepted, the role of divergent connectivities is less clear. It has been suggested by us and others that divergence may serve to generate well separated, sparse activity patterns which are known to be beneficial for recognition (classification) and memory formation.
In this work we analyze general implications of divergent connectivity in the framework of unspecific (randomly connected) feedforward networks of generic (McCulloch-Pitts) neurons. We find that the distribution for the number of active neurons in the target area is very skewed with large probabilities for no response as well as a fat tail of non-zero probabilities for overwhelming responses of many neurons. The results imply that in the example of the locust olfactory system, divergent connections by themselves, cannot generate the desirable, and experimentally observed, sparse activity patterns if the observed high density of PN-KC connections is correct. This apparent contradiction can be resolved by feedforward gain-control, which may be implemented by the antennal lobe - lateral horn - mushroom body pathway.
The remainder of the work is dedicated to elaborating a hypothesis why nature may have chosen a much more costly solution - dense PN-KC connections and feedforward gain control - over a much simpler solution - sparse PN-KC connections.
Our results are applicable to many questions in neuroscience as divergent connections are ubiquitous, another prominent example being the connections from the entorhinal cortex to the dentate gyrus in the mammalian hippocampal formation.
This work was partially funded by the Biotechnology and Biological Sciences Research Council (grant number BB/F005113/1).
Conference:
Computational and systems
neuroscience 2009, Salt Lake City, UT, United States, 26 Feb - 3 Mar, 2009.
Presentation Type:
Poster Presentation
Topic:
Poster Presentations
Citation:
(2009). Can divergent connectivity generate reliable sparse activity patterns?.
Front. Syst. Neurosci.
Conference Abstract:
Computational and systems
neuroscience 2009.
doi: 10.3389/conf.neuro.06.2009.03.008
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
29 Jan 2009;
Published Online:
29 Jan 2009.