Gustatory responses of cortical neurons in rats. II. Information processing of taste quality

The present report was designed to investigate neural coding of taste information in the cerebral cortical taste area of rats. The magnitude and/or type (excitatory, inhibitory, or no-response) of responses of 111 cortical neurons evoked by single concentrations of the four basic taste stimuli (sucrose, NaCl, HCl, and quinine HCl) were subjected to four types of analyses in the context of the four proposed hypotheses of taste-quality coding: across-neuron response-pattern, labeled-line, matrix-pattern, and across-region response-pattern notions (88 histologically located neurons). An across-neuron response-pattern notion assumes that taste quality is coded by differential magnitudes of response across many neurons. This theory utilizes across-neuron correlation coefficients as a metric for the evaluation of taste quality coding. Across-neuron correlations between magnitudes of responses to any pairs of the four basic taste stimuli across 111 cortical neurons were very high and were similar. However, calculations made with net responses (spontaneous rate subtracted) resulted in less positive correlations but still similar values among the various pairs of taste stimuli. This finding suggests that across-neuron response patterns of cortical neurons become less discriminating among taste qualities compared with those of the lower-order neurons. A labeled-line notion assumes that there are identifiable groups of neurons and that taste quality is coded by activity in these particular sets of neurons. Some investigators have classified taste-responsive neurons into best-stimulus categories, depending on their best sensitivity to any one of the four basic stimuli, such as sucrose-best, NaCl-best, HCl-best, and quinine-best neurons; they have suggested that taste can be classified along four qualitative dimensions that correspond to these four neuron types (i.e., four labeled lines). The present study shows that responsiveness of each of the four best-stimulus neurons had similar profiles between peripheral and cortical levels. That is, when the stimuli were arranged along the abscissa in the order of sucrose, NaCl, HCl, and quinine, there is a peak response in one place, and the responses decreased gradually from the peak. However, such response characteristics do not favor the labeled-line theory, since they can be explained in the context of the across-neuron pattern theory. A matrix-pattern notion assumes that taste quality is coded by a spatially arranged matrix pattern of activated neurons.(ABSTRACT TRUNCATED AT 400 WORDS)     

Taste quality and neural coding: implications from psychophysics and neurophysiology

Historically, taste research has often been guided by the concept that there are only four (or possibly five) basic taste qualities (sweet, sour, salty, and bitter, and possibly "umami"). All other tastes have been presumed to be combinations of these basic tastes. This psychophysical concept has been extended to electrophysiological data. That is, the neural code for each basic taste is hypothesized to be coded by a dedicated channel of neurons (the "Labeled-Line" theory); i.e., one group of neurons signals "salty" and another separate group signals "sweet." Numerous psychophysical and electrophysiological findings, however, cannot be accomodated by this quadripartite theory, which limits taste to four basic qualities and four basic neuron types. Rather, the data described in this article suggest that the range of taste is more extensive than four or five basic tastes, and that this breadth of taste quality results initially from the activation of a broad array of ion channels, receptors, and second messengers associated with taste cell membranes. These findings have implications for neural organization and provide support for the "Across-Fiber Pattern" theory in which the neural code for taste is represented by the pattern of activity across all of the neurons, i.e., neurons are not exclusively labeled for a particular sensation but cooperate with the others in the ensemble to encode taste quality.     

Primate insular/opercular taste cortex: neuronal representations of the viscosity, fat texture, grittiness, temperature, and taste of foods

It is shown that the primate primary taste cortex represents not only taste but also information about many nontaste properties of oral stimuli. Of 1,122 macaque anterior insular/frontal opercular neurons recorded, 62 (5.5%) responded to oral stimuli. Of the orally responsive neurons, some (53%) represented the viscosity, tested using carboxymethyl-cellulose in the range 1-10,000 cP. Other neurons (8%) responded to fat in the mouth by encoding its texture (as shown similar responses to nonfat oils), and 8% responded to gritty texture. Some neurons (35%) responded to the temperature of the liquid in the mouth. Some neurons responded to capsaicin, and others to fatty acids. Some neurons (56%) had taste responses. Some (50%) of these neurons were unimodal, responding to one of these types of stimulus, and the majority combined responsiveness to these types of stimulus, with 23% responding for example to both taste and temperature. Some neurons respond to taste, texture, and temperature unimodally, but others combine these inputs. None of these orally responsive neurons responded to odor or to the sight of food. These results provide fundamental evidence about the information channels used to represent the taste, texture, and temperature of food in the first cortical area involved in taste in the primate brain. The results are relevant to understanding the physiological and pathophysiological processes related to how the properties of oral stimuli are represented in the brain and thus to the control of food intake and food selection.     

Neural population coding and auditory temporal pattern analysis

Over the 2 decades that have elapsed since Robert Erickson first published his pioneering work on across-fiber patterns in the gustatory system, the idea that information is represented by a population code has become almost universally accepted among neuroscientists. Although the concept of a population code is an implicit theoretical assumption underlying most of the work done in neuroscience today, the details of how population codes operate in specific systems remain unclear in many respects. This article reviews electrophysiological studies of the auditory system of echolocating bats that show that information about sound is initially represented across both space and time by relative amounts of activity in populations of excitatory and inhibitory neurons with different discharge patterns, different sensitivity functions, and different latencies. At the next level, each neuron in the auditory midbrain receives convergent input from a specific population of these lower brainstem neurons and acts as a "readout" of activity within this population. As a result, midbrain neurons become selectively tuned to stimulus features, for example, signal duration, to which neurons at lower levels respond indiscriminately. Intracellular recordings from auditory midbrain neurons show some of the mechanisms by which population input is processed. The known projection patterns of the midbrain "readout" neurons indicate that they, in turn, must become part of a new spatio-temporal population code that is transmitted to neurons at the thalamus, where additional forms of selectivity and patterns of output arise.     

Gustatory responses in the frontal opercular cortex of the alert cynomolgus monkey

The responses of 165 single taste neurons in the anterior operculum of the alert cynomolgus monkey were analyzed. Chemicals were deionized water, blackcurrant juice, and the four basic taste stimuli: glucose, NaCl, HCl, and quinine HCl. Taste-evoked responses could be recorded from an opercular region that measured approximately 4.0 mm in its anteroposterior extent, 2.0 mm mediolaterally, and 3.0 mm dorsoventrally. Within this area, taste-responsive neurons were sparsely distributed such that multiunit activity was rarely encountered and neuronal isolation was readily achieved. Intensity-response functions were determined for nine cells.In each case, the lowest concentration of the dynamic response range conformed well to human electrophysiological and psychophysical thresholds for the basic taste stimuli. There was some evidence of chemotopic organization. Cells that responded best to glucose tended to be distributed toward the anterior operculum, whereas most acid-sensitive neurons were located more posteriorly. The proportion of cells responding best to NaCl peaked in the middle of the area, whereas quinine sensitivity was rather evenly distributed throughout. Opercular neurons in the monkey showed moderate breadth of sensitivity compared with taste cells of other species and at other synaptic levels. A breadth-of-tuning coefficient was calculated for each neuron. This is a metric that can range from 0.0 for a cell that responds specifically to only one of the four basic stimuli to 1.0 for one that responds equally to all four stimuli. The mean coefficient for 165 cells in the operculum was 0.67 (range = 0.12-0.99). Efforts were made to determine whether neurons could be divided into a discrete number of types, as defined by their responsiveness to the stimulus array used here. It was concluded that most taste cells may be assigned to a small number of groups, each of which is statistically independent of the others, but within which the constituent neurons are not identical. An analysis of taste quality indicated that the sweet and salty stimuli evoked patterns of activity that were significantly intercorrelated. Similarly, patterns representing HCl, quinine HCl, and water were related.     

The neural code for taste in the brain stem: response profiles

In the study of the neural code for taste, two theories have dominated the literature: the across neuron pattern (ANP), and the labeled line theories. Both of these theories are based on the observations that taste cells are multisensitive across a variety of different taste stimuli. Given a fixed array of taste stimuli, a cell's particular set of sensitivities defines its response profile. The characteristics of response profiles are the basis of both major theories of coding. In reviewing the literature, it is apparent that response profiles are an expression of a complex interplay of excitatory and inhibitory inputs that derive from cells with a wide variety of sensitivity patterns. These observations suggest that, in the absence of inhibition, taste cells might be potentially responsive to all taste stimuli. Several studies also suggest that response profiles can be influenced by the taste context, defined as the taste stimulus presented just before or simultaneously with another, under which they are recorded. A theory, called dynamic coding, was proposed to account for context dependency of taste response profiles. In this theory, those cells that are unaffected by taste context would provide the signal, i.e., the information-containing portion of the ANP, and those cells whose responses are context dependent would provide noise, i.e., less stimulus specific information. When singular taste stimuli are presented, noise cells would provide amplification of the signal, and when complex mixtures are presented, the responses of the noise cells would be suppressed (depending on the particular combination of tastants), and the ratio of signal to noise would be enhanced.     

Cracking taste codes by tapping into sensory neuron impulse traffic

Insights into the biological basis for mammalian taste quality coding began with electrophysiological recordings from "taste" nerves and this technique continues to produce essential information today. Chorda tympani (geniculate ganglion) neurons, which are particularly involved in taste quality discrimination, are specialists or generalists. Specialists respond to stimuli characterized by a single taste quality as defined by behavioral cross-generalization in conditioned taste tests. Generalists respond to electrolytes that elicit multiple aversive qualities. Na(+)-salt (N) specialists in rodents and sweet-stimulus (S) specialists in multiple orders of mammals are well characterized. Specialists are associated with species' nutritional needs and their activation is known to be malleable by internal physiological conditions and contaminated external caloric sources. S specialists, associated with the heterodimeric G-protein coupled receptor T1R, and N specialists, associated with the epithelial sodium channel ENaC, are consistent with labeled line coding from taste bud to afferent neuron. Yet, S-specialist neurons and behavior are less specific than T1R2-3 in encompassing glutamate and E generalist neurons are much less specific than a candidate, PDK TRP channel, sour receptor in encompassing salts and bitter stimuli. Specialist labeled lines for nutrients and generalist patterns for aversive electrolytes may be transmitting taste information to the brain side by side. However, specific roles of generalists in taste quality coding may be resolved by selecting stimuli and stimulus levels found in natural situations. T2Rs, participating in reflexes via the glossopharynygeal nerve, became highly diversified in mammalian phylogenesis as they evolved to deal with dangerous substances within specific environmental niches. Establishing the information afferent neurons traffic to the brain about natural taste stimuli imbedded in dynamic complex mixtures will ultimately "crack taste codes."     

Artificial neural networks estimate the contribution of taste neurons to coding

Taste qualities are believed to be coded in the activity of populations of taste neurons. However, it is not clear whether all neurons are equally responsible for coding. To clarify the point the relative contribution of each taste neuron to coding was assessed by constructing simple three-layer neural networks with input neurons that represent cortical taste neurons of the rat. The networks were trained by the back-propagation learning algorithm to classify the neural response patterns to the basic taste stimuli (sucrose, HCl, quinine-hydrochloride, and NaCl). The networks had four output neurons representing the basic taste qualities, the values of which provide a measure for similarity of test stimuli to the basic taste stimuli. We estimated relative contributions of input neurons to the taste discrimination of the network by examining their significance S(j), which is defined as the sum of the absolute values of the connection weights from the jth input neuron to the hidden layer. When the input neurons with a smaller S(j) (e.g., 15 out of 39 input units) were "pruned" from the trained network, the ability of the network to discriminate the basic taste qualities was not greatly affected. On the other hand, the taste discrimination of the network progressively deteriorated much more rapidly with pruning of input neurons with a larger S(j). These results suggest that cortical taste neurons differentially contribute to the coding of taste qualities. Input neurons with a larger S(j) tended to be with a larger variation of neural discharge rates to the basic taste stimuli. The variation of neural discharges may be important in the coding of taste qualities.     

Gustatory responses in the nucleus tractus solitarius of the alert cynomolgus monkey

Multiunit and single neuron responses to taste stimuli in the nucleus tractus solitarius (NTS) of alert cynomolgus monkeys were analyzed. Intensity-response functions, including neural thresholds to glucose and quinine HCl, agreed well with psychophysical reports, implying that the cynomolgus monkey and human share the same dynamic range of sensitivity to prototypical taste stimuli. The NTS is chemotopically organized: neurons most responsive to HCl are more common in the posterior gustatory area, whereas those most responsive to glucose and NaCl are located in the anterior NTS. Responsiveness to quinine is more widely distributed but tends toward the anterior. Efforts were made to determine if neurons could be divided into a discrete number of types, as determined by their sensitivities to the prototypical stimuli. The clearest distinction was between those that did or did not respond well to HCl. Beyond this, neuronal categories were not obvious. Individual neurons were quite broadly sensitive to our stimulus array, so that, for the typical NTS cell, no one of the four prototypes evoked a majority of the discharges. This extreme breadth of tuning suggests that taste-quality information in the monkey might be incorporated in relative discharge rates across the neuron population. Correlations among patterns of activity to the four prototypical stimuli indicated that only HCl and quinine HCl have closely related taste qualities.     

Lateral geniculate neurons in behaving primates. II. Encoding of visual information in the temporal shape of the response.

1. We used the Karhunen-Loève (K-L) transform to quantify the temporal distribution of spikes in the responses of lateral geniculate (LGN) neurons. The basis functions of the K-L transform are a set of waveforms called principal components, which are extracted from the data set. The coefficients of the principal components are uncorrelated with each other and can be used to quantify individual responses. The shapes of each of the first three principal components were very similar across neurons. 2. The coefficient of the first principal component was highly correlated with the spike count, but the other coefficients were not. Thus the coefficient of the first principal component reflects the strength of the response, whereas the coefficients of the other principal components reflect aspects of the temporal distribution of spikes in the response that are uncorrelated with the strength of the response. Statistical analysis revealed that the coefficients of up to 10 principal components were driven by the stimuli. Therefore stimuli govern the temporal distribution as well as the number of spikes in the response. 3. Through the application of information theory, we were able to compare the amount of stimulus-related information carried by LGN neurons when two codes were assumed: first, a univariate code based on response strength alone; and second, a multivariate temporal code based on the coefficients of the first three principal components. We found that LGN neurons were able to transmit an average of 1.5 times as much information using the three-component temporal code as they could using the strength code. 4. The stimulus set we used allowed us to calculate the amount of information each neuron could transmit about stimulus luminance, pattern, and contrast. All neurons transmitted the greatest amount of information about stimulus luminance, but they also transmitted significant amounts of information about stimulus pattern. This pattern information was not a reflection of the luminance or contrast of the pixel centered on the receptive field. 5. In addition to measuring the average amount of information each neuron transmitted about all stimuli, we also measured the amount of information each neuron transmitted about the individual stimuli with both the univariate spike count code and the multivariate temporal code. We then compared the amount of information transmitted per stimulus with the magnitudes of the responses to the individual stimuli. We found that the magnitudes of both the univariate and the multivariate responses to individual stimuli were poorly correlated with the information transmitted about the individual stimuli.(ABSTRACT TRUNCATED AT 400 WORDS)     

Concept of neuron types in gustation in the rat.

1. In taste neurophysiology, from Pfaffmann's (49, 50) pioneering work until the present, the possibility of types of neurons corresponding in some sense with the "primary" taste qualities of Henning (33) has been entertained: recently types of gustatory neurons in peripheral nerves have been established according to which of the four classical stimuli is the "best stimulus." However, considerable variation occurs in the response profiles within neurons classified as belonging to the same type. The purpose of this research is to determine, using mathematical techniques where appropriate, if the within-type variation is spurious or, instead, indicates the absence of a typology of taste neurons. The data used were counts of the spike discharges of 50 individual taste neurons in the nucleus of the solitary tract of the rat, evoked by 32 diverse chemical stimuli. 2. Using as input the matrix of Pearson r correlation coefficients calculated for the responses of all pairings of neurons to all stimuli, multidimensional scaling analysis revealed a two-dimensional space in which no clear groupings of neurons occurred. 3. In a hierarchical cluster analysis of the neuron response profile similarities, no evidence of grouping was found, suggesting a more-or-less continuous variation among neurons. 4. When the organization of the 32 stimuli utilized was studied by the same techniques, no clear evidence for stimulus types was found, although the possibility of two stimulus types--"sweet" and "nonsweet"--was raised. 5. Construction of a joint neuron-stimulus space supported a spatial model of taste neuron-stimulus interaction, while analysis of the number and pattern of high correlations among neurons--even after allowance for attenuation due to measurement error--failed to support the notion of types of taste neurons with identical response profiles. 6. Aspects of the logical role of types of neurons in gustatory coding were discussed, and the results and methods of the present investigation were related to classification schemes for neurons in general. Suggestions for a formal taxonomy of neurons were given. 7. It should be emphasized that the present study and conclusions are of second-order, CNS neurons, whereas the studies advocating the presence of neurons types were of peripheral neurons. Taken together, the implication to be drawn from these studies is that if neural types do exist in peripheral taste nerves, the typology is lost at the first synapse and is thus unavailable to the CNS for coding purposes, at least in the rat.     

Issues of gustatory neural coding: where they stand today

The basic issues of gustatory neural coding are revisited. Questions addressed and conclusions drawn are: (1) what is the physical dimension across which gustatory neurons are sensitive, and upon which taste perceptions are based? The dimension that unites the various taste qualities is not physical, but physiological: a dimension of well-being, bounded by toxins at one extreme and nutrients at the other. (2) How broadly tuned are taste cells across the dimension? There are instances of specificity, but most mammalian taste cells respond to a range of qualities. (3) Are there basic taste qualities? Sweet, salty, sour, and bitter are widely accepted as basic tastes. Umami and starch tastes are considered basic by some. (4) Is taste topographically organized? There is some degree of physical separation among neurons most responsive to different taste qualities, but this does not appear to be sufficient precision to act as a meaningful coding mechanism. (5) Are there gustatory neuron types? Neurons, separated into categories according to their response profiles, respond as members of their category to the challenges of conditioned aversions and preferences, sodium deprivation, hyperglycemia, and receptor blockade, while cells from other categories react differently. This indicates the existence of functionally distinct types of taste cells. (6) Is the quality signal coded within the activity of the single most appropriate category of neurons, or is it carried by the pattern of response across neuronal categories? Both the breadth of responsiveness and the logical ambiguity of the signal in any one category of neurons argue that the taste message is carried by a pattern of activity across gustatory neuron types.     

Gustatory responses of cortical neurons in rats. I. Response characteristics

The responses of 111 cortical neurons to the four classical taste stimuli (sucrose, NaCl, HCl, and quinine HCl) applied to the anterior part of the tongue were recorded extracellularly in lightly anesthetized rats. Basic response properties of these cortical taste neurons were analyzed. The location of 88 of 111 neurons were histologically identified. They were distributed from anterodorsal to posteroventral direction in the insular cortex just dorsal to the rhinal sulcus and ventral to the somatic sensory area I. The receptive fields of 17 cortical neurons were examined. Most (94%) of the neurons had a narrow focus on the ipsilateral, contralateral, or bilateral sides of the tongue surface. Half of the foci were surrounded by a less-sensitive receptive field of relatively wide size. No apparent relationship was detected between the location of the cortical neurons and the site or extent of the receptive fields of those neurons, indicating a lack of topographical organization in the cortical gustatory area. The mean rate of the spontaneous discharges was 7.1 impulses/3 s, which is about 3 times larger than that in a first-order taste nerve (chorda tympani). The statistically significant difference of spontaneous discharges among response types of cortical neurons was observed only between the neurons responding in an excitatory manner to only one or two kinds of basic stimuli (6.2 impulses/3 s) and the neurons responding in an inhibitory manner to more than three kinds of taste stimuli (14.2 impulses/3 s). When the net responses (spontaneous rate subtracted) to each of the four tastes were compared with the spontaneous discharges in each neuron, the magnitude of spontaneous discharges was significantly negatively correlated with the net response to sucrose. This fact indicates that a neuron with a larger spontaneous discharge rate has a tendency to respond less to sucrose. Response characteristics of cortical taste neurons were quite distinct from those of the first-order taste neurons in the following respects: 1) a decrease in the average evoked discharge rate, which resulted in a small signal-to-noise ratio; 2) a tendency toward equalization of effectiveness of the four basic taste stimuli; 3) about 27% of the neurons decreased their firing rate during the first 3 s after the onset of taste stimulation; and 4) no clear initial phasic response, with a fluctuation in impulse discharges in some neurons.(ABSTRACT TRUNCATED AT 400 WORDS)     

Role of inferior temporal neurons in visual memory. I. Temporal encoding of information about visual images, recalled images, and behavioral context.

1. Lesions of the inferior temporal (IT) cortex selectively hamper monkeys in tasks requiring visual memory. A system that recognizes images must be able to encode a current stimulus, recall the code of a previous stimulus, compare the codes of the two stimuli, and make a decision on the basis of the outcome of the comparison. Therefore, IT neurons must be involved in at least one of these processes. To determine the specific role of IT neurons in visual memory, we measured the information conveyed in the neuronal responses about current patterns, recalled patterns, and behavioral context. 2. Two monkeys were trained to perform a sequential matching task using a set of 32 black and white Walsh patterns. In the course of an experiment, each pattern was presented repeatedly in sample, match, and nonmatch behavioral contexts. While the monkeys were performing the task, we recorded the activity of 76 neurons from area TE of IT. The neuronal responses to the stimuli were converted to spike density functions, and the resultant waveforms were quantified using their principal components. The relationships between the responses and the stimuli were studied using analysis of variance and information theory. 3. The analysis of variance was applied to the neuronal response waveforms using the context (sample, match, or nonmatch) and the patterns of the stimuli as independent variables and the spike count or the coefficients of the principal components as the dependent variables. We found that the waveforms of most neurons were significantly modulated by both the pattern and the context of the stimulus presentation. 4. We also analyzed the stimulus-response relationships using information theory. The input codes were based on the pattern and context of the stimuli, and the output codes were based on the spike count or the principal components of the responses. The neuronal response waveforms were found to convey significant amounts of information about both the pattern and context of the stimuli. Transmitted information was greatest when the response of a neuron was interpreted as a message about the combination of pattern and context. Nevertheless, there was information about context independent of pattern and vice versa. 5. We also used information theory to determine whether the neuronal responses to the second, or test, stimulus conveyed information about the pattern of the first, or sample, stimulus. The input codes were based on the patterns of the sample stimuli, and the output codes were based on the responses to the nonmatch test stimuli.(ABSTRACT TRUNCATED AT 400 WORDS)     

Temporal encoding of two-dimensional patterns by single units in primate primary visual cortex. II. Information transmission

1. Previously, we studied how picture information was processed by neurons in inferior temporal cortex. We found that responses varying in both response strength and temporal waveform carried information about briefly flashed stationary black-and-white patterns. Now, we have applied that same paradigm to the study of striate cortical neurons. 2. In this approach the responses to a set of basic black and white pictures were quantified through use of a set of basic waveforms, the principal components (extracted from all the responses of each neuron). We found that the first principal component, which corresponds to the response strength, and others, which correspond to different basic temporal activity patterns, were significantly related to the stimuli, i.e., the stimulus drove both the response strength and its temporal pattern. 3. Our previous study had shown that, when information theory was used to quantify the stimulus-response relation, inferior temporal neurons convey over twice as much information in a response code that includes temporal modulation as in a response code that includes only the response strength. This study shows that striate cortical neurons also carry twice as much information in a temporal code as in a response strength code. Thus single visual neurons at both ends of a cortical processing chain for visual pattern use a multidimensional temporal code to carry stimulus-related information. 4. These results support our multiplex-filter hypothesis, which states that single visual system neurons can be regarded as several simultaneously active parallel channels, each of which conveys independent information about the stimulus.     

Difference in taste quality coding between two cortical taste areas,granular and dysgranular insular areas,in rats

Summary The responses of 84 taste neurons to stimulation of the oral cavity in rats were examined; most taste neurons were found in either a granular insular area (area GI; n = 55) or dysgranular insular area (DI; n = 25), and the others (n = 4) were in an agranular insular area (area AI). The fraction of neurons responding to only one of the four basic stimuli was significantly larger in area GI than in area DI. When neurons were classified by the stimulus which most excited the neuron among the four basic stimuli, every best-stimulus category of neurons was found in both GI and DI areas. Quinine-best and multistimulus-type neurons, whose responses to some non-best stimulus exceeded 90% of the maximum, were more numerous in the cortex than in the thalamocortical relay neurons. When responses were plotted against taste stimuli arranged in the order of sucrose, NaCl, HCl, and quinine along the abscissa (taste coordinate), response profiles of taste neurons often showed two peaks. The double-peaked type of response profiles were found in every best-stimulus category of neurons in both areas; though, a significantly large fraction of quinine-best neurons in area GI were of the double-peaked type. Some taste neurons in area GI (n = 21) and in area DI (n = 7) were inhibited by one to two taste stimuli, particularly by the stimuli present next to the best one along the taste coordinate. In correlation profiles — correlation coefficients between sucrose and NaCl and between HCl and quinine — pairs of stimuli which were located next to each other on the taste coordinate were significantly smaller in area GI than in area DI. It is thus highly probable that area GI plays an important role in fine taste discrimination and area DI in integration of taste information.     

Patterns in the brain. Neuronal population coding in the somatosensory system

The aim of this article is to review some basic principles of neural coding, with an emphasis on mechanisms of stimulus representation in ensembles of neurons. The theory of "across-neuron response patterns" (ANRPs), first suggested by Thomas Young (1802) and fully developed by Robert Erickson (1963-2000), is summarized and applied to the problem of coding in primary afferent fibers and cortical neurons of the somatosensory system. The basic premise of the theory is that precise information about stimulus features cannot be encoded by single neurons, but is encoded by patterns of activity across populations of neurons. Different stimuli produce uniquely different patterns of ensemble activity (ANRPs)-discrimination between two stimuli is based on the absolute difference in total amount of activity (neural mass difference) of the ANRPs for those stimuli. Review of the literature shows that ANRPs and related population codes can accurately represent and differentiate among various stimulus parameters that cannot be distinguished by single neurons alone. Finally, the behavior of neuronal ensembles can be used to account for the sensory-perceptual changes associated with plasticity of thalamocortical circuits following selective sensorimotor deprivation or experience.     

Role of precise spike timing in coding of dynamic vibrissa stimuli in somatosensory thalamus

Rats discriminate texture by whisking their vibrissae across the surfaces of objects. This process induces corresponding vibrissa vibrations, which must be accurately represented by neurons in the somatosensory pathway. In this study, we investigated the neural code for vibrissa motion in the ventroposterior medial (VPm) nucleus of the thalamus by single-unit recording. We found that neurons conveyed a great deal of information (up to 77.9 bits/s) about vibrissa dynamics. The key was precise spike timing, which typically varied by less than a millisecond from trial to trial. The neural code was sparse, the average spike being remarkably informative (5.8 bits/spike). This implies that as few as four VPm spikes, coding independent information, might reliably differentiate between 10⁶ textures. To probe the mechanism of information transmission, we compared the role of time-varying firing rate to that of temporally correlated spike patterns in two ways: 93.9% of the information encoded by a neuron could be accounted for by a hypothetical neuron with the same time-dependent firing rate but no correlations between spikes; moreover, 93.4% of the information in the spike trains could be decoded even if temporal correlations were ignored. Taken together, these results suggest that the essence of the VPm code for vibrissa motion is firing rate modulation on a submillisecond timescale. The significance of such a code may be that it enables a small number of neurons, firing only few spikes, to convey distinctions between very many different textures to the barrel cortex.     

Convergence of sensory systems in the orbitofrontal cortex in primates and brain design for emotion

In primates, stimuli to sensory systems influence motivational and emotional behavior via neural relays to the orbitofrontal cortex. This article reviews studies on the effects of stimuli from multiple sensory modalities on the brain of humans and some other higher primates. The primate orbitofrontal cortex contains the secondary taste cortex, in which the reward value of taste is represented. It also contains the secondary and tertiary olfactory cortical areas, in which information about the identity and also about the reward value of odors is represented. A somatosensory input is revealed by neurons that respond to the viscosity of food in the mouth, to the texture (mouth feel) of fat in the mouth, and to the temperature of liquids placed into the mouth. The orbitofrontal cortex also receives information about the sight of objects from the temporal lobe cortical visual areas. Information about each of these modalities is represented separately by different neurons, but in addition, other neurons show convergence between different types of sensory input. This convergence occurs by associative learning between the visual or olfactory input and the taste. In that emotions can be defined as states elicited by reinforcers, the neurons that respond to primary reinforcers (such as taste and touch), as well as learn associations to visual and olfactory stimuli that become secondary reinforcers, provide a basis for understanding the functions of the orbitofrontal cortex in emotion. In complementary neuroimaging studies in humans, it is being found that areas of the orbitofrontal cortex are activated by pleasant touch, by painful touch, by taste, by smell, and by more abstract reinforcers such as winning or losing money. Damage to the orbitofrontal cortex in humans can impair the learning and reversal of stimulus-reinforcement associations and thus the correction of behavioral responses when these are no longer appropriate because previous reinforcement contingencies change. It is striking that humans and other catarrhines, being visual specialists like other anthropoids, interface the visual system to other sensory systems (e.g., taste and smell) in the orbitofrontal cortex.     

Decoding the activity of neuronal populations in macaque primary visual cortex

Visual function depends on the accuracy of signals carried by visual cortical neurons. Combining information across neurons should improve this accuracy because single neuron activity is variable. We examined the reliability of information inferred from populations of simultaneously recorded neurons in macaque primary visual cortex. We considered a decoding framework that computes the likelihood of visual stimuli from a pattern of population activity by linearly combining neuronal responses and tested this framework for orientation estimation and discrimination. We derived a simple parametric decoder assuming neuronal independence and a more sophisticated empirical decoder that learned the structure of the measured neuronal response distributions, including their correlated variability. The empirical decoder used the structure of these response distributions to perform better than its parametric variant, indicating that their structure contains critical information for sensory decoding. These results show how neuronal responses can best be used to inform perceptual decision-making.