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 neural coding in the cortex of the alert cynomolgus macaque: the quality of bitterness

We sought to define the gustatory neural representation in primates for stimuli that humans describe as predominantly bitter. Thus we analyzed the responses of single neurons from the insular cortex of two alert, male cynomolgus macaques in response to the oral application of four basic taste stimuli (glucose, NaCl, HCl, and quinine HCl) and fruit juice, and to a series of 15 other chemicals to which humans ascribe a bitter component. Gustatory neurons occupied a volume of 109 mm3 across an area of 4.0 mm in the anterposterior plane, 4.4 mm in the mediolateral, and 6.2 mm in the dorsoventral. Taste cells represented 161 (8.6%) of the 1881 neurons tested for chemical sensitivity. Fifty of these could be monitored throughout the delivery of the entire stimulus series, and their responses constitute the data of this study. The mean spontaneous discharge rate of the cortical gustatory cells was 3.2 +/- 3.3 spikes/s (range = 0.2-17.7 spikes/s). The mean breadth-of-tuning coefficient was a moderate 0.77 +/- 0.15 (range = 0.25-0.99). Forty-eight neurons responded to taste stimuli with excitation, and two responded with inhibition. Forty-one of the 50 neurons were able to be classified into one of four functional types based on their responses to the four basic stimuli used here. These were sugar (n = 22), salt (n = 7), acid (n = 7), and quinine (n = 5). A two-dimensional space was generated from correlations among the response profiles elicited by the stimuli array. The 16 bitter chemicals formed a coherent group that was most closely related to HCl, moderately to NaCl, and bore no relationship with glucose. Within the bitter stimuli, six formed a subgroup that was most separated from all nonbitter chemicals: quinine HCl, phenlythiocarbamide, propylthiouracil, caffeine, theophylline, and phenylalanine. Humans describe these stimuli as rather purely bitter. Of the remaining 10 bitter compounds, 4 were on the fringe of the bitter group leading to NaCl: MgCl2, CaCl2, NH4Cl, and arginine. Humans characterize these as bitter-salty. Three were on the fringe leading to HCl: urea, cysteine and vitamin B1. Humans call these bitter-sour. The remaining three (nicotine, histidine, and vitamin B2) occupied the center of the bitter group. Taste quality, inferred from the position of each stimulus in the space, correlated well with human descriptions of the same stimuli, reinforcing the value of the macaque as a neural model for human gustation.

Electrophysiological responses to bitter stimuli in primate cortex

Studies investigating fine details of gustatory coding in the domain of each basic taste quality have been completed for sweet, salt, and sour stimuli. In the present experiment, we used chemicals that humans describe as predominantly bitter. We recorded the activity of 50 taste neurons in insular cortex of two cynomolgus macaques. Stimuli were water, fruit juice, glucose, NaCl, HCl, and 16 bitter solutions. In a multidimensional taste space the 16 bitter stimuli formed a coherent cluster composed of three main subgroups: (1) QHCl, phenylalanine, theophylline, caffeine, propyl-thiouracil (PROP), and phenylthiocarbamide (PTC), all of which humans describe as rather purely bitter, (2) MgCl2, CaCl2, NH4Cl, and arginine, which humans describe as salty-bitter, and (3) urea, cysteine, and vitamin B1, which are described as sour-bitter. Vitamin B2, histidine and nicotine were in the center of the bitter cluster. Human descriptions of taste qualities conformed well to the presumed quality of each stimulus as inferred from its position in the multidimensional space (MDS), reinforcing the use of the macaque as a neural model for human gustation.

Preference conditioning alters taste responses in the nucleus of the solitary tract of the rat

Aversive conditioning has an impact on the neural signal for the gustatory conditioned stimulus (CS). Here, we determined whether the code is also affected by preference conditioning. We paired the taste of MgCl2 (CS+) with intragastric nutrients in some rats (MG), and citric acid (CS+) with nutrients in others (CI). A control group (Control) experienced both tastants without nutrients. Preferences (>90%) developed for each CS+. We recorded responses to 16 taste stimuli in the nucleus of the solitary tract. Responsiveness of acid-oriented neurons to MgCl2 in MG rats was lower than in Controls, and its profile was more distinct from those of acidic and bitter stimuli. Total activity to citric acid was unchanged in CI rats. However, its temporal profile showed a decreased phasic component, making citric acid temporally distinct from nonsugars. Therefore, the responses to both CS+ were modified, each in its own manner, to be more distinct from those of aversive stimuli. The effects of preference conditioning, however, were weaker than those of aversive conditioning.

Extinction of a conditioned taste aversion in rats: II. Neural effects in the nucleus of the solitary tract

The formation of a conditioned taste aversion (CTA) in rats results in neural changes at several levels of the gustatory system. In the nucleus of the solitary tract (NTS), the outstanding feature of the response to a CS is a brief burst of activity that is absent in unconditioned animals. The burst occurs about 1 s after stimulus onset and is seen only in neurons that respond well to sugars and the CS (0.0025 M NaSaccharin). We recorded single neuron activity in response to 12 stimuli from taste cells in the NTS of 8 rats, in which a CTA to NaSaccharin had been created and fully extinguished, and in 8 unconditioned controls. The issue was if the neural effects of the CTA in NTS were reversed with extinction. We recorded the activity of 41 neurons in controls and 55 in CTA-extinguished rats. Responses measured across all neurons were not significantly different in spontaneous activity, breadth of tuning, overall response magnitude to each of the 12 stimuli, relationship among stimuli in taste spaces, or time-course. However, cells in the sugar-sensitive subgroup showed a clear vestige of the conditioning experience. They gave a well-defined burst of activity to the CS, though of reduced amplitude and slightly longer latency than in fully conditioned rats. This burst was no longer associated with the conditioned behavior-which was fully extinguished-though it may be a permanent marker for the once-salient CS that can influence subsequent reacquisition of the aversion.

Extinction of a conditioned taste aversion in rats: I. Behavioral effects

The literature is divided over whether a conditioned taste aversion (CTA) can be fully extinguished. In Experiment 1, we created a powerful aversion in 54 rats by pairing the taste of 0.0025 M NaSaccharin (CS) with intraperitoneal injections of 127 mg/kg LiCl (US) on 3 occasions. We then offered 23-h deprived rats NaSaccharin for 10 min/day to observe the course of recovery. Extinction occurred in three phases: static, dynamic, and asymptotic. During the static phase (mean = 9.6 days), rats consumed the CS at < 10% of their preconditioned rate. With dynamic recovery (6.0 days), they increased acceptance to > 80% of preconditioning levels. Finally, they achieved asymptote (3.1 days) at 100% acceptance. In Experiment 2, we used 8 additional conditioned rats and 8 unconditioned controls. We followed the same 1-bottle extinction procedure and, again, obtained 100% acceptance. Then we offered both NaSaccharin and water for 8 days at 23 h/day and monitored lick patterns every 6 s to determine taste preferences. The conditioned animals consumed less NaSaccharin than controls on Day 1, and less NaSaccharin as a percentage of total fluid as late as Day 3. For the last 5 days of 2-bottle preference testing, there were no significant differences between the groups with regard to 1. volume of NaSaccharin or water consumed, 2. percentage of total fluid taken as NaSaccharin, 3. consumption of each fluid associated with a meal or taken spontaneously, 4. intake during the light or dark periods, or 5. the characteristics of ingestion, including number of drinking bouts, duration of bouts, number of licks/bout, and rate of licking. Therefore, a robust CTA is subject to complete behavioral extinction.

Activity in rat nucleus tractus solitarius after recovery from sodium deprivation

Sodium depletion has powerful effects on ingestive behavior. Depleted rats consume NaCl avidly at first, but decrease their intake to normal levels as they restore their sodium balance. However, vestiges of the depletion experience are expressed as a more rapidly induced and robust sodium consumption when the rat is challenged with a second depletion. Thus, the salience of sodium to the rat is modified in a lasting manner by severe deprivation. Sodium depletion also causes changes in the responses of taste cells in the nucleus tractus solitarius (NTS). In the present study, we examined whether gustatory-evoked responses in rat NTS continue to reflect the condition induced by sodium deprivation after sodium balance is restored. Single-unit recordings were made in response to 13 taste stimuli in two groups of rats: an experimental group that underwent 10-16 days of sodium deprivation followed by a 2-week recovery period, and a control group that never experienced deprivation. Experimental animals were tested for daily intake of 0.5 M NaCl before and after deprivation; they demonstrated a clear salt appetite only on the first day of the recovery period. Electrophysiological recordings revealed no significant differences between the two groups in response to any single stimulus. Neurons from each group of rats were categorized into three subtypes: sugar-sensitive, salt-sensitive, and nonsugar cells. A comparison of responses in these three subtypes offered no significant differences across groups. Thus, as rats restore depleted sodium levels following deprivation, the responsiveness of cells in the NTS also returns to a predeprivation state.

Taste responses in the nucleus of the solitary tract in saccharin-preferring and saccharin-averse rats

A minority of rats consistently reject the taste of sodium saccharin at concentrations that the majority find palatable. We chose rats that selected either water (WP), or 0.03 M NaSaccharin (SP) in two-bottle preference tests and monitored single unit responses to a range of taste qualities in the nucleus of the solitary tract. WP rats gave significantly greater responses to Na/Li salts and QHCl. Their responses to sugars were equal to those from SP rats. Total activity to NaSaccharin did not differ between the two groups, but its distribution across the three identified neuron types did. The response was skewed from one in which sugar (S) and sodium salt (N) participated nearly equally (SP) to one dominated by the activity of N cells and nearly devoid of an S cell contribution (WP rats). Accordingly, the response profile for NaSaccharin was correlated nearly as well with those of the sugars (+ 0.60) as with the Na/Li salts (+ 0.73) in SP rats, but was reshaped in WP rats to be nearly identical with those of the salts (+ 0.85) and unlike sugars (+ 0.30). In their heightened sensitivity to stimuli that humans call salty and bitter, and in their rejection of the complex taste of NaSaccharin, WP rats showed many of the characteristics of human tasters of PTC/PROP.

Pancreatic glucagon suppresses gustatory responsiveness to glucose

Peripheral administration of the gut peptide pancreatic glucagon (GGN) alters hepatic metabolism and suppresses feeding. Other physical (gastric distension) and chemical factors (hyperglycemia, hyperinsulinemia) that reduce food intake also suppress taste-evoked activity. This may attenuate the reinforcement derived from feeding and so promote termination of the meal. To determine whether this mechanism was operative with GGN administration, we studied the effect of hepatic portal infusions of 40 micrograms/kg pancreatic GGN on taste responses in the nucleus tractus solitarius of the rat. Taste activity was elicited by oral application of NaCl, glucose, HCl, and quinine HCl. Responses were monitored before and after injections of GGN or a control vehicle. Blood glucose levels were measured in separate groups of GGN- and vehicle-injected rats. Blood glucose increased significantly after GGN infusion and returned to control levels within 35 min. Taste responsiveness to glucose was significantly reduced after the GGN injection and recovered to preinjection levels by 36 min. Activity evoked by NaCl, HCl, and quinine HCl was unaffected. The suppression of responsiveness to sugars may reduce the hedonic appeal of tastants and so serve as a mechanism by which GGN could contribute to postprandial satiety.

Gustatory neural coding in the amygdala of the alert macaque monkey

1. Neurons in the amygdala are implicated in mediating hedonic appreciation, emotional expression, and conditioning, particularly as these relate to feeding. The amygdala receives projections from the primary taste cortex in monkeys, offering a route by which it could gain access to the gustatory information required to guide feeding behavior. We recorded the activity of 35 neurons in the amygdala of alert rhesus macaques in response to a range of gustatory intensities and qualities to characterize taste-evoked activity in this area. 2. The stimulus array comprised 26 chemicals, including four concentrations of each of the four basic taste stimuli, a series of other sugars, salts, and acids, monosodium glutamate, and orange juice. 3. Neurons responsive to taste stimulation could be found in a 76-mm3 region of the amygdala, centered 9.1 mm lateral to the midline, 14.9 mm anterior to the interaural line, and 25.7 mm below the surface of the dura. They composed 7.2% (35/484) of the cells tested for gustatory sensitivity in the amygdala. 4. The mean spontaneous activity of taste cells was 8.2 +/- 2.3 (SE) spikes per second. This rather high level provided an opportunity for reductions from spontaneous rate that was used regularly in the amygdala. When these negative response rates were included, the mean breadth-of-tuning coefficient of this sample of taste cells was 0.82. There was no strong evidence for gustatory neuron types, nor were functionally similar cells located together in a chemotopic arrangement. 5. Responses across 1.5 log units of stimulus concentration were nearly flat, with increasing excitation in some neurons largely offset by increasing inhibition in others. Taking the absolute value of the evoked activity, concentration-response functions rose monotonically to all basic stimuli except HCl, but were not sufficiently steep to account for human psychophysical data. The neural response to HCl did not rise with stimulus concentration within the range used. 6. Neural patterns representing the taste qualities of the basic stimuli were less sharply separated in the amygdala than at lower-order gustatory relays. Glucose elicited activity patterns that were most distinct from those of the nonsweet chemicals; those associated with NaCl were next most distinct. There was no clear separation between the patterns generated by chemicals that humans describe as sour and bitter. Monosodium glutamate evoked responses that did not correlate well with those of any basic stimulus, implying that its quality cannot be subsumed under the four basic tastes.(ABSTRACT TRUNCATED AT 400 WORDS)

Administration of satiety factors and gustatory responsiveness in the nucleus tractus solitarius of the rat

The administration of certain factors associated with postprandial satiety decreases gustatory responsiveness. We compared the effects of intravenous injections of glucose, insulin, pancreatic glucagon (PG), and cholecystokinin (CCK) on multiunit activity evoked from taste responsive neurons in the nucleus tractus solitarius of rats. Glucose, insulin, and PG reliably suppressed evoked responses to lingual application of 1.0M glucose, whereas responses that followed CCK remained unchanged. A common physiological consequence of glucose, insulin, and glucagon is increased glucose availability which may impact directly on gustatory neurons or indirectly through modifications in ventral forebrain or vagal afferent activity.

Polysaccharides as taste stimuli: their effect in the nucleus tractus solitarius of the rat

Rats show a pronounced preference for the tastes of starch-derived polysaccharides. Three of these compounds--Polycose, maltotriose and amylopectin--were used along with a standard array of chemicals in a study of their effectiveness as taste stimuli, as monitored by evoked single unit activity in the nucleus tractus solitarii (NTS). Maltotriose and amylopectin elicited very few spikes and no clear quality-related pattern of neural activity. Polycose, however, was an effective taste stimulus. It evoked an activity profile across neurons and over time that was poorly correlated with that of the prototypical sugar (sucrose) and only moderately related to those of the non-sugar prototypes (NaCl, HCl and quinine-HCl). The 14 cells (23%) that responded particularly well to Polycose were all members of neuronal subgroups that emphasized salt, acid and quinine sensitivity. Thus, despite the strong behavioral preference shown to Polycose, its neural profile is unlike those of other preferred stimuli. Polycose may represent a unique taste stimulus whose quality cannot be readily associated with those of the traditional 4 basic tastes.

The effect of amiloride on taste-evoked activity in the nucleus tractus solitarius of the rat

Amiloride is an inhibitor of passive sodium transport. Its application to taste receptors blocks inward sodium current, suppresses sodium-induced neural activity and reduces the perceived intensity of NaCl. We recorded taste-evoked responses of single neurons in the nucleus tractus solitarius (NTS) of the rat before and after the lingual application of amiloride to determine which neurons would be affected, the degree of the effect and the subsequent form of the neural code for sodium. Responses to all 7 stimuli that contained Na+ or Li+ were suppressed by amiloride. Activity evoked by the 8 other stimuli was unaltered. NTS neurons could be divided into 4 subsets according to their response profiles: Group 1 (salt-sugar), Group 2 (salt), Group 3 (salt-acid) and Group 4 (acid-salt-bitter). The entire effect of amiloride was discharged on cells in Groups 1 and 2; those in Groups 3 and 4 were unaffected. Following amiloride application, the neural code for sodium and lithium salts was highly similar to those for acids, bitter salts and quinine. Thus the activity of neurons in Groups 1 and 2 may be responsible for the distinction between 'saltiness' and sour-bitter tastes. The results imply that specific receptors are responsible for the recognition and transduction of sodium salts and that this specificity is maintained in the peripheral taste nerves to be manifested in the NTS.

Gustatory neural coding in the monkey cortex: stimulus intensity

1. We analyzed the activity of single neurons in gustatory cortex of alert cynomolgus monkeys in response to a range of stimulus intensities. Chemicals were deionized water, fruit juice, and several concentrations of the four prototypical taste stimuli: 10(-3)-1.0 M glucose, 10(-3)-1.0 M NaCl, 10(-4)-3 x 10(-2) M HCl, and 10(-5)-3 x 10(-3) M quinine HCl. 2. Taste-evoked responses could be recorded from a cortical gustatory area that measured 2.5 mm in its anteroposterior extent, 6.0 mm dorsoventrally, and 3.0 mm mediolaterally. Taste-responsive cells constituted 62 (3.7%) of the 1,661 neurons tested. Nongustatory cells gave responses associated with mouth movement (10.1%), somatosensory stimulation (2.2%), and approach or anticipation (0.9%). 3. Intensity-response functions were determined across 62 gustatory neurons. Neural thresholds for each stimulus quality conformed well to human psychophysical thresholds. Mean discharge rate was a direct function of stimulus concentration for glucose, NaCl, and quinine HCl. The most effective of the basic stimuli was glucose. 4. Power function exponents were calculated from the responses of neural subgroups most responsive to each basic stimulus. Those for glucose, NaCl, and quinine were within the range of psychophysically derived values. Thus the perceived intensity of each basic quality is presumably based on the activity of the appropriate neural subgroup rather than on the mean activity of all taste cells. 5. The mean breadth-of-tuning (entropy) coefficient for 62 gustatory neurons was 0.65 (range, 0.00-0.98). 6. There was no clear evidence of chemotopic organization in the gustatory cortex. 7. An analysis of taste quality indicated that sweet stimuli evoked patterns of activity that were clearly distinct from those of the nonsweet chemicals. Among the latter group, NaCl was differentiable from HCl and quinine HCl, whose patterns were closely related. 8. The response characteristics of cortical taste cells imply gustatory thresholds and intensity-response functions for the nonhuman primate that conform well to those reported in psychophysical studies of humans, reinforcing the value of this neural model for human taste intensity perception.

Coding channels in the taste system of the rat

Basic taste qualities are thought to be perceived independently, yet discrete neural coding channels have not been demonstrated in the central nervous system. The response profiles of taste cells in the nucleus tractus solitarius (NTS) of the rat were categorized into four groups, and the effects of amiloride, a passive sodium channel blocker, on each were determined. NTS neurons that responded specifically to sodium chloride (NaCl) or to NaCl and sugars were suppressed by amiloride; those broadly sensitive to salts, acids, and bitter stimuli were unaffected. Moreover, the response profile evoked by NaCl lost its distinctiveness after treatment with amiloride, becoming similar to those evoked by acids and quinine. Receptors that respond to sodium must relay their information through independent coding channels to identifiable subgroups of NTS neurons, the activity of which is responsible for the perception of saltiness.

Effect of cholecystokinin on taste responsiveness in rats

Alterations in taste responsiveness have been suggested to mediate the suppression of feeding that accompanies exogenous administration of cholecystokinin (CCK). We tested this possibility in electrophysiological and behavioral experiments. First we monitored taste-evoked activity in the nucleus tractus solitarii of anesthetized rats during intravenous injection of 2 or 6 micrograms/kg of CCK or vehicle. We found no significant effects on taste activity during the 30-min period after CCK administration. Then we employed a conditioned taste-aversion paradigm to measure the rat's perceived intensity of a series of glucose concentrations under the same three experimental conditions. At 2 micrograms/kg, CCK had no effect; at 6 micrograms/kg there was a significant increase in the perceived intensity of 2 of the 15 test solutions, which we attribute to elevated vagal tone. The potential contribution of gastric distension was eliminated in the electrophysiological study and was minimized by brief exposures to stimuli in the behavioral experiment. Thus CCK administration, in the absence of significant gastric distension, does not appear to alter taste responsiveness.

Intravenous insulin infusions in rats decrease gustatory-evoked responses to sugars

Physiological factors that affect food intake have been shown to influence taste-evoked activity in the rat's central nervous system. Insulin appears to have a bimodal effect on feeding, inhibiting intake when its rise is within the normal physiological range, but, with further increases, causing hyperphagia. We studied the effect of low intravenous doses (0.5 U/kg) of regular insulin on taste-evoked responses in the nucleus tractus solitarius. Taste activity was elicited by application to the tongue of glucose, fructose, NaCl, HCl, and quinine. We monitored responses before and after intrajugular injections of insulin or a control vehicle. Taste responsiveness to glucose and fructose was significantly reduced for the period 7-22 min following the injection. Activity representing NaCl, HCl, and quinine was unaffected. The suppression of responsiveness to sweet stimuli could decrease the hedonic appeal of tastants and so serve as a mechanism by which physiological doses of insulin could contribute to a reduction in feeding.

A measure of taste intensity discrimination in the rat through conditioned taste aversions

The ability of rats to make intensity discriminations was determined by forming a conditioned taste aversion to a moderate concentration of each of four basic taste stimuli, and then measuring the level of acceptance (number of licks during a 15 sec exposure) shown to a range of concentrations of the same chemical. Rats (N = 66) could discriminate between glucose concentrations that were separated by as little as 0.074 M, between NaCl concentrations that differed by 0.029 M, between HCl concentrations that were 9 X 10(-4) M apart, and between quinine HCl concentrations that differed by as little as 2.4 X 10(-6) M.

Blood glucose level affects perceived sweetness intensity in rats

Electrophysiological data indicate that hyperglycemia is associated with decreased neural taste responsiveness to 1.0 M glucose, but not to 0.01 M quinine HCl, in the rat's hindbrain. The present behavioral experiment was conducted to determine whether this suppression of neural activity is manifested in a reduced intensity perception to glucose, but not quinine. Each rat learned to avoid 1.0 M glucose through development of a conditioned taste aversion. Perceived intensity was then measured in control and in hyperglycemic rats by the extent to which they generalized to each test concentration of glucose. Experimental subjects treated moderate glucose concentrations (0.6-2.0 M) as if their intensity perceptions were reduced by 47%. This is consistent with the mean reduction of 43% in taste-evoked neural activity associated with hyperglycemia. A corresponding experiment gave no indication of a change in intensity perception to quinine as a function of hyperglycemia, again in accord with earlier electrophysiological results. We conclude that nutritional state may selectively affect gustatory sensitivity in the rat.

Blood glucose selectively affects taste-evoked activity in rat nucleus tractus solitarius

Gustatory afferent activity and the perceptions it evokes have proven to be modifiable by physiological needs. We monitored changes in taste responses in the rat nucleus tractus solitarius (NTS) as a function of blood glucose levels. Taste stimuli were 0.1 M NaCl, 1.0 M glucose, 0.03 M HCl and 0.01 M QHCl. Gustatory activity was selectively modified as an inverse function of blood glucose. Intravenous injections caused blood glucose levels to rise from 90 to near 200 mg% 5-15 min post-injection, with slow recovery thereafter. During the 5-15 min period, NTS responses to glucose were depressed an average 43%, NaCl by 20%, HCl by 16% and QHCl by 3%. Time courses of all responses were unmodified. This effect, probably mediated either by glucoreceptors in NTS or by vagal afferents from the gut, suggests a particular suppression of more appetitive tastes, and so may provide a neural concomitant to the decreased appeal of food which accompanies satiety.