The neural processing of taste

Abstract

Although there have been many recent advances in the field of gustatory neurobiology, our knowledge of how the nervous system is organized to process information about taste is still far from complete. Many studies on this topic have focused on understanding how gustatory neural circuits are spatially organized to represent information about taste quality (e.g., "sweet", "salty", "bitter", etc.). Arguments pertaining to this issue have largely centered on whether taste is carried by dedicated neural channels or a pattern of activity across a neural population. But there is now mounting evidence that the timing of neural events may also importantly contribute to the representation of taste. In this review, we attempt to summarize recent findings in the field that pertain to these issues. Both space and time are variables likely related to the mechanism of the gustatory neural code: information about taste appears to reside in spatial and temporal patterns of activation in gustatory neurons. What is more, the organization of the taste network in the brain would suggest that the parameters of space and time extend to the neural processing of gustatory information on a much grander scale.

Figure 1

There are multiple configurations of central taste circuits that could account for the perceptual consequences that follow stimulation of specific taste receptors. In hypothetical model A, input from taste receptor cells that express sweet receptors is encoded along a labeled line in the central nervous system (CNS): information about a sweet stimulus is received exclusively by central neurons that respond only to sweets. A central labeled-line "decoder" could then know that a "sweet" stimulus is present when the sweet "line" is active. In model B, input from taste receptor cells that detect sweets is distributed across neurons and represented by a pattern code in the CNS. Here, a sweet stimulus produces a unique pattern of activation across cells. A central pattern decoder could recognize that a sweet stimulus is present through knowledge of this pattern. Under either coding strategy, the stimulation of sweet receptor cells results in the correct recognition of a sweet stimulus.

Figure 2

Stimuli of different taste qualities produce unique patterns of relative firing among central gustatory neurons during the first second of stimulus processing. Here, spiking rates to oral stimulation with sucrose (a prototypical "sweet" stimulus), NaCl ("salty"), HCl ("sour") or quinine ("bitter") were compared between taste neurons recorded from the rat NST using a theoretic technique based on statistical decision theory. This model bears on whether different cells fire at similar or reliably different spike rates when under the drive of a particular stimulus. The outcome of this analysis as applied to all possible neuron pairs among six randomly-selected cells is represented graphically as a set of half-matrices. A blackened matrix element represents that the i^th neuron (denoted along the matrix columns) of a particular pair fired at a detectably faster rate than the j^th (rows). A non-shaded element denotes similar spike rates (not different) between neurons, whereas halftone shading indicates that the j^th fired detectably faster than the i^th. It can be seen that different stimuli produce unique relative response relationships among these cells. A downstream processor of these neurons with knowledge of the stimulus associated with each response relationship could, in principle, compute discriminations among these stimuli. Reprinted from [40], with permission of the Journal of Neuroscience.