MIT and Brandeis
We are using intracellular, whole-cell recording techniques to record sensory responses from neurons in rat somatosensory cortex. Traditionally, extracellular recording techniques have been used to record action potentials evoked by sensory stimulation. These ``supra-threshold'' receptive fields provide information about the output of a given neuron when different sites in the sensory periphery are stimulated. However, extracellular recordings provide little information about the input to a given neuron. By recording intracellularly with whole-cell techniques, we can observe how supra-threshold, output receptive fields are transformed from the range of sub-threshold, excitatory inputs the cell receives. We can also observe in detail the inhibitory inputs (IPSPs) that are invisible to extracellular recording electrodes.
The preparation we are using to examine sub-threshold receptive field organization is the rat somatosensory (``barrel'') cortex (n = 46 neurons). In the rat somatosensory cortex, patches of densely packed neurons (called barrels) are laid out tangentially in the cortex in a series of rows that correspond anatomically to the rows of vibrissae on a rat's face (i.e., for each vibrissa on the rat's face, there is a corresponding barrel of cells in the cortex). Within a given barrel, the corresponding vibrissa is optimal for evoking action potentials and the receptive fields are small-usually only 1-3 vibrissa other than the corresponding vibrissa are able to evoke action potential responses.
Despite this focused anatomical and physiological organization of the cortical map, we found that most neurons had large sub-threshold receptive fields, with up to 14 whiskers providing input to a single neuron. Sub-threshold receptive fields were usually bell-shaped: A single, "central" vibrissa evoked a large, often supra-threshold response, while the amplitude of input from the surrounding whiskers decreased with increased distance from the central whisker (i.e., vibrissae distant from the central vibrissa evoked smaller responses than closer vibrissae). However, some receptive fields contained multiple supra-threshold vibrissae peaks, separated by vibrissae that did not provide any detectable input to the cell. IPSPs were present in all neurons examined. IPSPs were evoked by vibrissae throughout the receptive field, even by the central vibrissa.
We are currently examining the role that these large, sub-threshold receptive fields play in rapid sensory plasticity. Specifically, we are interested in how small changes in the timing of EPSPs and IPSPs may gate action potential firing, and how different patterns of correlated input may shift receptive field organization. We are also examining the relationship between the reliability of a subthreshold sensory input and the probability of evoking an action potential.