Physiology& Behavior,Vol. 52, pp. 535-540, 1992

0031-9384/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd.

Printed in the USA.

Is Starch Flavor Unitary? Evidence From Studies of Cooked Starch ISRAEL

RAMIREZ

Monell Chemical Senses Center, Philadelphia, PA 19104-3308 R e c e i v e d 22 N o v e m b e r

1991

RAMIREZ, I. Is starch flavor unitary? Evidence from studies of cooked starch. PHYSIOL BEHAV 52(3) 535-540, 1992.Although starch is the world's most abundant nutritive carbohydrate, its sensory properties are not as well understood as those of sugars. Previous researchers have assumed that all starches have the same flavor. The present experiments examined flavor differences among starches. Untrained rats were offered a choice of suspensions containing raw versus cooked starch. For some starches (potato and rice), rats strongly preferred cooked over uncooked starch. For other starches (regular corn, corn amylopectin, and wheat), rats showed little or no preference for cooked over uncooked starch. In order to determine whether the greater preference for cooked starch reflects a difference in flavor intensity, rats were conditioned to avoid potato or corn amylopectin starches by pairing ingestion of these substances with lithium chloride injections. Rats trained to avoid raw starch also avoided cooked starch, indicating that cooked and raw starch have similar flavors. However, when these trained rats were offered a choice between cooked and raw starch, they avoided the raw starch; this result is inconsistent with the assumption that cooking enhances the intensity of starch flavor. Similar results were obtained with corn amylopectin and potato starch, even though these starches differ greatly with regard to the effects of cooking on preference in untrained rats. However, rats trained to avoid potato starch avoided this starch to a greater degree than did rats trained to avoid corn amylopectin; conversely, rats trained to avoid corn amylopectin avoided this starch to a greater degree than did rats trained to avoid potato starch. Therefore, the flavor of starch is complex; there are specific flavor notes related to species and cooking. Nevertheless, the various types of starches share at least one flavor note in common. Flavor

Taste

Preference

Carbohydrates

Starch

A L T H O U G H several different classes of carbohydrates are abundant in natural foods, the h u m a n chemosensory system seems to recognize only one class of carbohydrate--sugars. Furthermore, chemically diverse simple sugars seem to evoke only one kind of sensation--sweet taste (3). For many years, it had been assumed that other m a m m a l s share this blindness to most carbohydrates (17). Recent studies have made this view untenable (14,18-23,25,26). Rats' detection thresholds for carbohydrates that are insipid to humans, such as malto-oligosaccharides, cellulose, and starch, are all below 0.1% (18,22,23,27). In addition to being able to detect low concentrations, rats can discriminate between sucrose, malto-oligosaccharide, cellulose, and starch (8,19,22,23,29); these substances seem to have qualitatively different flavors to rats. However, it is not clear whether rats can discriminate among members of different subclasses within each of these categories. Many different forms of starch exist that differ with regard to their physical and chemical attributes. Rats have a limited ability to discriminate between two different varieties of corn starch, but this is due to the presence of an off-taste c o m p o n e n t in amylomaize starch (20). The present report examines the ability of rats to discriminate among starches that differ with regard to species and heat treatment (cooking). Plants store starch as water-insoluble granules (2,4,5). Because the granules are fairly compact, most of the surface area of the starch molecules is buried inside the granules, and is, therefore,

unavailable to oral chemoreceptors. When a mixture of starch and water is heated to 70-90°C, the granules break up completely and the starch forms a solution or gel (2,5). Therefore, cooking greatly increases the surface area of starch molecules that is potentially available to oral chemoreceptors. This is why cooked starch is much more susceptible to enzymatic degradation than is raw starch (11,18). If chemoreception for starch is mediated by interactions between oral chemoreceptors with starch molecules, then cooking starch should greatly increase the intensity of starch flavor, just as cooking greatly increases susceptibility to amylase digestion. Alternatively, ifchemoreception for starch is mediated by tactile perception of granules (i.e., roughness), cooking starch should attenuate or abolish starch flavor. A single experiment did not detect large changes in preference resulting from cooking (18), but the possibility of more subtle effects has not been ruled out. Another factor that could influence starch flavor is the species of origin. Starches from different species differ with regard to diverse characteristics including granule size, susceptibility to amylase digestion, presence of phosphate groups, etc. (1,7,10,12,30). Although starches from different plants share some chemosensory similarities (19), rats prefer some starches over others (18,20,28). Simple preference tests in untrained rats do not tell us whether these differences in starch preference are due to quantitative differences (flavor intensity), or to qualitative differences (flavor quality).

535

RAMIREZ

536 The starches used in the present work were chosen to represent as diverse a sample as possible. Corn starch was used because it served as a typical starch in earlier studies (18,19). Corn amylopectin (waxy maze starch) and Lintner potato starch were used extensively because they appear to have minimal offtastes (1,7,20). Cooked starches may become stale upon standing at room temperature; this is known as retrogradation (6,13). Corn amylopectin and potato starch are less susceptible to retrogradation than most other starches (6,13). GENERAL METHOD

Animals A total of 32 female 7-week-old CD rats (Crl:CD®BR, Charles River Breeding Labs., Wilmington, MA) were used. The rats were housed individually in hanging stainless steel cages, maintained on a 12/12 hour light/dark cycle at 23-27°C, and given food ad lib. The food used, Purina Laboratory Chow (#5001), contains about 49% carbohydrate by weight, most of which is starch. Before the experiment began, the rats were given tap water in sipper bottles to drink ad lib. The rats were maintained in the laboratory for at least 1 week after arrival to permit them to acclimate to their surroundings before testing.

General Procedure The procedures are similar to those used in previous experiments [see (18) for additional details]. Fluids were offered in 120 ml glass jars placed inside the front of the cages. Food was in steel hoppers in the rear of the cages. All mixtures were formulated on a weight basis because intake was measured by weighing. Deionized water was used to make up suspensions. In most tests, rats were offered a choice of carbohydrate suspended in water versus another mixture. The position of the carbohydrate-containing mixture was reversed (left-right) in successive tests. In order to prevent precipitation of starch, 0.3% xanthan gum (16) (Sigma Chemical Co., St. Louis, MO) was added to fluids containing starch; the control fluid in choice tests also contained xanthan gum. For Experiment 1, preference tests lasted 24 h. For experiment 2, preference tests lasted for 30 rain during the first hour of the dark part of the light/dark cycle. Carbohydrate types. Regular corn, corn amylopectin (waxy maize), regular potato, Lintner potato (so called soluble starch), wheat, and rice starch were used. All carbohydrates were obtained from the Sigma Chemical Co. (St. Louis, MO), except for the regular corn starch which came from Teklad (Madison, WI). Lintner starch has been treated with acid to reduce the viscosity of the cooked starch solution (31). Except where otherwise stated, the concentration of starch was always 1%. Starch was cooked by heating it in water with a microwave oven. Cooking was interrupted to stir and resuspend the starch as needed. Heating was stopped when the starch granules appeared to dissolve in a hazy solution (this was usually at 8590°C). The mixture was brought to room temperature by placing in a water bath. Xanthan gum was added after the mixture cooled. For both cooked and uncooked starches, the mixture was stirred for at least 30 min after the addition ofxanthan gum.

Statistics Preference was computed as intake of the carbohydrate suspension divided by total fluid intake. A value of 0.5 indicates no preference and 1.0 indicates complete preference for carbohydrate suspension. Simple t-tests were used to determine

whether the preference statistic differed from 0.5. The results of these tests are given in the figures. Between-group comparisons were done by analysis of variance; significant overall F values were followed up by appropriate t-tests to determine which groups differed from one another. Results are always given as mean _+ SE. EXPERIMENT IA Experiment I consists of three subexperiments in which rats were given a choice between raw and cooked versions of the same starch. Several different starch types were used in order to determine which features of starch correlated with the degree of preference for cooked starch. In Experiment 1A, two starches that are resistant to retrogradation (precipitation) were compared to conventional corn starch.

Animals Eleven rats weighing 181 _+ 2 g were used.

Procedure The rats were given a choice of cooked versus raw amylopectin, Lintner potato, and regular corn starch on 3 successive days. They were given a 3-day rest, and retested again in the same sequence. The results of the two preference tests were averaged, since they did not differ from each other.

Results For all three starches, rats preferred cooked over raw starch (see Fig. 1). However, they showed a greater preference for cooked Lintner potato versus raw Lintner potato than for the other starches, F(2, 20) = 8.7, p < 0.002, individual t-tests p < 0.02. Total fluid intake was greater when rats were offered amylopectin (38 _+ 3 g/day) than when they were offered the other two starches [both were 32 + 2 g/day, F(2, 20) = 5.5, p < 0.02]. EXPERIMENT IB In Experiment 1A rats showed a greater preference for cooked over raw starch when Lintner-treated potato starch was used then for two other starches. Lintner treatment involves partial acid hydrolysis of starch (31). The reduced viscosity of cooked Lintner as opposed to regular starch might have influenced preference. Alternatively, species differences in starch type (corn versus potato) could have been involved. This experiment provided a more direct test by comparing Lintner potato starch to untreated potato starch.

Animals The same eleven rats from Experiment 1A were used.

Procedure The procedure is similar to that of Experiment 1A. The rats were offered a choice of raw versus cooked regular potato starch, and a choice of raw versus cooked Lintner starch on successive days. The rats were given a 1-day rest and then retested in the same order.

Results Preference for cooked starch was not affected by Lintner treatment [see Fig. 2, F(1, 10) = 0.0008, NS]. Total fluid intake

COOKED STARCH @ O C

537

was the same for both starch types [33 _+ 2 g/day, F ( I , 10) = O.4].

1.00

L

Q NO

0.80

a

EXPERIMENT IC Since Lintner t r e a t m e n t did n o t affect preference, it seemed likely that species differences in starch types m a y influence preference for cooked starch. This e x p e r i m e n t e x a m i n e d starches from two additional species, rice a n d wheat.

L

ft.

0.60

0 k. 4,e

0.40

Animals

O~

T h e same eleven rats from E x p e r i m e n t s 1A a n d 1B were used.

0.20 f///~ . . . .

0 0 0

Procedure

o.oo Amylopectin

0 0

1.00

b

Lintner

Regular

Potato

Corn b

L_

0.80 L,,

I1. tO

0.60

U////////~

Resbtlts • .

k. 0.40 i////////// /////////// ///////////

~///////// E~ v, 0 0 0

0.20

///////////

0.00

0 0 C @

1.00 "

@

0.80 "

T h e procedure is similar to that of E x p e r i m e n t 1B. T h e rats were offered a choice of raw versus cooked rice starch, a n d a choice of raw versus cooked wheat starch o n successive days. T h e rats were given a l-day rest a n d t h e n retested. T h e following week, a n additional test was c o n d u c t e d to confirm the results of Experiments 1A a n d lB. T h e rats were offered a choice of raw versus cooked regular corn o n the first day, a n d a choice of raw versus cooked regular potato starch on the second day. Unlike the other tests, this test was n o t repeated.

Regular

Lintner

Potato

Potato

Rats preferred cooked rice over raw rice starch, but this effect was not significant for w h e a t starch (see Fig. 3). T h e degree of preference was greater for rice t h a n for w h e a t starch [see Fig. 3, F ( I , 10) = 9.0, p < 0.02]• Rats also ingested m o r e fluid w h e n offered rice t h a n wheat starch [36 + 3 versus 33 + 3 g/day, F(1, 10) = 9.5, p < 0.02]. In the follow-up test, the rats did not show a significant preference for cooked regular corn starch [preference o f 0.38 + 0.10, t(10) = 1.2], whereas they showed a strong preference for cooked regular potato starch [preference o f 0.87 _+ 0.04, t(10) = 9, p < 0.001].

b Substance

Rats

Were

Tested

With

L

@ a. .C 0

Cooked Potato

® 0 t-

1.00

'6-

0.80

~

Cooked AmylopeotJn

0.60 "

I.

0.40

a. tO

0.20

0.60

0.40

O O

0

o.oo Rice

Wheat

Starch Type FIG. 1. Mean preference for cooked starch by rats given a choice of raw versus cooked starch in Experiment i. Each bar represents the mean + SEM of two tests. The dashed line represents 50% preference (i.e., no evidence for discrimination). "Preference different from 50% p < 0.05. bPreference different from 50% p < 0.01.

"1~

0.20

0 0 0

o.oo

b

Amylopectin

None

Substance

Rats

Were

Trained

Potato to

Avoid

FIG. 2. Mean preference for cooked starch by rats given a choice of cooked starch versus vehicle in Experiment 2. Mean + SEM. The dashed line represents 50% preference (i.e., no evidence for discrimination). bpreference different from 50% p < 0.01.

538

RAMIREZ Substance

Rats

Tested

I~

Raw Potato

0 o to 0 o

Were

With

Raw Am ylopectln

1.00 b a

0.80

0.60

L

0.40

b

~'~

b

CO

0.20

rr

o.oo

b

Control

Substance

Amylopectin

Rats

Were

Trained

Potato

to

Avoid

FIG. 3. Mean preference for raw starch by rats given a choice of raw starch versus vehicle in Experiment 2. Mean _+ SEM. The dashed line represents 50% preference (i.e., no evidence for discrimination). aPreference different from 50% p < 0.05. bPreferencedifferent from 50% p < 0.01.

DISCUSSION OF EXPERIMENTS 1A, lB, AND IC

For all of the starch types examined, rats preferred cooked over raw starch. An earlier study failed to find such a preference, but it used a lower starch concentration (0.5% in the earlier study, 1% in the present work). However, the degree of preference varied greatly with the type of starch used. Preference for cooked starch was weak when corn and wheat starch were used. Indeed, no significant preference for cooked corn starch occurred in the final test of Experiment 1C as well as in an earlier report (18). Preference for cooked starch is high when potato and rice starch are used. Preference does not seem to be correlated with total intake; corn amylopectin, and regular corn induced different levels of fluid intake but not different preference values (Experiment 1A). Rice and potato starch arc virtually the most dissimilar starch types imaginable (1,7,12). There is no known trait shared by potato and rice starch, that is lacking in regular corn, corn amylopectin, and wheat starch. This makes it difficult to identify the physical or chemical basis for the observed difference among starch types. It is hypothesized that preference for cooked starch is due to unidentified impurities locked up within starch granules. Cooking releases impurities that are normally trapped inside starch granules (7,9). The flavor of these impurities cannot be predicted from the physicochemical attributes of different starches, but could influence preference for cooked starch. EXPERIMENT 2

Oreater preference for cooked starch than for raw starch implies that these two forms of starch differ with regard to flavor intensity or quality. These alternatives were tested in the second experiment, in which rats were trained to avoid raw starch, and were subsequently tested for their avoidance of raw and cooked starch. Two different starches were used--corn amylopectin and Lintner potato. These starches differ with respect to the degree to which rats prefer the cooked form (see Experiment 1A). After training, a series of preference tests was conducted to test various hypotheses. In each case, rats were tested for their

avoidance of starches from both species. If starches from the two species have precisely the same flavor, the degree of avoidance should be unrelated to starch type. In the first set of tests, rats were tested for their preference for cooked starch versus vehicle. If cooking drastically alters starch flavor, a conditioned aversion to raw starch should not generalize to the cooked version of that starch. The rats were then tested for their avoidance of raw starch to assess the degree of aversion to the original training stimulus. The third set of tests provided a sensitive test for the effects of cooking on flavor intensity. The rats were given a choice of raw versus cooked starch. The concentration of starch in these tests was reduced to one-half that used in training. If cooked starch has a more intense flavor than raw starch, rats should avoid the cooked to a greater degree than the raw starch. Alternatively, if flavor quality is affected by cooking, the raw starch may resemble the original training stimulus more, even though the concentration was reduced. Finally, the rats were retested for their preference for raw starch versus vehicle to assess how much extinction occurred.

Animals A total of 21 rats weighing 186 _+ 2 g were used.

Method The procedures are similar to those described in another publication [Experiment 5 in (24)]. All tests were conducted on Mondays, Wednesdays, and Fridays during the first hour of the dark part of the light/dark cycle. The rats were given plain water ad lib from Saturday through Sunday each week. The only illumination during these tests was provided by red lights. For habituation trials (two per week), rats were water deprived for 24 h and given xanthan gum vehicle to drink overnight. On training trials (one per week), the rats were then given either xanthan gum vehicle (control group), corn amylopectin, or Lintner potato starch to drink; within 10 rain after they began drinking, they were injected intraperitoneally with 1 ml 0.15 M lithium chloride per 100 g body weight. About one-half hour after the lithium chloride injection, the rats were given plain water to drink. A total of four training trials were given (one per week). On the first trial, the concentration of starch was 3%; on subsequent trials, the concentration was reduced to 1%. The higher concentration was used in the first trial in an effort to make the stimulus more salient. Control rats were given the same series of injections as experimental rats, but they were not offered starch suspensions until the preference tests began. Preference tests were conducted using procedures similar to those used for training (only one test per day). The rats were water deprived for 24 h and given a choice of two fluids for 30 min. At the end of each test, the rats were given plain water to drink. Each preference test was conducted twice, the first time using Lintner potato starch, the second time using corn amylopectin. The following choice tests were conducted: 1% cooked starch versus vehicle, 1% raw starch versus vehicle, 0.5% raw versus 0.5% cooked starch, and raw starch versus vehicle. Starch was cooked and suspended in water by the same method described for Experiment 1.

Results As shown in Fig. 2, rats trained to avoid either starch avoided both forms of cooked starch (all p < 0.01). However, rats trained to avoid Lintner potato starch avoided cooked

COOKED STARCH

539

Lintner potato starch to a greater degree than did rats trained to avoid corn amylopectin, t(18) = 2.6, p < 0.05. Conversely, rats trained to avoid corn a m y l o p e c t i n avoided cooked corn amylopectin to a greater degree than did rats trained to avoid Lintner potato starch, although this was not significant, t(18) = 1.4, NS. A similar pattern appeared w h e n the rats were tested for their avoidance o f raw starch (see Fig. 3). Thus, rats trained to avoid one starch avoided both forms of raw starch (all p < 0.001). However, rats trained to avoid Lintner potato starch avoided cooked Lintner potato starch to a greater degree than did rats trained to avoid corn amylopectin, t(18) = 3.0, p < 0.01. Conversely, rats trained to avoid corn a m y l o p e c t i n avoided cooked corn amylopectin to a greater degree than did rats trained to avoid Lintner potato starch, t(18) = 2.2, p < 0.05. Since tests involving a choice of starch versus vehicle did not give a clear picture of possible differences between cooked and raw starch, the rats were offered a choice of cooked and raw starch (Fig. 4). Rats that were trained to avoid raw starch, prefenced cooked over raw starch, that is, they avoided raw starch to a greater degree (all p < 0.05). Furthermore, rats trained to avoid Lintner potato starch showed a greater degree of avoidance of raw Lintner potato starch than rats trained to avoid raw corn amylopectin, t(18) = 2.8, p < 0.05. However, rats trained to avoid corn amylopectin did not avoid raw corn amylopectin to a greater degree than did rats trained to avoid Lintner potato starch, t(18) = 0.04, NS. The final tests of preference for raw starch versus vehicle closely paralleled the initial preference tests (Fig. 5). Rats trained to avoid one starch avoided both forms of cooked starch (all p _< 0.05). Rats trained to avoid Lintner potato starch avoided cooked Lintner potato starch to a greater degree than did rats trained to avoid corn amylopectin, t(18) - 2.4, p < 0.05. Conversely, rats trained to avoid corn amylopectin avoided cooked corn amylopectin to a greater degree than did rats trained to avoid Lintner potato starch, t(18) = 2.3, p < 0.05.

Substances

Rats

R a w vs. Cooked Potato

® o t-

Were ~

Tested

With

R a w vs. Cooked Amylopectin

1.00

0.80

® In

n tO I=,

0.60

...../ /

.....

............

0.40

0.20

rr

0.00 Control

Substance

Arnylopectin

Rats

Were

Trained

Substance

Rats

Were

Tested Raw Amylopectin

Raw Potato

@ o t-

1.00

[

0.80

L

@

With

b

® 0.60 co I-

a 0.40

0.20

0~

n"

O.OQ Control

Substance

Amylopectin

Rats

Were

Trained

Potato

to

Avoid

FIG. 5. Mean preference for raw starch by rats given a choice of raw starch versus vehicle in Experiment 2. Mean _+ SEM. The dashed line represents 50% preference (i.e., no evidence for discrimination). aPreference different from 50% p < 0.05. bPreference different from 50% p < 0.01.

Discussion When the rats were offered two starches to chose from, the concentration of starch (0.5%) was less than that used during training (1-3%). If cooked starch possess a more intense starch flavor than raw starch, one would expect that rats trained to avoid starch would show a greater avoidance of cooked than raw starch. Since this prediction was not confirmed, it must be concluded that cooking alters qualitative aspects of starch flavor. The observed differences between potato and corn starch are best interpreted in terms of qualitative differences in starch flavor. If the starches differ in flavor intensity, corn amylopectin should have the more intense flavor because rats show a stronger preference for it than for potato starch (18,20). Previous studies indicate that rats trained to avoid a flavor tend to avoid more intense flavors to a greater degree than milder flavors (15). Therefore, if species differences in starch preference are due to intensity differences, both groups of rats that were trained to avoid starch should have avoided corn amylopectin more than Lintner potato starch. Since rats showed maximal avoidance for the starch species that they were trained to avoid, rather than for the one more preferred by untrained rats (presumably more intensely flavored starch), it seems likely that species differences in starch flavor are due to qualitative differences in flavor intensity. It is of interest that species differences in starch flavor seem to survive cooking (see Fig. 2), even though cooking alters some qualitative aspect of starch flavor. This result seems to suggest that the qualitative flavor changes induced by cooking are different from species-related qualitative flavor differences.

Potato

to

Avoid

FIG. 4. Mean preference for cooked starch by rats given a choice of cooked starch versus raw starch in Experiment 2. Mean _+ SEM. The dashed line represents 50% preference (i.e., no evidence for discrimination), aPreference different from 50% p < 0.05. bpreference different from 50% p < 0.01.

GENERAL DISCUSSION

When it was first hypothesized that starch had a flavor, it was assumed that malto-oligosaceharides, malto-polysaccharides and starch all shared a single taste quality (25,26,28). Recent studies indicate that there is a qualitative difference between the flavors of malto-oligosaccharides, corn starch, and cellulose (19,22,23).

RAMIREZ

540 The present studies complicate the picture further by indicating that different starches share some flavor components in common, but differ with regard to other flavor components. It seems that rats' ability to sense carbohydrates may be richer than anyone anticipated. Nevertheless, diverse starch types (cooked, raw, potato, corn) share at least one flavor note in c o m m o n , because rats trained to avoid one type, also avoid the other types. The present results are not easily reconciled with the assumption that starch perception is based on the detection of the textural properties of starch. Since cooking completely destroys the granular structure of starch (5), it would be expected to obliterate textural cues related to starch granules. Cooking modifies, but does not obliterate, starch flavor. Cooking does not even block rats ability to discriminate between starches from different species (Experiment 2). Therefore, texture cannot play more than a m i n o r role in starch perception by rats. The present results indicate that at least part of the perception of starch is mediated by perception of water-soluble or volatile impurities. Rats usually prefer cooked over raw starch. The degree of preference for cooked over raw starch varies with the

species from which the starch was derived, but does not seem to be correlated with any obvious physical attribute of the starch (e.g., particle size, gelatinization temperature, etc.). It is known that minor impurities, trapped inside starch particles, are released by disruption of the granules by cooking (7,9). Because they are normally trapped deep inside the granules, these impurities would not be sensed in raw starch, but could enhance preference for cooked starch. Impurities may also be invoked to explain the differences in flavor a m o n g starches from different species. hnpurities may not account for all of the flavor components of starch. Other experimental methods indicate that rats do not show reliable preferences for fluids containing starch impurities but no starch (18,20,21). On the other hand, if impurities do account for much of starch flavor, they are probably unstable once they have been removed from starch (21). This instability would make it difficult to identify their chemical identity. ACKNOWLEDGMENTS This work was supported by an NIH grant (1 R01 DC01107), and the Monell Chemical Senses Center. Highly skilled technical assistance was provided by Samantha Keezer.

REFERENCES 1. AVEBE America Inc. Potato starch: The natural foodstarch for thickening and binding. Hopelawn, NJ: AVEBE America Inc.; 1988: 17p. 2. Banks, W.; Greenwood, C. T. Starch and its components. Edinburgh: Edinburgh University Press; 1975. 3. Bartoshuk, L. M. Is sweetness unitary? An evaluation of the evidence for multiple sweets. In: Dobbing, J., ed. Sweetness. London: SpringerVerlag; 1987:33-47. 4. Blanshard, J. M. V. Starch granule structure and function: A physicochemical approach. In: Galliard, T., ed. Starch, properties and potenti~il. Chichester: John Wiley & Sons; 1987:16-54. 5. Collison, R. Swelling and gelation of starch. In: Radley, J. A., ed. Starch and its derivatives. 4th ed. London: Chapman and Hall; 1968: 168-193. 6. Collison, R. Starch retrogradation. In: Radley, J. A., ed. Starch and its derivatives. 4th ed. London: Chapman and Hall; 1968:194-203. 7. Galliard, T.; Bowler, P. Morphology and composition of starch. In: Galliard, T., ed. Starch, properties and potential. Chichester: John Wiley & Sons; 1987:55-78. 8. Giza, B. K.; Scott, T. R.; Sclafani, A.; Antonucci, R. F. Polysaceharides as taste stimuli: Their effect in the nucleus tractus solitarius of the rat. Brain Res. 555:1-9; 1991. 9. Gracza, R. Minor constituents of starch, ln: Whistler, R. L.; Paschall, E. F.; BeMiller, J. N.; Roberts, H. J., eds. Starch: Chemistry and technology, vol l. Fundamental aspects. 1st ed. New York: Academic Press; 1965:105-131. 10. Hixon, R. M.; Brimhall, B. Waxy cereals and red iodine starches. In: Radley, J. A., ed. Starch and its derivatives. 4th ed. London: Chapman and Hall; 1968:247-281. 11. Holm, J.; Lundquist, I., Bjrrck, I., Eliasson, A.; Asp, N. Degree of starch gelatinization, digestion rate of starch in vitro, and metabolic responses in rats. Am. J. Clin. Nutr. 47:1010-1016; 1988. 12. Juliano, B. O. Rice starch: Production, properties, and uses. In: Whistler, R. L.; BeMiller, J. N.; Paschall, E. F., eds. Starch: Chemistry and technology. 2nd ed. Orlando, FL: Academic Press; 1984:507-528. 13. Miles, M. J.; Morris, V. J.; Orford, P. D.; Ring, S. G. The roles of amylose and amylopectin in the gelation and retrogradation of starch. Carbohydr. Res. 135:271-281; 1984. 14. Nissenbaum, J. W.; Sclafani, A. Qualitative differences in polysaccharide and sugar tastes in the rat: A two-carbohydrate taste model. Neurosci. Biobehav. Rev. 11:187-196; 1987.

15. Nowlis, G. H. Conditioned stimulus intensity and acquired alimentary aversions in the rat. J. Comp. Physiol. Psychol. 86:1173-1184; 1974. 16. Pettitt, D. J. Xanthan gum. In: Glicksman, M., ed. Food hydrocolloids, vol. I. Boca Raton, FL: CRC Press; 1982:12%149. 17. Ramirez, I. Why do sugars taste good? Neurosci. Biobehav. Rev. 14:125-134; 1990. 18. Ramirez, I. Chemoreception for an insoluble nonvolatile substance: Starch taste? Am. J. Physiol. 260:RI92-RI99; 1991. 19. Ramirez, I. Does starch taste like Polycose? Physiol. Behav. 50:389392; 1991. 20. Ramirez, 1. Starch flavor: Apparent discrimination between amylopectin and amylose by rats. Physiol. Behav. 50:1181-1186; 1991. 21. Ramirez, I. Starch flavor: Can rats taste an insoluble substance? Chem. Senses 16:361-371; 1991. 22. Ramirez, I. Thresholds for starch and Polycose are lower than for sucrose in rats. Physiol. Behav. 50:699-703; 1991. 23. Ramirez, I. Chemoreception for cellulose. Chem. Senses 16:691698; 199l. 24. Ramirez, I. Chemoreception for fat: Do rats sense triglycerides directly or do they rely on surrogate compounds? Appetite (in press). 25. Sclafani, A. Carbohydrate taste, appetite, and obesity: An overview. Neurosci. Biobehav. Rev. 11:131-153; 1987. 26. Sclafani, A. Starch and sugar tastes in rodents: An update. Brain Res. Bull. 27:383-386; 1991. 27. Sclafani, A.; Nissenbaum, J. W. Taste preference thresholds for Polycose, maltose, and sucrose in rats. Neurosci. Biobehav. Rev. 11:181-185; 1987. 28. Sclafani, A.; Nissenbaum, J. W.; Vigorito, M. Starch prefbrence in rats. Neurosci. Biobehav. Rev. 11:253-262; 1987. 29. Somenarain, L.; Jakinovich, W., Jr. Antagonism of the gerbil's sweetener and Polycose gustatory responses by copper chloride. Brain Res. 522:83-89; 1990. 30. Watson, S. A. Corn and sorghum starches: Production. In: Whistler, R. L.; BeMiller, J. N.; Paschall, E, F., eds. Starch: Chemistry and technology. 2nd ed. Orlando, FL: Academic Press; 1984:417468. 31. Wurtzburg, O. B. Converted starches. In: Wurtzburg, O. B., ed. Modified starches: Properties and uses. Boca Raton, FL: CRC Press; 1986:17-40.

Is starch flavor unitary? Evidence from studies of cooked starch.

Although starch is the world's most abundant nutritive carbohydrate, its sensory properties are not as well understood as those of sugars. Previous re...
726KB Sizes 0 Downloads 0 Views