We explore information processing in the nervous system, and use the visual system as a convenient model. We employ systems analysis, borrowed from communication engineering, as the conceptual and methodological framework. We use single and multi-neuron recordings and optical imaging of neuronal activity, and have developed sophisticated techniques for stimulation of the visual system, and for the analysis of neuronal responses to these stimuli. We also record simultaneously the input and output of neurons in the visual thalamic nucleus, the lateral geniculate nucleus (LGN), in order to learn how information transmission from the retina to the visual cortex through the LGN is controlled. Applied mathematics and computers are used extensively for stimulation, analysis and modeling of the systems under study.
Currently we are investigating the properties of visual neurons in the retina, LGN and visual cortex of monkeys and cats, with an emphasis on the temporal, spatial and chromatic selectivities of these cells, and on the nature, source and function of the variability (noise) of neuronal responses. Since perception and behavior involve the coordinated activity of many interacting neurons, we also use optical imaging to visualize the spatio-temporal distribution of activity in neuronal ensembles in the visual cortex. In addition, we are recording from groups of neurons in the LGN and visual cortex, and study their dynamical interractions and the information that these neuronal ensembles carry. The experimental approaches are augmented by computational modeling of LGN neurons and cortical circuits. The topics we are focused on currently include: the representation of color in the primate visual cortex, the transmission of information from retina to cortex via the LGN, and the role of the descending pathways in the brain.
Recently we have started to record from populations of neurons in the brains of mice, from regions involved in depression, drug addiction and Post-Traumatic-Stress-Disorder (PTSD). The recorded regions incluide the Nucleus Accumbens and the Ventral Tegmental Area (VTA). We subject the recordings to various mathematical methods that quantify the population behavior, in order to determine whether pathological behavior is reflected in the network behavior.
Our research is supported by grants from the NIH's Eye Institute, Institute for General Medical Sciences, and the Institute for Mathematical & Biological Synthesis.