- Email: [email protected]
Anion Size of Sodium Salts and Simple Taste Reaction Times
Physiology & Behavior, Vol. 66, No. 1, pp. 27–32, 1999 © 1999 Elsevier Science Inc. Printed in the USA. All rights reserved 0031-9384/99/$–see front matter
PII S0031-9384(98)00273-X
Anion Size of Sodium Salts and Simple Taste Reaction Times JEANNINE F. DELWICHE,*1 BRUCE P. HALPERN† AND JOHN A. DESIMONE‡ *Monell Chemical Senses Center, 3500 Market Street, Philadelphia, PA 19104-3308, †Department of Psychology and Section of Neurobiology and Behavior, Cornell University, Ithaca, NY 14853-7601, ‡Department of Physiology, Virginia Commonwealth University, P.O. Box 980551-MCV, Richmond, VA 23298-0551 Received 22 June 1998; Accepted 31 August 1998 DELWICHE, J. F., B. P. HALPERN AND J. A. DESIMONE. Anion size of sodium salts and simple taste reaction times. PHYSIOL BEHAV 66(1) 27–32, 1999.—Simple taste reaction times (RT) and taste intensities were measured in adult humans for 100-mM aqueous solutions of sodium chloride, acetate, glutamate, ascorbate, and gluconate flowed over the anterodorsal tongue with a closed liquid delivery system. Results from 12 subjects showed a significant increase in RT with molecular weight of the tastant, and a correlation of 0.941 between RT and the square roots of anionic weights. A multiple regression analysis controlling for perceived taste intensity indicated that RT increased linearly with the square root of the anionic weight. These findings support a model that includes both the permeability of ions through the tight junctions between the taste receptor cells of fungiform papillae taste buds and the effects of ions at apical portions of the receptor cells. They also suggest that gustatory transduction of sodium salts in humans normally involves intercellular spaces of taste buds as part of the functional sensory structures, in addition to individual taste receptor cells. © 1999 Elsevier Science Inc. Sodium salt
Reaction time
Intensity
Taste
Taste bud
for taste responses involving salts of differing anions (39,40) that can produce a directly testable quantitative prediction of human gustatory perception. A temporal aspect of this model, in relation to the molecular weight of tastants, is the subject of the present article. As do several earlier (10,18) and contemporary (11,30) models, the model developed by Ye et al. (39,40) includes the hypothesis that anions in solution with Na1 may affect taste responses through diffusion potentials across the paracellular shunt barrier in the tight junctions between taste bud receptor cells. Ye et al. (38–40) directly measured lingual diffusion potentials in laboratory rats while monitoring gustatory neural responses. Their data indicate that the paracellular shunt barrier in the tight junctions behaves as a weak cation exchange membrane, so the Na1 ion permeability exceeds that of the anions. Accordingly, electrodiffusion of Na-salts into intercellular regions evokes an electropositive potential outside the basolateral membranes of the taste cells. This would result in hyperpolarization of the taste receptors, and consequently, reduced neural response in the gustatory afferent neurons of the chorda tympani nerve. The larger (i.e., less mobile) the anion, the larger the hyperpolarizing potential and the smaller the neural response for a given Na1 concentration (38). Because
THE taste intensity and taste quality of aqueous solutions containing sodium ions (Na1) differ as a function of the anions that are present (11,32,34). These relationships could indicate that saltiness is primarily a function of the Na1 cation, while all anions modulate intensity or saltiness, with larger anions reducing saltiness more than smaller anions (2). However, the anions themselves can also serve as primary taste stimuli, as is the case for sodium saccharin which tastes sweet (26). The reduction of taste intensity, change in perceived taste quality, and active taste stimulation all have been reported as effects of anions in solution with Na1 (2,26), with increase in anion size being the determining factor for all these outcomes. Because of these apparently contradictory findings, it is difficult to develop explicit, quantitative, and testable predictions of sodium-associated “anion effects” if only psychophysical data are used. The qualitative models of “anion effects” for human taste perception of sodium salts derived solely from psychophysical data state that saltiness may decrease, or perhaps increase, as the size of an anion associated with Na1 becomes larger than Cl2, and that other taste qualities may appear and may become intense (2,26). Fortunately, gustatory data from electrophysiological studies of the anterior tongue of laboratory rats suggest a model 1To
whom requests for reprints should be addressed. E-mail: [email protected].
27
28 the cation exchanger property of the paracellular shunt barrier in the tight junctions appears to be imperfect, anions that are not very immobile will permit measurable transfer of Na1 ions into the intercellular regions. In the case of laboratory rats, Cl2 is such an anion (39,40). The increased cation concentration in the low-volume intercellular spaces can then directly interact with cation channels in the basolateral membranes of taste bud receptor cells. In this view (the cellular-paracellular model), the functional peripheral taste sensory unit becomes, at minimum, taste cell groups within a taste bud and the intercellular spaces that connect them. This would seem to conflict with earlier views of peripheral taste function that focused on the apical membrane of the individual taste receptor cell as the only region of the taste bud where contact with the stimulus occurred. For some classes of tastants, transduction events may indeed be limited to taste bud receptor cells’ apical membranes (22,23,25), as an individual gustatory receptor cell model would envision. However, the individual cell model [e.g., (24)] cannot easily accommodate the modulatory effects of various anions on sodium and potassium-salt gustatory responses. The latter model must posit a variety of specific anion-binding sites on the apical membranes of taste receptor cells that modulate the permeability of the cation through the taste cell apical membrane to account for the effects of anions on taste responses to salts (24). Empirical evidence in support of such anion binding sites on the apical membranes of taste cells remains lacking. In addition to providing a simple and empirically supported explanation of modulation of gustatory responses by anions (18,29,30,33,41), the cellular-paracellular model of taste responses to salts predicts that when cation and anion diffusion through the tight junctions is the predominant mode of stimulus presentation, consequences should be observed in the time course of the gustatory neural response. This condition was met for K1 salts for laboratory rats, and the time to maximum gustatory neural response in the chorda tympani nerve increased significantly with increasing anion size (40). This observation, combined with relations derivable from the cellular-paracellular model of gustatory responses to salts (39,40), allows several psychophysical predictions: (a) for a given small cation (e.g., Na1 or K1), perceived taste intensity will decrease as anion size increases; (b) for a given small cation, as anion size increases, the perceptual time course will be slowed. The latter prediction is the focus of the present report. The general prediction was that simple taste reactions times [i.e., the time intervals between the arrival at a gustatory receptor organ of any detectable change in taste, and an appropriate overt response made as quickly as possible by an organism to any change in taste (14)] in response to anterodorsal tongue stimulation with various sodium salts would increase with increasing molecular weight. A repeated-measures design assessing simple taste reaction times to NaCl, sodium acetate (Na-acetate), monosodium glutamate (MSG), sodium ascorbate (Na-ascorbate), and sodium gluconate (Na-gluconate), was employed. More specifically, the validity of a theoretically derived relationship between simple taste reaction times and the square root of anionic weight was investigated. Finally, because taste reaction times have been shown to decrease with increasing taste intensity or stimulus concentration (4,14, 15,17,35–37), and the intensities of the salts employed in the present study are known to decrease with increasing molecular weight (32,34), taste intensity data were collected so that the impact of taste intensity could be con-
DELWICHE ET AL. trolled statistically. A more detailed presentation is contained in a Ph.D. thesis (6), and brief reports have been made (7,8). MATERIALS AND METHODS
Subject Screening Before experimentation, a proposal of all testing procedures were submitted to, and approved by, the Cornell University Committee on Human Subjects. Thirty-nine nonsmoking paid volunteers were recruited from Cornell University’s Ithaca, NY campus (16 men, 23 women, 18–30 years old). None had participated in other taste experiments. They were instructed not to eat or drink anything other than water for the hour preceding screening, and were screened with a tip of the tongue procedure (9). Only those volunteers providing >4 conceptually similar taste quality descriptors for seven presentations of 25 mM sucrose, 2 mM HCl, 40 mM NaCl, and 8 mM quinine sulfate were eligible to become subjects (6,9). Thirteen of the subjects screened were found to be eligible (four men and nine women, 18–30 years old); all were recruited for participation in the main experiment, but one male subject (22 years old) was excused from the study after the training sessions (see below) because he was not able to position his tongue correctly. Procedure Subjects were instructed not to eat or drink anything other than water for at least an hour preceding a session. In each session, subjects were presented with 100 mM aqueous solutions of NaCl, Na-acetate, MSG, Na-ascorbate, and Na-gluconate, made from reagent grade compounds dissolved in water purified by a polished reverse osmosis method (conductivity ,1.5 microSiemens, refractive index 5 1.3330). No attempt was made to match the intensities of the different solutions given the large differences among individuals as to what constitutes equi-intensity solutions (Michael O’Mahony, 1995, personal communication). Each subject participated in a total of six sessions: three training and three experimental sessions. Within each session, each solution, plus water, was presented as a stimulus three times. The stimuli were presented in a unique restricted random order, with the five solutions not being allowed to occur three times in a row, the water presentations not being allowed to occur twice in a row, and a session never beginning with water. Subjects received identified water presentations at least three times before the data collection phase of a session began, with additional identified water presentations if requested. Stimuli were delivered via an automated closed-system flow machine (20), which controlled stimulus onset time and duration, avoided salivary dilution or vapor phase stimulation, and restricted solutions to z39 mm2 on the anterodorsal tip region of the tongue. Ten and a half seconds before a trial began, a 0.5-s warning tone was sounded, alerting the subject to get into the proper position. Each trial began with a 10-s water rinse, followed by a 1-s stimulus flow and then a 5-s water rinse. Between trials there was a 65.5 s rest. A double tone signaled the end of each trial and the beginning of the rest. Liquid flow rates were 9.0 to 11.6 mL s21. Simple taste reaction time was measured by having subjects release a telegraph key as soon as they detected a taste change (14,21). During the final experimental session, at the end of each trial subjects rated the total taste intensity of the stimulus solution. The identified water presentations that began this ses-
ANION SIZE
29
sion were defined as having zero intensity. Immediately following these identified water presentations and preceding the data collection portion of the session, 100 mM NaCl was presented and was defined as an example of a 10 in intensity. Subjects were told that if a subsequent presentation seemed to be higher in intensity than the example of 10 had seemed, they could give it a higher rating. Intensity ratings were reported during the 65.5-s rest. Subjects were required to use whole numbers in their ratings. Statistical Analysis Median reaction times were determined for each subject and salt (Table 1). These medians were converted to a rank order of salts for each subject [fastest reaction time 5 1; slowest, (5)] and tested for ranked tendency with the Page test (27). Means of the median reaction times were obtained for each solution (Table 1); the relationship between the molecular weights of the tastants and mean simple taste reaction times was plotted (Fig. 1), and the nonparametric correlation measure Spearman’s Rho (27) was calculated for this relationship. To evaluate a physicochemical model of the effects at mammalian anterodorsal tongue taste buds of the anionic component of tastants (see below), mean median simple taste reaction times were plotted against the square roots of the anionic weights of the tastants (Fig. 2), and the Pearson’s product-moment correlation was determined. To examine possible relationships between sodium salt type and reaction time independently from intensity, statistical control for a linear relationship between simple taste reaction time and judged taste intensity was provided by a multiple regression analysis in which simple reaction time was the dependent variable, and subject, intensity, and the square root of the anionic weight of each salt type were independent variables. An additional, statistically stricter multiple regression was also done, in which the dependent variable of reaction time was modeled by the independent variables of sub-
FIG. 1. Reaction time versus molecular weight.
ject, the square root of the anionic weight, intensity, the square root of intensity, and the log of the intensity. Theoretical Model A model was developed that predicted that taste reaction times would have a linear relationship to the square root of anionic weights (see equation 6). The underlying reasoning was that the higher a salt’s permeability, the shorter the time would be for the salt to reach the basolateral locus of action, and thus the shorter simple reaction times would be, as stated in equation 1: τ MA = α ⁄ P MA + τ i
where tMA is the reaction time for salt MA, PMA is the permeability constant of that salt, ti is the portion of the reaction time independent of the salt permeability, and a is a constant.
TABLE 1 INDIVIDUAL MEDIAN SIMPLE TASTE REACTION TIMES (IN SECONDS) 100 mM Aqueous Solutions Subject
NaCl
NaAce*
MSG
NaAsco†
NaGlu‡
1 2 3 4 5 6 7 8 9 10 11 12 Mean SD
0.638 0.483 0.623 0.550 0.571 0.451 0.711 0.709 0.562 0.672 0.684 0.522 0.598 0.088
0.903 0.609 1.193 0.630 0.619 0.544 0.831 1.170 0.488 1.932 0.607 0.629 0.846 0.414
1.081 1.217 1.009 0.695 0.674 0.765 1.277 1.540 0.535 1.689 0.720 0.992 1.016 0.362
1.120 1.038 1.137 0.592 0.679 0.598 0.911 2.167 0.512 2.203 0.668 0.692 1.026 0.581
1.829 2.052 1.259 0.753 1.034 0.551 2.020 0.915 0.707 2.499 0.752 0.677 1.254 0.667
*Na-acetate. †Na-ascorbate. ‡Na-gluconate.
(1)
FIG. 2. Reaction time versus square root of anionic weight.
30
DELWICHE ET AL.
The permeability constant, PMA, consists of two separable parts, the permeability constant of the cation, M, and the anion, A. The relationship between these permeability constants is shown in equation 2: 1 ⁄ P MA = 1 ⁄ 2 ( 1 ⁄ P M + 1 ⁄ P A )
(2)
Numerous studies on membrane permeation suggest that the permeability constants should vary with the inverse square root of the molecular weight of the diffusing species [e.g., (5)]. Thus, equation 2 can be reexpressed in terms of molecular weight: 1 ⁄ P MA = 1 ⁄ 2 ( aÎW M + bÎW A )
(3)
where a and b are constants that may depend on the particular ion, WM is the molecular weight of the cation, and WA is the molecular weight of the anion. By combining equations 1 and 3, equation 4, relating taste reaction time and molecular weight, can be written: τ MA = ( α ⁄ 2 ) [ aÎW M + bÎW A ] + τ i
(4)
Equation 4 can be made specific for a particular set of salts such as sodium salts: τ NaA = { ( α ⁄ 2 ) [ aÎW Na ] + τ i } + ( α ⁄ 2 )bÎW A
(5)
For a series of a particular salt, the first term of equation 5 can be treated as a constant. To further simplify the equation, it was assumed that b does not vary much with the anion, allowing the “(a/2), b” portion of the second term also to be treated as a constant. This allows equation 5 to be simplified: τ NaA = C 1 + C 2 ÎW A
(6)
where C1 and C2 are constants for sodium salts. Equation 6 presents a theory-derived testable prediction that, for sodium salts, simple taste reaction times will be linearly related to the square roots of the anion weights. RESULTS
Responses on Water Presentation Trials. All subjects responded to ,44% of trials in which water was presented as a stimulus, thus satisfying the preestablished criterion for data acceptance of ,50% responses on water trials. Reaction Times to Salts Median simple taste reaction times were determined for each subject and salt (Table 1). The rank orders of these reaction times increased as the molecular weights of the tastant became greater (p , 0.001, Page test). Means of the median reaction times (Table 1), when plotted against the molecular weight of each salt (Fig. 1), produced a relationship between the mean reaction times and molecular weights with a statistically significant Spearman’s Rho (rho 5 0.993, p , 0.001). As specified by the theoretical model for relationships between reaction times and anion weights of sodium salts that had been developed from physicochemical considerations (equation 6), simple taste reaction times were plotted against the square roots of the anionic weights (Fig. 2). The linear correlation coefficient between these reaction times and the square roots of the anionic weights was 0.941 (p , 0.01). A multiple regression analysis was performed with simple taste reaction time as the dependent variable, and subject, intensity, and the square root of the anionic weight of each salt
type as independent variables. The results indicated that there is a significant linear relationship between reaction time and the square root of the anionic weight (p , 0.035), as well as a significant linear relationship between reaction time and intensity (p , 0.001). Furthermore, with a multiple regression in which the dependent variable of reaction time was modeled by the independent variables of subject, the square root of the anionic weight, intensity, the square root of intensity, and the log of the intensity (which provides a stricter control for the influence of intensity on reaction times), a significant linear relationship (p , 0.055) between simple taste reaction time and the square root of the anionic weight of each salt type was found. DISCUSSION
The present data demonstrated a linear relationship between simple taste reaction time and the square root of the anionic weights. The observed direct relationship between molecular weight and simple taste reaction time was expected from previous studies [e.g., (4,14,15,17,35–37)]. However, the quantitative manner in which simple taste reaction time increased with the anionic weight of sodium salts was of substantial interest, because it could be especially compatible with some models of gustatory function (10,18,37,39,40). This relationship conformed to a physiological and physicochemical model that emphasized the permeability of anions and utilized standard results from the theory of cell membrane permeability [e.g., (5)] to suggest the nature of the specific connection between the molecular weight of the diffusing species and the speed of the behavioral responses. It would appear that the cellular-paracellular model of salt taste transduction mechanisms (10,11,18,29,39,40), based upon neurophysiological work in various rodents, can be generalized to humans. It is, therefore, likely that transduction of gustatory responses to salts normally involves taste buds as functional sensory structures, that transduction sites on both the apical membrane of taste bud receptor cells and on the basolateral membrane contribute, and that the basolateral sites are normally accessible only by diffusion through the tight junctions and into the paracellular space between cells (29,39,40). Molecular and anionic weight were used as approximations of molecular and anionic volume. Volume was of special importance because the reasoning underlying the tested model suggested that the square root of anionic molecular volume should directly relate to simple taste reaction time. This relationship was posited based upon previous nonhuman neurophysiological and biophysical data indicating that anions could act at mammalian fungiform papillae taste buds by diffusing through size-limited tight junctions that connected the receptor cells of taste buds (10,11,29,39,40). Threshold measurements for some of the same sodium salts indicate a different relationship between molecular weight and detection threshold (31). This is not surprising, because thresholds do not typically yield predictions of suprathreshold intensity (1,2), and reaction times near threshold are usually very long, frequently more than five times the reaction times for sodium chloride reported here (12,14,19). The correspondence between a model prompted in part by the gustatory neural data from a peripheral taste nerve of a strain of laboratory rat, and the human taste-dependent responses measured in the present study, suggests that a basic and meaningful common factor exists. This shared element may result from the way in which anions of monomolecular aqueous solutions distribute themselves on and between the
ANION SIZE
31
receptor cells of the taste buds found in fungiform papillae. A possible interpretation is that human simple taste reaction times relate quite directly to peripheral gustatory biophysical and neural events, with central nervous system (CNS) processing contributing little beyond that necessary to detect a change in taste and initiate a motor response. Another related, but somewhat different, interpretation is that CNS processing for simple taste reaction times may be similar for all tastants, allowing dissimilarities in peripheral events to become apparent. It may seem surprising to propose that the observed association between human simple taste reaction times and square roots of anionic molecular weights can be understood in terms of events at gustatory receptor organs. The substantial differences beyond the level of the hindbrain in the organization and connectivity of gustatory CNS between “higher primates,” and that of most other vertebrates that have been studied (3,28), might lead one to expect that CNS gustatory processing would have a sizable role in human taste-dependent behavior. In addition, a comparison and analysis of human versus Sprague–Dawley rat simple taste reaction times concluded that a substantially larger fraction of human taste reaction times could be ascribed to CNS events (14). In fact, a gross temporal difference between Sprague– Dawley rat and human taste reaction times is readily apparent: the absolute magnitude of human taste reaction times is much greater than that of Sprague–Dawley rats (13,14,16). Differences in the length of afferent and efferent conduction pathways could be a factor in this temporal disparity, but appear to account for only a small proportion of the much slower human times. It seems likely that, in adult humans, obligatory forebrain gustatory processing happens before any overt taste-dependent behavior can occur. This would result
in an addition of time to any human taste reaction interval. However, for simple taste reaction times, which presumably have minimal cognitive content, the temporal sequence of fastest and slowest could, nonetheless, be fully determined by the events at the level of the lingual taste bud population, perhaps with an equivalent duration of CNS processing across taste stimuli. Judged taste intensity decreases as the molecular weight of sodium salts increases (32,34), whereas taste reaction time increases under the same conditions. It could be the case that the present experiment’s observed change in simple taste reaction time with anionic weight was actually a restatement of the inverse relationship between taste intensity and the molecular weight of sodium salts. This possibility was examined by statistically separating reaction time from taste intensity. Although intensity differences did influence the simple reaction times, the analysis also indicated that when the contribution of taste intensity was removed, a significant relationship between simple taste reaction time and anionic molecular weight remained. Furthermore, previous studies found that judgments of taste intensity had very long reaction times (.1000 ms longer than simple taste reaction times), thus suggesting that simple taste reaction time and taste intensity are not determined by the same processes (14,15,19). In the present instance, the anionic component of sodium salts was seen to have an effect on the speed of taste responses beyond that which could be accounted for by intensity. ACKNOWLEDGEMENTS
We thank Paul A. S. Breslin, Michael G. Tordoff, and the anonymous reviewers for their thoughtful comments on earlier drafts of this article.
REFERENCES 1. Bartoshuk, L. M.: Is mixture suppression related to compression? Physiol. Behav. 14:643–649; 1975. 2. Bartoshuk, L. M.: Taste. In: Atkinson, R. C.; Herrnstein, R. J.; Lindzey, G.; Luce, R. D., eds. Steven’s handbook of experimental psychology. New York: Wiley; 1988:461–499. 3. Beckstead, R. M.; Morse, J. R.; Norgren, R.: The nucleus of the solitary tract in the monkey. Projections to the thalamus and brain stem nuclei. J. Comp. Neurol. 90:259–282; 1980. 4. Bujas, Z.: Le temps de réaction aux excitations gustatives d’intensité différente. Comp. Rend. Seances Soc. Biol. Filiales 119:1360– 1362; 1935. 5. Davson, H.: A textbook of general physiology, vol. 1, 4th ed. Baltimore: The Williams and Wilkins Company; 1970. 6. Delwiche, J. F.: Psychophysical evidence supporting a physiologically derived model of salt transduction mechanisms. Ph.D. Dissertation, Psychology, Cornell University, Ithaca, NY; 1997. 7. Delwiche, J. F.; Halpern, B. P.; DeSimone, J. A.: Anion size, perceived taste intensity and simple taste reaction times. Chem. Senses 23:213; 1998. 8. Delwiche, J. F.; Halpern, B. P.; DeSimone, J. A.; Lee, M. Y.: Anion size and simple taste reaction times in humans. Chem. Senses 21:595; 1996. 9. Delwiche, J. F.; Halpern, B. P.; Lee, M. Y.: A comparison of tip of the tongue and sip and spit screening procedures. Food Quality Pref. 7:293–297; 1996. 10. Elliot, E. J.; Simon, S. A.: The anion in salt taste: A possible role for paracellular pathways. Brain Res 535:9–17; 1990. 11. Erickson, R. P.; Schiffman, S. S.: Psychophysics: Insights into transduction mechanisms and neural codings. In: Simon, S. A.; Roper, S. D., eds., Mechanisms of taste transduction. Boca Raton, FL: CRC Press; 1993:395–424.
12. Hahn, H.; Gunther, H.: Über die Reize und die Reizbedingungen des Geschmackssinnes. Pflüg. Arch. Ges. Physiol. 231:48–67; 1933. 13. Halpern, B. P.: Time as a factor in gustation: Temporal patterns of taste stimulation and response. In: Pfaff, D. W., ed., Taste, olfaction, and the central nervous system. New York: Rockefeller University Press; 1985:181–209. 14. Halpern, B. P.: Constraints imposed on taste physiology by human taste reaction time data. Neurosci. Biobehav. Rev. 10:135– 152; 1986. 15. Halpern, B. P.: More than meets the tongue: Temporal characteristics of taste intensity and quality. In: Lawless, H. T.; Klein, B. P., eds., Sensory science theory and applications in foods. New York: Marcel Dekker; 1991;37–105. 16. Halpern, B. P.; Tapper, D. N.: Taste stimuli: Quality coding time. Science 171:1256–1258; 1971. 17. Hara, S.: Interrelationship among stimulus intensity, stimulated area and reaction time in the human gustatory sensation. Bull. Tokyo Med. Dental Univ. 2:147–158; 1955. 18. Harper, H. W.: A diffusion potential model of salt taste receptors. In: Roper, S. D.; Atema, J., eds. Annals of the New York Academy of Sciences. New York: Academy of Sciences; 1987:349– 351. 19. Kelling, S. K.; Halpern, B. P.: Taste judgments and gustatory stimulus duration: Taste quality, taste intensity, and reaction time. Chem. Senses 13:559–586; 1988. 20. Kelling, S. T.; Halpern, B. P.: The physical characteristics of open flow and closed flow taste delivery apparatus. Chem. Senses 11:89– 104; 1986. 21. Kelling, S. T.; Halpern, B. P.: Taste judgments and gustatory stimulus duration: Simple taste reaction times. Chem. Senses 12:543– 562; 1987.
32 22. Kinnamon, S. C.: Taste transduction: A diversity of mechanisms. Trends Neurosci. 11:491–496; 1988. 23. Kinnamon, S. C.: Taste transduction: Linkage between molecular mechanisms and psychophysics. Food Quality Pref. 7:153–159; 1996. 24. Kitada, Y.: Effect of anions on the responses to Ca, Mg, and Na in single water fibers of the frog glossopharyngeal nerve. In: Kurihara, K.; Suzuki, N.; Ogawa, H., eds., Olfaction and taste XI. Tokyo: Springer Verlag; 1994:110. 25. Kurihara, K.; Yoshii, K.; Kashiwayangi, M.: Transduction mechanisms in chemoreception. Comp. Biochem. Physiol. 85A:1–22; 1986. 26. Miller, I. J.; Bartoshuk, L. M.: Taste perception, taste bud distribution, and spatial relations. In: Getchell, T. V.; Doty, R. L.; Bartoshuk, L. M.; Snow, J. B., Jr., eds. Smell and taste in health and disease. New York: Raven Press; 1991:205–233. 27. O’Mahony, M.: Sensory evaluation of food. New York: Marcel Dekker, Inc.; 1986. 28. Pritchard, T. C.: The primate gustatory system. In: Getchell, T. V.; Bartoshuk, L. M.; Doty, R. L.; Snow, J. B., eds. Smell and taste in health and disease. New York: Raven Press; 1991:109–125. 29. Rehnberg, B. G.; MacKinnon, B. I.; Hettinger, T. P.; Frank, M. E.: Anion modulation of taste responses in sodium-sensitive neurons of the hamster chorda tympani nerve. J. Gen. Physiol. 101:453– 465; 1993. 30. Roper, S. D.: The microphysiology of peripheral taste organs. J. Neurosci. 12:1127–1134; 1992. 31. Schiffman, S. S.; Crumbliss, A. L.; Warwick, Z. S.; Graham, B. G.: Thresholds for sodium salts in young and elderly human subjects: Correlation with molar conductivity of anion. Chem. Senses 15:671–678; 1990.
DELWICHE ET AL. 32. Schiffman, S. S.; McElroy, A. E.; Erickson, R. P.: The range of taste quality of sodium salts. Physiol. Behav. 24:217–224; 1980. 33. Simon, S. A.: Influence of tight junctions on the interaction of salts with lingual epithelia: Responses of chorda tympani and lingual nerves. Mol. Cell. Biochem. 114:43–48; 1992. 34. Van der Klaauw, N. J.; Smith, D. V.: Taste quality profiles of fifteen organic and inorganic salts. Physiol. Behav. 58:295–306; 1995. 35. Yamamoto, T.; Kato, T.; Matsuo, R.; Araie, N.; Azuma, S.; Kawamura, Y.: Gustatory reaction time under variable stimulus parameters in human adults. Physiol. Behav. 29:79–84; 1982. 36. Yamamoto, T.; Kawamura, Y.: Gustatory reaction time in human adults. Physiol. Behav. 26:715–719; 1981. 37. Yamamoto, T.; Kawamura, Y.: Gustatory reaction time to various salt solutions in human adults. Physiol. Behav. 32:49–53; 1984. 38. Ye, Q.; Heck, G. L.; DeSimone, J. A.: The anion paradox in sodium taste reception: Resolution by voltage-clamp studies. Science 254:724–726; 1991. 39. Ye, Q.; Heck, G. L.; DeSimone, J. A.: Voltage dependence of the rat chorda tympani response to Na1 salts: Implications for the functional organization of taste receptor cells. J Neurophysiol 70:167–178; 1993. 40. Ye, Q.; Heck, G. L.; DeSimone, J. A.: Effects of voltage perturbation of the lingual receptive field on chorda tympani responses to Na1 and K1 salts in the rat: Implications for gustatory transduction. J. Gen. Physiol. 104:885–907; 1994. 41. Yoshie, S.; Wakasugi, C.; Kanazawa, H.; Fugita, T.: Receptosecretory nature of the gustatory cell. In: Kurihara, K.; Suzuki, N.; Ogawa, H., eds, Olfaction and taste XI. Tokyo: Springer Verlag; 1994:5–8.
PII S0031-9384(98)00273-X
Anion Size of Sodium Salts and Simple Taste Reaction Times JEANNINE F. DELWICHE,*1 BRUCE P. HALPERN† AND JOHN A. DESIMONE‡ *Monell Chemical Senses Center, 3500 Market Street, Philadelphia, PA 19104-3308, †Department of Psychology and Section of Neurobiology and Behavior, Cornell University, Ithaca, NY 14853-7601, ‡Department of Physiology, Virginia Commonwealth University, P.O. Box 980551-MCV, Richmond, VA 23298-0551 Received 22 June 1998; Accepted 31 August 1998 DELWICHE, J. F., B. P. HALPERN AND J. A. DESIMONE. Anion size of sodium salts and simple taste reaction times. PHYSIOL BEHAV 66(1) 27–32, 1999.—Simple taste reaction times (RT) and taste intensities were measured in adult humans for 100-mM aqueous solutions of sodium chloride, acetate, glutamate, ascorbate, and gluconate flowed over the anterodorsal tongue with a closed liquid delivery system. Results from 12 subjects showed a significant increase in RT with molecular weight of the tastant, and a correlation of 0.941 between RT and the square roots of anionic weights. A multiple regression analysis controlling for perceived taste intensity indicated that RT increased linearly with the square root of the anionic weight. These findings support a model that includes both the permeability of ions through the tight junctions between the taste receptor cells of fungiform papillae taste buds and the effects of ions at apical portions of the receptor cells. They also suggest that gustatory transduction of sodium salts in humans normally involves intercellular spaces of taste buds as part of the functional sensory structures, in addition to individual taste receptor cells. © 1999 Elsevier Science Inc. Sodium salt
Reaction time
Intensity
Taste
Taste bud
for taste responses involving salts of differing anions (39,40) that can produce a directly testable quantitative prediction of human gustatory perception. A temporal aspect of this model, in relation to the molecular weight of tastants, is the subject of the present article. As do several earlier (10,18) and contemporary (11,30) models, the model developed by Ye et al. (39,40) includes the hypothesis that anions in solution with Na1 may affect taste responses through diffusion potentials across the paracellular shunt barrier in the tight junctions between taste bud receptor cells. Ye et al. (38–40) directly measured lingual diffusion potentials in laboratory rats while monitoring gustatory neural responses. Their data indicate that the paracellular shunt barrier in the tight junctions behaves as a weak cation exchange membrane, so the Na1 ion permeability exceeds that of the anions. Accordingly, electrodiffusion of Na-salts into intercellular regions evokes an electropositive potential outside the basolateral membranes of the taste cells. This would result in hyperpolarization of the taste receptors, and consequently, reduced neural response in the gustatory afferent neurons of the chorda tympani nerve. The larger (i.e., less mobile) the anion, the larger the hyperpolarizing potential and the smaller the neural response for a given Na1 concentration (38). Because
THE taste intensity and taste quality of aqueous solutions containing sodium ions (Na1) differ as a function of the anions that are present (11,32,34). These relationships could indicate that saltiness is primarily a function of the Na1 cation, while all anions modulate intensity or saltiness, with larger anions reducing saltiness more than smaller anions (2). However, the anions themselves can also serve as primary taste stimuli, as is the case for sodium saccharin which tastes sweet (26). The reduction of taste intensity, change in perceived taste quality, and active taste stimulation all have been reported as effects of anions in solution with Na1 (2,26), with increase in anion size being the determining factor for all these outcomes. Because of these apparently contradictory findings, it is difficult to develop explicit, quantitative, and testable predictions of sodium-associated “anion effects” if only psychophysical data are used. The qualitative models of “anion effects” for human taste perception of sodium salts derived solely from psychophysical data state that saltiness may decrease, or perhaps increase, as the size of an anion associated with Na1 becomes larger than Cl2, and that other taste qualities may appear and may become intense (2,26). Fortunately, gustatory data from electrophysiological studies of the anterior tongue of laboratory rats suggest a model 1To
whom requests for reprints should be addressed. E-mail: [email protected].
27
28 the cation exchanger property of the paracellular shunt barrier in the tight junctions appears to be imperfect, anions that are not very immobile will permit measurable transfer of Na1 ions into the intercellular regions. In the case of laboratory rats, Cl2 is such an anion (39,40). The increased cation concentration in the low-volume intercellular spaces can then directly interact with cation channels in the basolateral membranes of taste bud receptor cells. In this view (the cellular-paracellular model), the functional peripheral taste sensory unit becomes, at minimum, taste cell groups within a taste bud and the intercellular spaces that connect them. This would seem to conflict with earlier views of peripheral taste function that focused on the apical membrane of the individual taste receptor cell as the only region of the taste bud where contact with the stimulus occurred. For some classes of tastants, transduction events may indeed be limited to taste bud receptor cells’ apical membranes (22,23,25), as an individual gustatory receptor cell model would envision. However, the individual cell model [e.g., (24)] cannot easily accommodate the modulatory effects of various anions on sodium and potassium-salt gustatory responses. The latter model must posit a variety of specific anion-binding sites on the apical membranes of taste receptor cells that modulate the permeability of the cation through the taste cell apical membrane to account for the effects of anions on taste responses to salts (24). Empirical evidence in support of such anion binding sites on the apical membranes of taste cells remains lacking. In addition to providing a simple and empirically supported explanation of modulation of gustatory responses by anions (18,29,30,33,41), the cellular-paracellular model of taste responses to salts predicts that when cation and anion diffusion through the tight junctions is the predominant mode of stimulus presentation, consequences should be observed in the time course of the gustatory neural response. This condition was met for K1 salts for laboratory rats, and the time to maximum gustatory neural response in the chorda tympani nerve increased significantly with increasing anion size (40). This observation, combined with relations derivable from the cellular-paracellular model of gustatory responses to salts (39,40), allows several psychophysical predictions: (a) for a given small cation (e.g., Na1 or K1), perceived taste intensity will decrease as anion size increases; (b) for a given small cation, as anion size increases, the perceptual time course will be slowed. The latter prediction is the focus of the present report. The general prediction was that simple taste reactions times [i.e., the time intervals between the arrival at a gustatory receptor organ of any detectable change in taste, and an appropriate overt response made as quickly as possible by an organism to any change in taste (14)] in response to anterodorsal tongue stimulation with various sodium salts would increase with increasing molecular weight. A repeated-measures design assessing simple taste reaction times to NaCl, sodium acetate (Na-acetate), monosodium glutamate (MSG), sodium ascorbate (Na-ascorbate), and sodium gluconate (Na-gluconate), was employed. More specifically, the validity of a theoretically derived relationship between simple taste reaction times and the square root of anionic weight was investigated. Finally, because taste reaction times have been shown to decrease with increasing taste intensity or stimulus concentration (4,14, 15,17,35–37), and the intensities of the salts employed in the present study are known to decrease with increasing molecular weight (32,34), taste intensity data were collected so that the impact of taste intensity could be con-
DELWICHE ET AL. trolled statistically. A more detailed presentation is contained in a Ph.D. thesis (6), and brief reports have been made (7,8). MATERIALS AND METHODS
Subject Screening Before experimentation, a proposal of all testing procedures were submitted to, and approved by, the Cornell University Committee on Human Subjects. Thirty-nine nonsmoking paid volunteers were recruited from Cornell University’s Ithaca, NY campus (16 men, 23 women, 18–30 years old). None had participated in other taste experiments. They were instructed not to eat or drink anything other than water for the hour preceding screening, and were screened with a tip of the tongue procedure (9). Only those volunteers providing >4 conceptually similar taste quality descriptors for seven presentations of 25 mM sucrose, 2 mM HCl, 40 mM NaCl, and 8 mM quinine sulfate were eligible to become subjects (6,9). Thirteen of the subjects screened were found to be eligible (four men and nine women, 18–30 years old); all were recruited for participation in the main experiment, but one male subject (22 years old) was excused from the study after the training sessions (see below) because he was not able to position his tongue correctly. Procedure Subjects were instructed not to eat or drink anything other than water for at least an hour preceding a session. In each session, subjects were presented with 100 mM aqueous solutions of NaCl, Na-acetate, MSG, Na-ascorbate, and Na-gluconate, made from reagent grade compounds dissolved in water purified by a polished reverse osmosis method (conductivity ,1.5 microSiemens, refractive index 5 1.3330). No attempt was made to match the intensities of the different solutions given the large differences among individuals as to what constitutes equi-intensity solutions (Michael O’Mahony, 1995, personal communication). Each subject participated in a total of six sessions: three training and three experimental sessions. Within each session, each solution, plus water, was presented as a stimulus three times. The stimuli were presented in a unique restricted random order, with the five solutions not being allowed to occur three times in a row, the water presentations not being allowed to occur twice in a row, and a session never beginning with water. Subjects received identified water presentations at least three times before the data collection phase of a session began, with additional identified water presentations if requested. Stimuli were delivered via an automated closed-system flow machine (20), which controlled stimulus onset time and duration, avoided salivary dilution or vapor phase stimulation, and restricted solutions to z39 mm2 on the anterodorsal tip region of the tongue. Ten and a half seconds before a trial began, a 0.5-s warning tone was sounded, alerting the subject to get into the proper position. Each trial began with a 10-s water rinse, followed by a 1-s stimulus flow and then a 5-s water rinse. Between trials there was a 65.5 s rest. A double tone signaled the end of each trial and the beginning of the rest. Liquid flow rates were 9.0 to 11.6 mL s21. Simple taste reaction time was measured by having subjects release a telegraph key as soon as they detected a taste change (14,21). During the final experimental session, at the end of each trial subjects rated the total taste intensity of the stimulus solution. The identified water presentations that began this ses-
ANION SIZE
29
sion were defined as having zero intensity. Immediately following these identified water presentations and preceding the data collection portion of the session, 100 mM NaCl was presented and was defined as an example of a 10 in intensity. Subjects were told that if a subsequent presentation seemed to be higher in intensity than the example of 10 had seemed, they could give it a higher rating. Intensity ratings were reported during the 65.5-s rest. Subjects were required to use whole numbers in their ratings. Statistical Analysis Median reaction times were determined for each subject and salt (Table 1). These medians were converted to a rank order of salts for each subject [fastest reaction time 5 1; slowest, (5)] and tested for ranked tendency with the Page test (27). Means of the median reaction times were obtained for each solution (Table 1); the relationship between the molecular weights of the tastants and mean simple taste reaction times was plotted (Fig. 1), and the nonparametric correlation measure Spearman’s Rho (27) was calculated for this relationship. To evaluate a physicochemical model of the effects at mammalian anterodorsal tongue taste buds of the anionic component of tastants (see below), mean median simple taste reaction times were plotted against the square roots of the anionic weights of the tastants (Fig. 2), and the Pearson’s product-moment correlation was determined. To examine possible relationships between sodium salt type and reaction time independently from intensity, statistical control for a linear relationship between simple taste reaction time and judged taste intensity was provided by a multiple regression analysis in which simple reaction time was the dependent variable, and subject, intensity, and the square root of the anionic weight of each salt type were independent variables. An additional, statistically stricter multiple regression was also done, in which the dependent variable of reaction time was modeled by the independent variables of sub-
FIG. 1. Reaction time versus molecular weight.
ject, the square root of the anionic weight, intensity, the square root of intensity, and the log of the intensity. Theoretical Model A model was developed that predicted that taste reaction times would have a linear relationship to the square root of anionic weights (see equation 6). The underlying reasoning was that the higher a salt’s permeability, the shorter the time would be for the salt to reach the basolateral locus of action, and thus the shorter simple reaction times would be, as stated in equation 1: τ MA = α ⁄ P MA + τ i
where tMA is the reaction time for salt MA, PMA is the permeability constant of that salt, ti is the portion of the reaction time independent of the salt permeability, and a is a constant.
TABLE 1 INDIVIDUAL MEDIAN SIMPLE TASTE REACTION TIMES (IN SECONDS) 100 mM Aqueous Solutions Subject
NaCl
NaAce*
MSG
NaAsco†
NaGlu‡
1 2 3 4 5 6 7 8 9 10 11 12 Mean SD
0.638 0.483 0.623 0.550 0.571 0.451 0.711 0.709 0.562 0.672 0.684 0.522 0.598 0.088
0.903 0.609 1.193 0.630 0.619 0.544 0.831 1.170 0.488 1.932 0.607 0.629 0.846 0.414
1.081 1.217 1.009 0.695 0.674 0.765 1.277 1.540 0.535 1.689 0.720 0.992 1.016 0.362
1.120 1.038 1.137 0.592 0.679 0.598 0.911 2.167 0.512 2.203 0.668 0.692 1.026 0.581
1.829 2.052 1.259 0.753 1.034 0.551 2.020 0.915 0.707 2.499 0.752 0.677 1.254 0.667
*Na-acetate. †Na-ascorbate. ‡Na-gluconate.
(1)
FIG. 2. Reaction time versus square root of anionic weight.
30
DELWICHE ET AL.
The permeability constant, PMA, consists of two separable parts, the permeability constant of the cation, M, and the anion, A. The relationship between these permeability constants is shown in equation 2: 1 ⁄ P MA = 1 ⁄ 2 ( 1 ⁄ P M + 1 ⁄ P A )
(2)
Numerous studies on membrane permeation suggest that the permeability constants should vary with the inverse square root of the molecular weight of the diffusing species [e.g., (5)]. Thus, equation 2 can be reexpressed in terms of molecular weight: 1 ⁄ P MA = 1 ⁄ 2 ( aÎW M + bÎW A )
(3)
where a and b are constants that may depend on the particular ion, WM is the molecular weight of the cation, and WA is the molecular weight of the anion. By combining equations 1 and 3, equation 4, relating taste reaction time and molecular weight, can be written: τ MA = ( α ⁄ 2 ) [ aÎW M + bÎW A ] + τ i
(4)
Equation 4 can be made specific for a particular set of salts such as sodium salts: τ NaA = { ( α ⁄ 2 ) [ aÎW Na ] + τ i } + ( α ⁄ 2 )bÎW A
(5)
For a series of a particular salt, the first term of equation 5 can be treated as a constant. To further simplify the equation, it was assumed that b does not vary much with the anion, allowing the “(a/2), b” portion of the second term also to be treated as a constant. This allows equation 5 to be simplified: τ NaA = C 1 + C 2 ÎW A
(6)
where C1 and C2 are constants for sodium salts. Equation 6 presents a theory-derived testable prediction that, for sodium salts, simple taste reaction times will be linearly related to the square roots of the anion weights. RESULTS
Responses on Water Presentation Trials. All subjects responded to ,44% of trials in which water was presented as a stimulus, thus satisfying the preestablished criterion for data acceptance of ,50% responses on water trials. Reaction Times to Salts Median simple taste reaction times were determined for each subject and salt (Table 1). The rank orders of these reaction times increased as the molecular weights of the tastant became greater (p , 0.001, Page test). Means of the median reaction times (Table 1), when plotted against the molecular weight of each salt (Fig. 1), produced a relationship between the mean reaction times and molecular weights with a statistically significant Spearman’s Rho (rho 5 0.993, p , 0.001). As specified by the theoretical model for relationships between reaction times and anion weights of sodium salts that had been developed from physicochemical considerations (equation 6), simple taste reaction times were plotted against the square roots of the anionic weights (Fig. 2). The linear correlation coefficient between these reaction times and the square roots of the anionic weights was 0.941 (p , 0.01). A multiple regression analysis was performed with simple taste reaction time as the dependent variable, and subject, intensity, and the square root of the anionic weight of each salt
type as independent variables. The results indicated that there is a significant linear relationship between reaction time and the square root of the anionic weight (p , 0.035), as well as a significant linear relationship between reaction time and intensity (p , 0.001). Furthermore, with a multiple regression in which the dependent variable of reaction time was modeled by the independent variables of subject, the square root of the anionic weight, intensity, the square root of intensity, and the log of the intensity (which provides a stricter control for the influence of intensity on reaction times), a significant linear relationship (p , 0.055) between simple taste reaction time and the square root of the anionic weight of each salt type was found. DISCUSSION
The present data demonstrated a linear relationship between simple taste reaction time and the square root of the anionic weights. The observed direct relationship between molecular weight and simple taste reaction time was expected from previous studies [e.g., (4,14,15,17,35–37)]. However, the quantitative manner in which simple taste reaction time increased with the anionic weight of sodium salts was of substantial interest, because it could be especially compatible with some models of gustatory function (10,18,37,39,40). This relationship conformed to a physiological and physicochemical model that emphasized the permeability of anions and utilized standard results from the theory of cell membrane permeability [e.g., (5)] to suggest the nature of the specific connection between the molecular weight of the diffusing species and the speed of the behavioral responses. It would appear that the cellular-paracellular model of salt taste transduction mechanisms (10,11,18,29,39,40), based upon neurophysiological work in various rodents, can be generalized to humans. It is, therefore, likely that transduction of gustatory responses to salts normally involves taste buds as functional sensory structures, that transduction sites on both the apical membrane of taste bud receptor cells and on the basolateral membrane contribute, and that the basolateral sites are normally accessible only by diffusion through the tight junctions and into the paracellular space between cells (29,39,40). Molecular and anionic weight were used as approximations of molecular and anionic volume. Volume was of special importance because the reasoning underlying the tested model suggested that the square root of anionic molecular volume should directly relate to simple taste reaction time. This relationship was posited based upon previous nonhuman neurophysiological and biophysical data indicating that anions could act at mammalian fungiform papillae taste buds by diffusing through size-limited tight junctions that connected the receptor cells of taste buds (10,11,29,39,40). Threshold measurements for some of the same sodium salts indicate a different relationship between molecular weight and detection threshold (31). This is not surprising, because thresholds do not typically yield predictions of suprathreshold intensity (1,2), and reaction times near threshold are usually very long, frequently more than five times the reaction times for sodium chloride reported here (12,14,19). The correspondence between a model prompted in part by the gustatory neural data from a peripheral taste nerve of a strain of laboratory rat, and the human taste-dependent responses measured in the present study, suggests that a basic and meaningful common factor exists. This shared element may result from the way in which anions of monomolecular aqueous solutions distribute themselves on and between the
ANION SIZE
31
receptor cells of the taste buds found in fungiform papillae. A possible interpretation is that human simple taste reaction times relate quite directly to peripheral gustatory biophysical and neural events, with central nervous system (CNS) processing contributing little beyond that necessary to detect a change in taste and initiate a motor response. Another related, but somewhat different, interpretation is that CNS processing for simple taste reaction times may be similar for all tastants, allowing dissimilarities in peripheral events to become apparent. It may seem surprising to propose that the observed association between human simple taste reaction times and square roots of anionic molecular weights can be understood in terms of events at gustatory receptor organs. The substantial differences beyond the level of the hindbrain in the organization and connectivity of gustatory CNS between “higher primates,” and that of most other vertebrates that have been studied (3,28), might lead one to expect that CNS gustatory processing would have a sizable role in human taste-dependent behavior. In addition, a comparison and analysis of human versus Sprague–Dawley rat simple taste reaction times concluded that a substantially larger fraction of human taste reaction times could be ascribed to CNS events (14). In fact, a gross temporal difference between Sprague– Dawley rat and human taste reaction times is readily apparent: the absolute magnitude of human taste reaction times is much greater than that of Sprague–Dawley rats (13,14,16). Differences in the length of afferent and efferent conduction pathways could be a factor in this temporal disparity, but appear to account for only a small proportion of the much slower human times. It seems likely that, in adult humans, obligatory forebrain gustatory processing happens before any overt taste-dependent behavior can occur. This would result
in an addition of time to any human taste reaction interval. However, for simple taste reaction times, which presumably have minimal cognitive content, the temporal sequence of fastest and slowest could, nonetheless, be fully determined by the events at the level of the lingual taste bud population, perhaps with an equivalent duration of CNS processing across taste stimuli. Judged taste intensity decreases as the molecular weight of sodium salts increases (32,34), whereas taste reaction time increases under the same conditions. It could be the case that the present experiment’s observed change in simple taste reaction time with anionic weight was actually a restatement of the inverse relationship between taste intensity and the molecular weight of sodium salts. This possibility was examined by statistically separating reaction time from taste intensity. Although intensity differences did influence the simple reaction times, the analysis also indicated that when the contribution of taste intensity was removed, a significant relationship between simple taste reaction time and anionic molecular weight remained. Furthermore, previous studies found that judgments of taste intensity had very long reaction times (.1000 ms longer than simple taste reaction times), thus suggesting that simple taste reaction time and taste intensity are not determined by the same processes (14,15,19). In the present instance, the anionic component of sodium salts was seen to have an effect on the speed of taste responses beyond that which could be accounted for by intensity. ACKNOWLEDGEMENTS
We thank Paul A. S. Breslin, Michael G. Tordoff, and the anonymous reviewers for their thoughtful comments on earlier drafts of this article.
REFERENCES 1. Bartoshuk, L. M.: Is mixture suppression related to compression? Physiol. Behav. 14:643–649; 1975. 2. Bartoshuk, L. M.: Taste. In: Atkinson, R. C.; Herrnstein, R. J.; Lindzey, G.; Luce, R. D., eds. Steven’s handbook of experimental psychology. New York: Wiley; 1988:461–499. 3. Beckstead, R. M.; Morse, J. R.; Norgren, R.: The nucleus of the solitary tract in the monkey. Projections to the thalamus and brain stem nuclei. J. Comp. Neurol. 90:259–282; 1980. 4. Bujas, Z.: Le temps de réaction aux excitations gustatives d’intensité différente. Comp. Rend. Seances Soc. Biol. Filiales 119:1360– 1362; 1935. 5. Davson, H.: A textbook of general physiology, vol. 1, 4th ed. Baltimore: The Williams and Wilkins Company; 1970. 6. Delwiche, J. F.: Psychophysical evidence supporting a physiologically derived model of salt transduction mechanisms. Ph.D. Dissertation, Psychology, Cornell University, Ithaca, NY; 1997. 7. Delwiche, J. F.; Halpern, B. P.; DeSimone, J. A.: Anion size, perceived taste intensity and simple taste reaction times. Chem. Senses 23:213; 1998. 8. Delwiche, J. F.; Halpern, B. P.; DeSimone, J. A.; Lee, M. Y.: Anion size and simple taste reaction times in humans. Chem. Senses 21:595; 1996. 9. Delwiche, J. F.; Halpern, B. P.; Lee, M. Y.: A comparison of tip of the tongue and sip and spit screening procedures. Food Quality Pref. 7:293–297; 1996. 10. Elliot, E. J.; Simon, S. A.: The anion in salt taste: A possible role for paracellular pathways. Brain Res 535:9–17; 1990. 11. Erickson, R. P.; Schiffman, S. S.: Psychophysics: Insights into transduction mechanisms and neural codings. In: Simon, S. A.; Roper, S. D., eds., Mechanisms of taste transduction. Boca Raton, FL: CRC Press; 1993:395–424.
12. Hahn, H.; Gunther, H.: Über die Reize und die Reizbedingungen des Geschmackssinnes. Pflüg. Arch. Ges. Physiol. 231:48–67; 1933. 13. Halpern, B. P.: Time as a factor in gustation: Temporal patterns of taste stimulation and response. In: Pfaff, D. W., ed., Taste, olfaction, and the central nervous system. New York: Rockefeller University Press; 1985:181–209. 14. Halpern, B. P.: Constraints imposed on taste physiology by human taste reaction time data. Neurosci. Biobehav. Rev. 10:135– 152; 1986. 15. Halpern, B. P.: More than meets the tongue: Temporal characteristics of taste intensity and quality. In: Lawless, H. T.; Klein, B. P., eds., Sensory science theory and applications in foods. New York: Marcel Dekker; 1991;37–105. 16. Halpern, B. P.; Tapper, D. N.: Taste stimuli: Quality coding time. Science 171:1256–1258; 1971. 17. Hara, S.: Interrelationship among stimulus intensity, stimulated area and reaction time in the human gustatory sensation. Bull. Tokyo Med. Dental Univ. 2:147–158; 1955. 18. Harper, H. W.: A diffusion potential model of salt taste receptors. In: Roper, S. D.; Atema, J., eds. Annals of the New York Academy of Sciences. New York: Academy of Sciences; 1987:349– 351. 19. Kelling, S. K.; Halpern, B. P.: Taste judgments and gustatory stimulus duration: Taste quality, taste intensity, and reaction time. Chem. Senses 13:559–586; 1988. 20. Kelling, S. T.; Halpern, B. P.: The physical characteristics of open flow and closed flow taste delivery apparatus. Chem. Senses 11:89– 104; 1986. 21. Kelling, S. T.; Halpern, B. P.: Taste judgments and gustatory stimulus duration: Simple taste reaction times. Chem. Senses 12:543– 562; 1987.
32 22. Kinnamon, S. C.: Taste transduction: A diversity of mechanisms. Trends Neurosci. 11:491–496; 1988. 23. Kinnamon, S. C.: Taste transduction: Linkage between molecular mechanisms and psychophysics. Food Quality Pref. 7:153–159; 1996. 24. Kitada, Y.: Effect of anions on the responses to Ca, Mg, and Na in single water fibers of the frog glossopharyngeal nerve. In: Kurihara, K.; Suzuki, N.; Ogawa, H., eds., Olfaction and taste XI. Tokyo: Springer Verlag; 1994:110. 25. Kurihara, K.; Yoshii, K.; Kashiwayangi, M.: Transduction mechanisms in chemoreception. Comp. Biochem. Physiol. 85A:1–22; 1986. 26. Miller, I. J.; Bartoshuk, L. M.: Taste perception, taste bud distribution, and spatial relations. In: Getchell, T. V.; Doty, R. L.; Bartoshuk, L. M.; Snow, J. B., Jr., eds. Smell and taste in health and disease. New York: Raven Press; 1991:205–233. 27. O’Mahony, M.: Sensory evaluation of food. New York: Marcel Dekker, Inc.; 1986. 28. Pritchard, T. C.: The primate gustatory system. In: Getchell, T. V.; Bartoshuk, L. M.; Doty, R. L.; Snow, J. B., eds. Smell and taste in health and disease. New York: Raven Press; 1991:109–125. 29. Rehnberg, B. G.; MacKinnon, B. I.; Hettinger, T. P.; Frank, M. E.: Anion modulation of taste responses in sodium-sensitive neurons of the hamster chorda tympani nerve. J. Gen. Physiol. 101:453– 465; 1993. 30. Roper, S. D.: The microphysiology of peripheral taste organs. J. Neurosci. 12:1127–1134; 1992. 31. Schiffman, S. S.; Crumbliss, A. L.; Warwick, Z. S.; Graham, B. G.: Thresholds for sodium salts in young and elderly human subjects: Correlation with molar conductivity of anion. Chem. Senses 15:671–678; 1990.
DELWICHE ET AL. 32. Schiffman, S. S.; McElroy, A. E.; Erickson, R. P.: The range of taste quality of sodium salts. Physiol. Behav. 24:217–224; 1980. 33. Simon, S. A.: Influence of tight junctions on the interaction of salts with lingual epithelia: Responses of chorda tympani and lingual nerves. Mol. Cell. Biochem. 114:43–48; 1992. 34. Van der Klaauw, N. J.; Smith, D. V.: Taste quality profiles of fifteen organic and inorganic salts. Physiol. Behav. 58:295–306; 1995. 35. Yamamoto, T.; Kato, T.; Matsuo, R.; Araie, N.; Azuma, S.; Kawamura, Y.: Gustatory reaction time under variable stimulus parameters in human adults. Physiol. Behav. 29:79–84; 1982. 36. Yamamoto, T.; Kawamura, Y.: Gustatory reaction time in human adults. Physiol. Behav. 26:715–719; 1981. 37. Yamamoto, T.; Kawamura, Y.: Gustatory reaction time to various salt solutions in human adults. Physiol. Behav. 32:49–53; 1984. 38. Ye, Q.; Heck, G. L.; DeSimone, J. A.: The anion paradox in sodium taste reception: Resolution by voltage-clamp studies. Science 254:724–726; 1991. 39. Ye, Q.; Heck, G. L.; DeSimone, J. A.: Voltage dependence of the rat chorda tympani response to Na1 salts: Implications for the functional organization of taste receptor cells. J Neurophysiol 70:167–178; 1993. 40. Ye, Q.; Heck, G. L.; DeSimone, J. A.: Effects of voltage perturbation of the lingual receptive field on chorda tympani responses to Na1 and K1 salts in the rat: Implications for gustatory transduction. J. Gen. Physiol. 104:885–907; 1994. 41. Yoshie, S.; Wakasugi, C.; Kanazawa, H.; Fugita, T.: Receptosecretory nature of the gustatory cell. In: Kurihara, K.; Suzuki, N.; Ogawa, H., eds, Olfaction and taste XI. Tokyo: Springer Verlag; 1994:5–8.