David Ferster - US grants

The striate cortex stands out among the different areas of the early visual syste chiefly because of the unique receptive field properties of its neurons. Intricate synaptic connections between cortical neurons and geniculate afferents, and amount cortical neurons themselves, produce essential changes in the form in which the visual image is encoded, a change that is of interest not only in its own right, but as a model for functions throughout the neocortex. An explanation of cortical receptive fields must include a description of synaptic inputs onto individual cells: their sources, their receptive field properties and how they interact. The experiments in this proposal are designed to provide such a description by examining synaptic potentials directly with intracellular recording techniqes. Electrical stimulation will be used to reveal the sources of synaptic inputs; visual stimulation will be used to characterize their visual response properties. The largest effort will be devoted to studying the origins of orientation selectivity in cortical neurons, in particular to determining whether the spatial organization of thalamic input is by itself sufficient to construct this unique cortical property, or whether intracortical inhibitory connections make a substantial contribution. The major experiment will be to record carefully the orientation selectivity of visually evoked IPSPs and EPSPs in cortical cells. Various methods will be used to insure that all synaptic potentials present in the cell are detected. In addition, analysis of recorded potentials will make it possible to determine the receptive field properties of individual presynaptic inputs. By performing the same experiment in the presence of a drug that inactivates ON-center cells of the lateral genieulate nucleus (APB), it will also be possible to determine how the interactions between On-center and OFF-center cells contributes to orientation selectivity. Finally, the interaction of X and Y cells of the LGN will be studied by activating them separately with an electrical stimulus in the optic nerve.

DESCRIPTION (provided by applicant): The neurons of the primary visual cortex are highly selective for specific features of the visual image. How this integration is accomplished by the intricate connections of the cortical neurocircuitry is the subject of this proposal. The mechanisms of integration will be studied by recording intracellularly from cortical neurons in the intact cat and analyzing the synaptic potentials evoked by visual stimulation. These experiments directly explore the visual information a neuron receives, from which presynaptic neurons it receives that information, and by what mechanisms the neuron processes the information. Three series of experiments are proposed for the next 5 years of the project. 1) We will explore the origins of cross-orientation suppression. This nonlinear property of simple cells, now 30 years old, represents some of the strongest evidence for feedback models of orientation selectivity in visual cortex. Our preliminary data indicate that cross-orientation suppression might arise not from intracortical inhibition, as is often proposed, but from amplification by threshold of small cross-orientation effects in the relay cells of the lateral geniculate. We will test this proposal by comparing the behavior of relay cells, membrane potential responses and spike rate responses in simple cells. 2) Cortical simple cells exhibit strong contrast-dependent nonlinearities: changes in the amplitude, time course and selectivity of visual responses that occur with increasing stimulus contrast. These nonlinearities, like cross orientation suppressions, appear to be inconsistent with feed forward models of cortical processing. As with cross-orientation suppression, however, we propose to test whether small nonlinear effects in geniculate relay cells, when amplified by spike threshold, might account for the bulk of contrast-dependent nonlinearity in simple cells. 3) Complex cells in cortex exhibit distinct state transitions: abrupt, spontaneous shifts of membrane potential between distinct ranges of potential termed UP and DOWN states. We will examine the interaction between these transitions and visual stimulation, asking whether visual stimuli affect the probability of state transitions, and conversely, whether the current state of a neuron affects the size and latency of visual responses.

DESCRIPTION: The neurons of the primary visual cortex are highly selective for specific features of the visual image such as orientation, size, depth, and direction of motion. The analysis of these features requires the integration of visual information not only from different regions of the visual field, but in the case of motion selectivity, from different moments in time. How this integration is accomplished by the intricate connections of the cortical neurocircuitry is the subject of this proposal. The mechanisms of integration will be studied by recording intracellularly from cortical neurons in the intact animal and analyzing the synaptic potentials evoked by visual stimulation. These potentials give a direct indication of what type of visual information a neuron receives, from which presynaptic neurons it receives that information, and by what mechanisms the neuron processes the information. Three series of experiments are proposed for the next 5 years of the project. 1. The contribution of synaptic input from the lateral geniculate nucleus (LGN) to the spatial selectivity of cortical neurons will be investigated. The region of cortex containing an intracellularly- recorded neuron will be cooled. The cooling will inactivate surrounding cortical neurons, leaving the neurons of the LGN as the only remaining functional synaptic input. The visual responses of the synaptic potentials arising from the LGN can then be measured quantitatively. This experiment could resolve a long-standing question about the origin of spatial selectivity in cortical neurons. 2. Current models of motion sensitivity in cortical neurons rely on multiple synaptic inputs, each of which responds to visual stimulation with a different time delay. The source of these delays is not known, but in one model, the delays are generated in a class of neurons in the LGN called lagged cells. This model will be tested by selectively inactivating lagged cells by the application of pharmacological agents to the LGN while recording from motion selective neurons in the visual cortex. 3. A little understood feature of the visual cortex is contrast normalization. Contrast normalization is critical to the function of cortical neurons in that it enables cells to maintain their selectivity for stimulus features in the face of changing stimulus contrast. Without it, the way in which cells encoded a visual stimulus would change with changing contrast. Intracellular recording will be used to examine the cellular mechanisms underlying contrast adaptation.

The neurons of the primary visual cortex are highly selective for specific features of the visual image such as orientation, size, depth, and direction of motion. How this integration is accomplished by the intricate connections of the cortical neurocircuitry is the subject of this proposal. The mechanisms of integration will be studied by recording intracellularly from cortical neurons in the intact animal and analyzing the synaptic potentials evoked by visual stimulation. These potentials give a direct indication of what type of visual information a neuron receives, from which presynaptic neurons it receives that information, and by what mechanisms the neuron processes the information. Four series of experiments are proposed for the next 5 years of the project. 1) The precise spatial geometry of the receptive fields of simple cells will be mapped using rapidly flashing small stimuli. From these maps, the orientation selectivity of the neurons will be predicted with a linear model, and the predictions compared to the measured orientation selectivity of the neurons. The results will help to evaluate the merits of two competing models of cortical function. 2) We will measure changes in orientation selectivity that occur during the early phases of the response to flashing stimuli. A recently introduced method will be used to make precise measurements at times when the responses are just emerging from the background. These measurements will be compared to the predictions of orientation selectivity obtained in first experiment. 3) We will explore the cellular mechanisms underlying the phenomenon of contrast adaptation, in which prolonged exposure to a pattern renders the nervous system less and less sensitive to that pattern over time. Two possible mechanisms will be explored: changes in the electrical properties intrinsic to the affected neuron, and changes in the efficacy of the synaptic input to the neuron. 4) Although they are the primary determinate of cortical function, the properties of individual synapses in the cortex have never been studied in the intact animal. We will do so by recording intracellularly from nearby pairs of cortical neurons, searching for synaptic potentials triggered in one neuron by spikes in the other.

EIA-0217879
Northwestern University
Ferster, David

Collaborative Research: CRCNS: Detection and Recognition of Objects in Visual Cortex

A three way collaboration between the laboratories of Profs. T. Poggio at MIT, D. Ferster at Northwestern University and C. Koch at Caltech is exploring and evaluating the hypotheses that the cortical organization and the neural mechanisms of visual recognition can be explained by a coherent theoretical framework built on two existing computational models for recognition and attention and, secondly, that a combination of physiological work on monkeys and cats, together with visual psychophysics can be used to test and refine the theory. The research is organized into three main projects. The work at MIT is guided by a quantitative hierarchical model of recognition, probing the relations between identification and categorization and the properties of selectivity and invariance of the neural mechanisms in IT cortex. The work at Northwestern University is testing a key prediction of the model about the nature of the pooling operation (a max operation vs. a linear sum) performed by complex cells in V1. The experiments are done in the anesthetized cat, intracellularly, to allow for a characterization of the underlying circuit and biophysical mechanisms. Finally, work at Caltech is extending the basic model of recognition by integrating it with a saliency-based attentional model. The computational component of this work, centered around the development of a quantitative model of visual recognition, constitutes the primary tool to enforce interactions between the investigators: the model suggests experiments and guides planning and interpreting new experiments.