A Computational Model of Neocortical-Hippocampal Cooperation and Its Application to Self-Localization
Recently many computational modules of hippocampal system have been proposed, investigating mainly the development of place cells, similar to mammals, but without employed by other structures for further use. We propose a biologically plausible computational model of neocortical-hippocampal cooperation, which is based on familiarity recognition by neocortex, followed by a recall process in the hippocampus. Our model is implemented and tested in a simulated robotic platform, which shows that neocortex is able to interact with hippocampus for the development of a self-localization behaviour.
KeywordsDentate Gyrus Inhibitory Neuron Place Cell Association Cortex Recurrent Connection
Unable to display preview. Download preview PDF.
- 3.Hafner, V.V.: Learning places in Newly Explored Environments. In: Proc. SAB 2000 (2000)Google Scholar
- 6.Kali, S., Dayan, P.: The involvement of recurrent connections in area CA3 in establishing the properties of place fields: a model. J. Neurosc. 20(19), 7463–7477 (2000)Google Scholar
- 7.Samsonovich, A., McNaughton, B.L.: Path integration and Cognitive Mapping in a Continuous Attractor Neural Network Model. J. Neurosc. 17(15), 5900–5920 (1997)Google Scholar
- 12.Norman, K.A., O’Reilly, R.C.: Modelling Hippocampal and Neocortical Contributions to Recognition Memory: A complementary Learning Systems Approach University of Colorado, Boulder, ICS, Technical Report 01-02 (2001)Google Scholar
- 16.Schraudolph, N.N., Sejnowski, T.J.: Competitive Anti-Hebbian Learning of Invariants. Advances in Neural Information Processing Systems 4, 1017–1024 (1992)Google Scholar
- 17.Urzelai J., Floreano D.: Evolutionary Robotics: Coping with Environmental change. In: Proc. GECCO 2000 (2000)Google Scholar