作者: Theodore W Berger , Robert E Hampson , Dong Song , Anushka Goonawardena , Vasilis Z Marmarelis
DOI: 10.1088/1741-2560/8/4/046017
关键词:
摘要: A primary objective in developing a neural prosthesis is to replace circuitry the brain that no longer functions appropriately. Such goal requires artificial reconstruction of neuron-to-neuron connections way can be recognized by remaining normal circuitry, and promotes appropriate interaction. In this study, application specially designed using multi-input/multi-output (MIMO) nonlinear model demonstrated trains electrical stimulation pulses substitute for MIMO derived ensemble firing patterns. Ensembles CA3 CA1 hippocampal neurons, recorded from rats performing delayed-nonmatch-to-sample (DNMS) memory task, exhibited successful encoding trial-specific sample lever information form different spatiotemporal patterns, identified online real-time, were employed within closed-loop behavioral paradigm. Results showed was able predict performance on same trial. Also, model-derived delivered as electrodes, improved under testing conditions and, more importantly, capable recovering when animals with activity compromised pharmacologic blockade synaptic transmission. These integrated experimental-modeling studies show first time that, sufficient about coding memories, real-time diagnosis manipulation process restore even enhance cognitive, mnemonic processes.