Chem. Percept.

(2010) 3:68–84
DOI 10.1007/s12078-010-9067-z

Does Food Color Influence Taste and Flavor Perception
in Humans?
Charles Spence & Carmel A. Levitan &
Maya U. Shankar & Massimiliano Zampini

Received: 3 September 2009 / Accepted: 8 February 2010 / Published online: 9 March 2010
# 2010 Springer Science+Business Media, LLC

Abstract In this paper, we review the empirical literature research findings concerning the second question clearly
concerning the important question of whether or not food support the view that people’s judgments of flavor identity
color influences taste and flavor perception in humans. are often affected by the changing of a food or drink’s color
Although a superficial reading of the literature on this topic (be it appropriate, inappropriate, or absent). We discuss the
would appear to give a somewhat mixed answer, we argue possible mechanisms underlying these crossmodal effects
that this is, at least in part, due to the fact that many and suggest some of the key directions for future research
researchers have failed to distinguish between two qualita- in order to move our understanding in this area forward.
tively distinct research questions. The first concerns the role
that food coloring plays in the perception of the intensity of Keywords Flavor . Taste . Color . Perception . Crossmodal .
a particular flavor (e.g., strawberry, banana, etc.) or taste Multisensory . Expectancy . Attention
attribute (e.g., sweetness, saltiness, etc.). The second
concerns the role that food coloring plays in the perception
of flavor identity. The empirical evidence regarding the first Introduction
question is currently rather ambiguous. While some
researchers have reported a significant crossmodal effect Does food coloring influence taste and flavor perception in
of changing the intensity of a food or drink’s coloring on humans? Although researchers have been investigating this
people’s judgments of taste or flavor intensity, many others important (both on a theoretical and practical level)
have failed to demonstrate any such effect. By contrast, the question for more than 70 years now (see Duncker 1939;
Masurovsky 1939; Moir 1936 for early research), an
unequivocal answer to the question has not, as yet, been
C. Spence : M. U. Shankar
reached. That, at least, would seem to be the conclusion
Department of Experimental Psychology, University of Oxford,
Oxford, UK drawn by the majority of researchers in the field. Take, for
example, Lavin and Lawless’s (1998, p. 284) claim that
C. A. Levitan “The literature on the effects of color on taste and flavor
Cognitive Science Program, Occidental College,
judgments is consistent in its inconsistency” or Koch and
1600 Campus Road,
Los Angeles, CA 90041, USA Koch’s (2003, p. 240) statement that “In fact, it may be that
color has nothing to do with the taste of food or drink.”
M. Zampini Meanwhile, Bayarri et al. (2001, p. 399) have also
Center for Mind/Brain Sciences & Department of Cognitive
suggested that “…the possible influence of color on flavor
Sciences and Education, University of Trento,
Rovereto (TN), Italy perception is under discussion and no clear conclusions
have been attained yet.” In the present article, we argue that
C. Spence (*) one important reason why our understanding of the nature
Crossmodal Research Laboratory,
of any crossmodal effects of food coloring on taste and
Department of Experimental Psychology, University of Oxford,
Oxford OX1 3UD, UK flavor perception in humans has progressed so slowly
e-mail: relates to the fact that many researchers have failed to

1980. or else have demonstrated complex (and/or not aware of any research having been conducted assessing unexpected) interactions that have proved rather more the effect of color on the identification of basic tastes (e. many other studies have evidence unequivocally supports an affirmative answer to the either failed to demonstrate any such crossmodal effect (e. when solutions containing an odorless tastant such as sugar 1980. 1995). Does food color influence taste intensity? One of the Which Senses Contribute to the Perception of Flavor? classic studies to investigate color’s influence on taste sensitivity was conducted by Maga (1974). 1983. this research focus on colored drinks (be they perception of a food’s flavor by influencing the gustatory carbonated or uncarbonated) reflects both the ease of stimulus control qualities of the food. food’s flavor. Given this bias Koza et al. we have also chosen to focus our review primarily on qualities of the food. Christensen 1985. beverages. Chan and Kane-Martinelli 1997. banana.Chem. We discuss several of the possible Strugnell 1997. significant effect of changing the intensity of the coloring while a number of studies have demonstrated a significant added to a food on people’s judgments of flavor and/or taste effect of increasing the level of food coloring on people’s intensity (e. Roth et al. by influencing the oral–somatosensory in the literature.. ratings of taste or flavor intensity across a range of different 1982. 2009). there is nothing much to say about vision’s influence. thermal. . Fletcher et al. The flavor and Engelen 2006 for a review).g. 137. 1987).. more recently. Johnson et al. to these two questions. 2 may be influenced by tactile. de “complex combination of the olfactory. such as cognitive expectancy. green. He investigated Before we proceed. researchers have looked at color’s role in influencing kinaesthetic effects” (see Delwiche 2004. Romeu and De Frank et al. etc. Stevenson 2009. syrups (Kanig 1955). on the oral–somatosensory attributes of flavor (see Christensen 1983. 2007).. Moir 1936. ISO 5492 2008). DuBose et al. second question. 1983. may then modify the choice. experiment 2).) or taste (such as sweetness. that color provides one of the few distinctive non-olfactory features of such stimuli (see Christensen 1985. 2005). McCullough et al. 1988). intrinsic to it (though see also Auvray and gums (Teerling 1992). and yogurt (Norton and Johnson 1987). Johnson et al. and attention. drinks. olfactory. Roth et al. In fact. Does Food Color Influence Perceived Taste or Flavor sourness. Levitan et al. 1989. we review the evidence regarding color’s effect on two relatively independent research questions: (1) Does the taste and flavor. the evidence pertaining to each of these is dealt with separately in the following sections. by influencing the olfactory attributes and creation that such experimental materials afford and also the fact of the food (as perceived orthonasally and/or retronasally. the International Standards It should be noted that the near absence of research means that.1 the color present in a food or drink influence people’s perception of the intensity of a particular flavor (e. experiment 1. and solutions as the stimuli of cues. to demonstrate any such effect (e. Zampini et al. or gustatory.. 1978.. Johnson and Clydesdale 1982. with people’s judgments of a flavor’s Alley and Alley 1998. By contrast. (2010) 3:68–84 69 distinguish between the evidence pertinent to evaluating article. chocolate (Duncker 1939. and/or by influencing the overall those studies that have investigated the effect of color on taste and multisensory flavor percept (or Gestalt.. intensity would appear to be rather more ambiguous than bottom-up multisensory integration. cake (DuBose et al. Visual However.g. 2007. see Table 1). identity often being reliably affected by a food’s color. Below. painful and/or Over the years. or change in the intensity. wine definition. Alley and Alley 1998. if any. but they are not. such as a food’s color. at least according to the ISO Shankar et al. it is important to note that the effects of coloring an aqueous solution red.)? (2) Does food coloring influence the correct Intensity? identification of a food or drink’s flavor? The evidence concerning the first question is indeed The evidence pertaining to the question of whether food rather mixed (Auvray and Spence 2008. and oral–somatosensory cues (in contrast to visual and auditory cues) all contribute directly 1 to flavor perception. however. We also discuss the evidence regarding presence versus absence. Frost and Janhoj 2007 for exceptions. 1). see Verhagen trigeminal sensations perceived during tasting.2 Given mechanisms that may underlie these crossmodal effects on that the evidence regarding color’s influence on taste taste and flavor identification. gustatory and Wijk et al. 1982. Gifford and Clydesdale 1986. 1988). difficult to interpret (e. sherbets (Hall 1958). of color’s influence on olfactory judgments where relevant. 1987). see Fig. 1991.g.g. Most likely. strawberry. as Organization (ISO 5492 1992) has defined flavor as a yet. etc. the majority of the research has tended to use various different drinks. we review the evidence relating Lawless 1998. Visual and auditory cues may modify a Tom et al. including jellies (Moir 1936). Kostyla 1978. or absent.g. Oram et al. see also people’s perception of the taste and flavor of many different foods. the published Vicente 1968. Johnson and Clydesdale 1982. 1989. Note here that we are al. Lavin and or salt are sampled). 2004. coloring influences people’s perception (or ratings) of taste Zampini et al. Gifford et appropriate. Stevenson 2009 for alternative views).g. Percept. 2008. and beverages. with some researchers observing a or flavor intensity is currently rather ambiguous: that is. Pangborn 1960. In this flavor perception in various solutions. Spence 2008. p. inappropriate. while many others have failed drinks (Hyman 1983. be it Frank et al. its effect on flavor intensity.

(2010) 3:68–84 Table 11 Summary Summary of of the thestudies studiesthat thathave havebeen beenpublished publishedto date to date thatthat have table investigated highlights thethefact effect that of thevarying majoritytheofpresence research vs. So. meaning that the participants were presumably never demonstrated an effect of food coloring on taste perception uncertain with regard to the identity of the tastant whose in sweetened solutions (that sometimes contained a cherry presence they were trying to detect. Sig significant result. (1983) Sw Sig Gifford and Clydesdale (1986) Sa n. in an oft-cited study.color absence. in this absence. (1995) Sw Sig Chan and Kane-Martinelli (1997) Sa n. (2007) Fl Complex Zampini et al. sensitivity to sweet taste. (1982) Sw Sig Hyman (1983) Fl Sig Johnson et al. addedthe to a solution focusedon on participants’ the influence tasteof and/or colorflavor on sweetness (Fl) intensity perception. in the pattern The of table results also highlights that has been the inconsistency a solution on participants’ in the patterntaste of results and/orthatflavor has(Fl) beenintensity reportedratings. tastes was tested in a separate part of the experiment. ratings. and bitter). So Fl Complex Sw sweet.s. the addition of green coloring ipants could more easily detect the presence of sucrose than to a sweet solution significantly increased taste sensitivity. In increased in order for his participants to be able to detect its the threshold task. with the darker-colored solutions bitter taste sensitivity. In the magnitude estimation task.s. Finally. (1989) Sw n. Fletcher et al. sweet. the fact that the majorityor of the research intensity.s. (1980) Fl Complex Johnson and Clydesdale (1982) Sw Sig Johnson et al. have investigated the effect of varying or the intensity. Johnson and Clydesdale (1982). to dateThe across the reported various to studies date across that have the various been reported studies that have been reported Study Tastant Flavor Result Pangborn (1960) Sw. Johnson and Clydesdale found that changing the sensitivity.s. both the yellow and green coloring of solutions level of food coloring had a significant effect on partic- decreased participants’ sensitivity. So. in this of area the color has focused added toon the also influence highlights of color theon inconsistency sweetness perception. Johnson and Clydesdale found that on presence in the colored (as compared to the uncolored) average. Bi Sig Kostyla (1978) Sw. (1988) Sw Sig Frank et al. Maga 1974). (2001) Sw Fl Complex Zampini et al. Percept. (1963) Sw Sig Romeu and De Vicente (1968) Fl Sig Maga (1974) Sw. n. non-significant result. So sour. Gifford et al. though the while yellow color decreased taste sensitivity (see Table 2). red coloring had no significant effect on their performance.s. with red coloring again ipants’ perception of the sweetness of both odorless and having no effect. Complex typically a mixture of significant and non-significant results yellow on perceptual thresholds for four of the basic tastes coloring had no such effect. Note that each of the basic effect on taste detection thresholds for salt solutions. partic- solutions. (2008) Sw. So Sig Pangborn et al. Maga flavoring). Sa salt. In many cases.70 Chem. The The table highlights appropriateness/inappropriateness. Lavin and Lawless (1998) Sw Complex Bayarri et al. (1978) Sw Complex DuBose et al. when odorless solutions were colored red. for example.s. (1987) Sa n. adding color had no (salty. area has the appropriateness/inappropriateness. (1991) Sw Complex Philipsen et al. With respect to sour taste however. Coloring a clear solution red decreased cherry-flavored solutions. Roth et al. the presence of the vs. The participants in their study had to perform observed that the concentration of the tastant had to be both a threshold task and a magnitude estimation task. Sa. intensity of the color did not have a significant effect on Interestingly. while the addition of yellow and green being rated as 2–10% sweeter than the lighter-colored . So Fl Sig McCullough et al. Strugnell (1997) Sw Sig Alley and Alley (1998) Sw n. sour. when they were uncolored (cf.

Zellner and Whitten consistent pattern of results has been reported more recently 1999). That is. many fruits show a transition from colors at the exist. In further studies. such as chicken broth (see Chan and Kane-Martinelli 1997.g. perception only once the olfactory. Maga 1974) of sweetness and that increasing the intensity reference solutions despite the fact that the actual concen- tration of sucrose was 1% lower. 1983) increased presented in uncolored solutions) by 2–13% when the intensity of the cherry red coloring was Color of Taste increased. green end of the spectrum. Complicating matters still further.. Red No effect No effect No effect Decrease Color cues have also been reported to influence parti.Chem. however. see (citric acid) (sucrose) chloride) (caffeine) Kostyla 1978. Interestingly. Maga suggested that Oral. the foods) might help explain why red/green coloring should have nature of the crossmodal influence of visual cues on flavor perception such a profound effect on sweetness/sourness perception. 2007 on this point). By contrast. Strugnell 1997). but light green solutions as sweeter than dark date. Zellner and Kautz 1990. To red solutions. Morrot et al. Yellow Decrease Decrease No effect No effect cipants’ ratings of sourness/tartness (e. there is only very limited research detailing any visual green solutions. through yellow. Visual cues (such as the color of a Maga (1974) argued that there is a natural correlation beverage) may exert a crossmodal influence on olfaction. A (e. Such crossmodal effects. Engen 1972. Kostyla 1978.(Johnson et al. then look to see what color the filling of the chocolate has. may well vary with the task at hand. several further studies have also solution documented a significant effect of the addition of (specifical. people have few Fig. absence) of concerning the influence of visual (color) cues on perceived taste and food coloring leads to a significant effect on ratings of flavor intensity the presence (vs. Percept. and/or on oral somatosensation. sory cues have been integrated into a multisensory flavor percept (so. between increases in redness and sweetness in integrated flavor percept). as is the rather more mixed evidence appear to demonstrate that the addition (vs. In order to try and explain this seemingly inconsistent effect of the addition of Flavour color on salt and sweetness perception. gustation. however.. Pangborn 1960. might influence flavor perception. should they That is. Over the years. to colors at the Alternatively. Maga suggested that prior exposure to this natural color–taste for example.g. we are aware of no research that has directly addressed the question of whether color to rate for sweetness using a nine-point categorical scale. (2010) 3:68–84 71 Vision detection threshold for the presence of salt in aqueous solu- tions reported by Maga (1974) has been extended in several other well-controlled laboratory studies in which coloring was added to more ecologically valid food substrates. There is robust psychophysical evidence that sweetness ratings while enhancing sourness ratings in pear visual (color) cues can modulate people’s perception of the identity nectar. There is also convincing psychophysical evidence that visual in a study by Lavin and Lawless (1998). 1 This figure highlights the multiple ways in which visual cues direct color associations with saltiness. The cues can influence a participant’s ability to discriminate (and/or participants rated the dark red solutions as sweeter than light identify) the identity of basic tastants when presented in solution. Pangborn (1960) demon- observed as a function of whether participants are asked to report on the pleasantness or intensity of a flavor versus having to identify or strated that the addition of green food coloring reduced discriminate the flavor. Johnson Table 2 Summary of the results from Maga’s (1974) study and his colleagues went on to use the same magnitude highlighting the effect of the addition of color on participants’ estimation procedure in order to show that sweetness sensitivity to each of the four traditional basic tastants when dissolved ratings for both cherry. Johnson and Clydesdale 1982. then color may well influence the correlation (i. might then have a carryover effect on the experienced flavor percept once the various unisensory cues have been integrated. Blackwell 1995. 2001. Zellner et al. 1991. visual information might influence flavor red end of the spectrum (see also Brice 1954. Green Decrease Increase No effect No effect Pangborn 1960). color might not have any effect on salt perception because Somatosensation salty foods can come in any color. 1987). if you bite into a filled chocolate. contributions to the oral–somatosensory attributes of food.e. The participants in (color) cues can influence people’s perception of the intensity of the this study were given four strawberry-flavored drinks of basic tastants when presented in odorless solutions (Johnson and different colors (light and dark red and light and dark green) Clydesdale 1982. Kostyla 1978. 1982) and in solution (and compared to performance when the tastants were strawberry-flavored drinks (Johnson et al.. the null result of color on the . and/or oral–somatosen. between redness and sweetness levels as many fruits ripen. Lavin and Lawless 1998. Gifford and Clydesdale 1986. Davis 1981. absence. different results may be Consistent with this suggestion. Maga 1974). Zampini et al. Sour Sweet Salty (sodium Bitter ly) red food coloring on the perception of sweetness (e. and hence. Evidence demonstrating the robust influence of visual cues on flavor identifi- While the results discussed thus far in this section would cation is discussed in the text.g. gustatory. though it should be noted that Pangborn and Hansen and intensity of both orthonasally and retronasally presented odors (1963) subsequently failed to replicate this finding. Gifford et al..

(1980) reported that overall flavor that color cues do not influence the perception of saltiness (see intensity was affected by color intensity. problematic. might all be expected to influence participants’ perceived intensity of different basic tastes. in turn. Kostyla perceptual sensitivity. 2007 for an exception). Pangborn 1960. rather than their actual taste sweetness intensity (four levels).. for example. 1988). the answer appears expectations (see Shankar et al. Using signal detection techniques solutions might have helped participants to realize that the (e. it seems possible that drinks in this study were uncolored. the perception of sweetness can be modified by the addition of red (or green) 3 Perceptual effects are usually defined in terms of a change in food coloring (e. each study. odor. This is unfortunate because of the susceptibility baseline on participants’ ratings of the sweetness of either to response bias. possibly reflecting a Alley 1998. Percept. Maga 1974). no-color-added solutions. 1978. see studies of color’s influence on perceived flavor intensity. red-looking drinks relative to participants’ assessment of the 2003). For example. Null effects are often reported in likely to respond that a flavor. the specific tasks and ranges of stimuli used in labeled magnitude scale rating procedure. Note that the participants in this latter study had procedure is also vulnerable to demand characteristics. odorless or strawberry odor-sweetened aqueous samples and thus. changed in response to changes in the color and odorant (present vs. DuBose et al. 1982. Frank et al. such as a change in the d′ measure that can be 1978. with higher color Chan and Kane-Martinelli 1997. intensity solutions giving rise to stronger flavor evaluation Gifford et al. Similarly. also be constraints. measure of bias (c or beta). see Johnson et al. under investigation (Maga 1974) and perhaps the precise Does food color influence flavor intensity? As visual presentation protocol adopted (such as how many incongru. 1987. see Green and Swets 1966) could help separate out color was not necessarily predictive of the presence/intensity perceptual from decisional effects here. but derived using signal detection theory (e. studies have failed to demonstrate any such effect (see many of the findings reported above are potentially Alley and Alley 1998. Frank et al.. Gifford and Clydesdale 1986. may show up in one’s (though see Zampini et al. it could be argued that the presence sweetened water or gelatine samples. By contrast. that to rate the intensity of 16 stimuli on a 21-point sweetness is. or taste is present whenever a studies in which colored drinks are compared to clear drinks color is added to a liquid. see Engen 1972). there have been far fewer ently colored trials the participant may be presented with. Odgaard et al. the story concerning see Gifford and Clydesdale 1986 for one null effect of color on color’s effect on flavor intensity seems much more clear-cut salt perception reported using the magnitude estimation and convincing.3 Due to space of the taste. this effect is not always observed (Alley and By contrast. they were more willing to respond that they did failed to increase perceived sweetness ratings of the orange indeed perceive a taste (cf. This raises the possibility under certain conditions. under closer examination. 1983. participants could easily be influenced by the coloring scale: the stimuli were generated by crossing the factors of such that their responses. 1983. see also Johnson threshold task. McCullough et response bias or change in the criteria for responding adopted by participants (such as. Christensen 1985. absent). detailing each of the various factors that might put forward to account for the null results reported by Alley help account for the occurrence of false positive results and Alley (1998). In other words. Johnson 1982. Pangborn 1960). as well as the demand characteristics of each Thus. with regard to the question of whether or not adding experiment and differences in the pools of participant color to an otherwise colorless substrate influences the tested. Certainly. 1989. yellow. (2007) also failed to (if that is what Johnson and colleagues’ results actually demonstrate an easily interpretable effect of variations in reflect) lies beyond the scope of the present paper. in their description of the McCullough et al. it is important to note that many other tion of taste intensity. suggest that color does indeed influence people’s percep- Roth et al. participants may simply want to that the inclusion of so many sweetened but uncolored please the experimenter). color (present vs. Zampini et al. 1982. to depend on a number of factors including the particular taste have influenced the pattern of results that were observed. 2009). Engen 1972. importantly. The consensus view would currently seem to be in this area.72 Chem. However. half of the intensity of the solutions (indeed. A similar explanation could. Lavin and Lawless 1998. inspection of Table 1 makes clear. It is important to note that the magnitude estimation clear drinks.. Pangborn 1960). in contrast to the rather mixed picture regarding rise to more pronounced effects than other methods (though the effect of color on taste intensity. Green and Swets 1966).g. footnote 4). which could. changes in decisional criteria. For example. absent). participants simply being more al.g. Johnson et al. Johnson and Clydesdale (1982) make no and Clydesdale 1982). that is. In one of the most frequently cited studies technique). for color intensity on perceived sweetness intensity using a instance. 1978. Alley and Alley reference to interleaving any “catch trials” in which reported no effect of the addition of color (red. Frank et al. unsweetened solutions were mixed in with the sweetened and green) when compared to a clear. of course. (2010) 3:68–84 of color can increase perceived sweetness intensity ratings Taken at face value.g. these results would appear to (Johnson and Clydesdale 1982. of the coloring in certain solutions simply led the (1989) reported that adding red food coloring to either an participants in those studies to expect a taste to be present. 1989. magnitude estimation typically gives However. . blue. percepts.

2001. taste and/or flavor intensity have rarely bothered to conduct and strawberry-flavored beverages decreased flavor ratings any kind of analysis to ensure they had sufficient power in by around 4%. Oram et al. ratings of the intensity Kanig 1955. plausible arguments can be made in both directions. Stillman 1993.g. Engen 1972. Moir 1936. Cain Note that a similar stimulus specificity has also been reported 1979. DuBose et al. Blackwell 1995. however. Zampini et al.g.g. Blue color reduced fruit flavor by 20% (and their experimental design to observe a significant effect had the addition of red coloring increased sweetness by 5–10%). 2007. while for the mint odor. other sensory cues regarding the odor’s identity (e. (e. presented retronasally (i.g. five or less stimuli in certain of the experiments orthonasally presented odors. Laing et al. most of these correctly identify odors more rapidly when colored appro- crossmodal studies involved investigating the effect of color priately than when uncolored or else inappropriately colored. in the perceived intensity of the very same odors when Lavin and Lawless 1998.. More complete tends to be rather limited. see Frank and Byram 1988. of the strawberry odor peaked at the middle color intensity. 2005. 2007.g. would their judgments of the identity of a specific tastant. sour. Sakai 2004. 2007. This may be particularly important as people were making flavor judgments. 4 the results of a study by Koza et al. Kostyla (1978) reported that null results of the addition of food coloring on ratings of the addition of yellow color to sweetened cherry-. .g. (1991) have shown that participants intensity judgments. (2010) 3:68–84 73 responses by participants for orange. Engen 1972. Shankar et al.. In this regard. the number of trials a given participant can also been reported by Romeu and De Vicente (1968).(but not for the cherry-) It is worth noting that researchers who have documented flavored beverages. For example. Complicating this issue even further is varying the intensity of the color on flavor intensity are rarer the fact that the effects of food coloring on ratings of taste and (Zampini et al. Zellner and Whitten 1999). 2008). Morrot et al. note that we are aware of no with increasing color intensity. identity are influenced by a food’s color (e. particular taste (see Stevenson et al. (2005) are critical. green mint solutions. there was a monotonic increase 2008. appear to quickly learn that the color of a beverage no longer predicts a Zellner and Durlach 2003). 2009. Institute of Food Technologists 1980.Chem. 2008. they demonstrated that while the addition of color studies in which the participants were only given a very small number of resulted in an increase in the perceived intensity of stimuli to evaluate (e. several studies have investigated the effect of color or inappropriate) had little effect. robust effects of color cues on judgments of odor intensity have now been reported in numerous studies (e. one been present (see Frick 1995). For Note here also that many of the demonstrations of the influence of color on taste and/or flavor in the literature have actually been reported in those example. see also Christensen 1983. it actually led to a reduction reported by DuBose et al. for it would seem likely that color should also influence flavor example. It is therefore currently unclear whether the discre- the basic tastes (e. or 6–15 stimuli in the studies reported by Alley and Alley 1998. in fact. 2008). Indeed. Desor and Beauchamp 1974.. By intensity. 2000). Jönsson et al. as they would be if participants Shankar et al. Zellner and Whitten reported By contrast. or on the detection of the presence of a tastant It should be noted here that people are very poor at in solution. Maga 1974. Percept. on orthonasal olfactory odor identification judgments. see identifying orthonasally presented odors in the absence of any Bayarri et al. Hall 1958. 1995. and hence in this section. Does Food Color Influence Perceived Flavor Identity? 2001. sweet. So. Because of the nature of Elsewhere. 2009). 1986. the specific color that was added (be it appropriate contrast. 2002). Meanwhile.. we only discuss color played a large role in modulating perceived odor studies of color’s influence on flavor identification). see also Dolnick 2008. studies that have examined the effect of food color on taste Zellner and Whitten showed that while the intensity of the identification. Hoegg and Alba 2007b.. Hyman 1983. raspberry-. In a subsequent experiment. and null results of 2001.g. then. many (and. appear to be rather stimulus-specific (e. 1991). Zellner et al.e. Zellner and Kautz 1990. Stillman 1993. flavor intensity. 2001). though see also Valentin et people appear to be much better at identifying the presence of al. Rouanet 1996). 2006). Interestingly. and bitter. significant effects of color on flavor intensity have flavor research. on participants’ orthonasal odor judgments.4 hence constraining the complex results have been reported in other studies where ability of experimenters to detect what may well be a certain color–flavor combinations appear to give rise to bigger relatively small behavioral effect (see also Lecoutre and Derzko effects than others (Bayarri et al. salty. By contrast. Parr et al. 2001. 1980. 2003.. we would argue all) studies that the addition of color (presented at one of four different published to date support the claim that judgments of flavor color intensity levels) influenced participants’ odor intensi. it would groups of researchers over the last 50 years are best explained seem likely that participants’ judgments of odor (and hence of in terms of a failure to properly control for all of the potential flavor) identity would be much more likely to be influenced confounding variables or by a lack of statistical power in the by the presentation of an incongruent visual color cue than experimental designs used by certain researchers in the field.. see pant results of color on taste intensity reported by different Bartoshuk 1975. Given results such as these. ty responses when orthonasally sniffing red strawberry and 1980. Davis 1981. Zellner et al. Eskenazi et previously for the case of olfactory–gustatory interactions al.. Morrot et al. A priori. That said. Zampini et al. Levitan et al.

in colors of the drinks that were presented to the participants..g. raspberry. As such. 1995). grapefruit. The participants were given a checklist of 14 possible simply by the color of the foodstuff itself.g. lemon lime. it is unclear whether the the results from one of the experiments conducted by results reported in many of the early studies demonstrating DuBose et al. for example. The partic. 2007. meaning that gated crossmodal influences of food coloring on flavor the color cues actually modulated the flavor percept itself. DuBose et al. 1976). Zampini et al. or flavorless) and color (red. perception in humans (e. how participants respond when placed in an ecologically invalid The numerical values indicate the percentages of each flavor response laboratory context than they do about color’s influence on flavor for each color perception in the real world. laboratory studies often find themselves.g. Oram et al. though see Shankar et al.. and/or flavor cherry-flavored drink colored red). if participants variety of differently colored fruit-flavored drinks. Moir see Green and Swets 1966. then they may simply have decided to respond on the deemed “inappropriate” (e. foods that they were being asked to evaluate (cf. Hoegg and Alba 2007a). Percept.. the participants in these earlier studies of multisensory So. it is entirely possible that the participants in these the notion of appropriateness).. other. a (or taste) on the basis of gustatory. 2009 on Hence. to date. orange. in and of themselves. The participants in this Note that we do not wish to argue that decisional biases in the study had to try and identify 16 different sequentially presented context of food colorings’ influence on multisensory flavor perception beverages created by fully crossing the factors of flavor (cherry-. This is a particularly important issue One means by which researchers have attempted to given that it means that the participants in the majority of circumvent (or at least reduce) this potential uncertainty regarding the cause of color’s influence on flavor identifi- cation responses is by explicitly informing their participants Table 3 Partial summary of the results from DuBose et al. green as compared to no-lime-flavor responses when the the obviously changing color of the drink from one trial to drink was colored red instead (see Table 3 for a summary of the next. the participants identified in the literature documenting the visual capture. (1980). or lime-flavored. That is.. Certain found it difficult to discriminate the identity of the flavor color–flavor pairings were deemed “appropriate” (e. are not. interesting. while others were cues. 2003 for a similar argument). . with the majority of other studies that have investi. They most certainly are. olfactory. That responses (see Partan and Marler 1999. What is more. drink was colored red. often made what could be classed as visually dominant or dominance. lemon. cf. In other words.5 Hall 1958. (1980) of the potential deception prior to the start of the study (see (experiment 2) Stillman 1993. the results Reported flavor Color of drink nevertheless confirm the suggestion that food coloring can Red (%) Orange (%) Green (%) still affect people’s flavor identification responses under such conditions. 2008). Orne 1962). a genuine crossmodal perceptual effect (i. apricot. 2001. (1980) conducted one fre. or no flavor). grape. Bertelson participants’ incorrect answers often seemed to be driven by and Aschersleben 1998 for a similar confound that has been the colors of the drinks themselves. point here is rather that the decisional biases that are elicited not colorless). is that the participants were not some unknown combination of these two effects (see below informed of the deception that was taking place (e. Posner et al. 26% of the participants reported that a flavor perception may have felt some pressure to respond in cherry-flavored drink tasted of lemon/lime when colored line with the salient information provided visually (namely.). flavor identification may simply have assumed that the quently cited study addressing this question. see for a fuller discussion of this issue). as when the lime-flavored basis of the more easily discriminable color cues instead. DuBose et al. p. green. and. Our orange-. informative (see also Garber et al. over perceived auditory localization). orange. fact.g. the 310 untrained observers Cherry 70 41 37 (visitors to an open day at the University of Auckland) in a Orange 0 19 0 study reported by Stillman (1993) were each given a Lime 0 0 26 The results highlight the profound effect that food coloring can have 5 on participants’ flavor identification responses. The table highlights the distribution of responses from Our concern is that the results of such studies may say more about the three most common flavor responses for the cherry-flavored drink. colors of the drinks were meant to be informative with ipants in their study attempted to identify the flavors of a regard to their likely flavor. taken this precaution. but rather by the responses (including 12 fruit flavors) to choose from when trying to ecologically invalid context in which the participants in these identify each of the drinks (strawberry. Although only a few studies have. cherry. lime. color’s influence on flavor identification responses reflect a One potentially important limitation with regard to the response bias elicited by the clearly visible (and changing) interpretation of the study of DuBose et al.e. 229).74 Chem. blueberry. is. (2010) 3:68–84 In terms of studies looking at color’s effect on flavor the studies that have investigated the effect of color on identification. may not be especially apple. Kanig 1955. found that studies may simply have used the salient visual cues as a participants misidentified the flavor of a number of the cognitive shortcut with regard to the likely flavors of the drinks when the coloring was inappropriate. For example. or 1936. Hall 1958.

yellow. Hoegg and Alba 2007b). of any studies that have reported results that conflict with or red than when it was presented as a colorless solution. orange yellow. 3). this null result may simply reflect a were significant (asterisk) using independent samples chi-square tests lack of statistical power.4 reflect a decisional effect. the results of that atypically colored drinks were identified by their color Zampini et al. where the color of a food was (noticeably) The participants in another study reported by Zampini changed but where this change had no effect on partic- et al. 2008) in which the participants had to try and identify biasing of participants’ responses on a certain proportion of the flavor of blackcurrant. DuBose et al. responses were based on color on 13% of trials and on flavor on 79% of trials.e. a perceptual effect. 1980. The drinks were either raspberry. 2) showed that participants were significantly of a food/drink can exert a powerful influence on people’s better at correctly identifying the raspberry flavor when the flavor identification responses (e. 2007. DuBose et al. for example.or orange-flavored and were either colored red. it is also important to bear in mind that informative (cf. Engen 1972). 2 Summary of the results of Stillman’s (1993) study in which were not significantly affected by the changes that were introduced 310 participants were given a single drink to taste and identify.6 Red effects on participants’ flavor identification responses Orange 0. or gray solution. That is. or else was presented as a colorless solution. flavor identification responses are concerned. Similar flavor identification responses do not always match the color results were also reported in a follow-up study (Zampini that they see. Experiment 2) study in which participants’ flavor identification responses for the strawberry-flavored solutions Fig. at least as far as the effects of food coloring on people’s or as a red. (1980) showed colorless (see Fig. 4). 2008. to the claim that the crossmodal effect of color on multi. What is more.Chem. the participants in Stillman’s (1993) study were sensory flavor perception in humans does not reflect able to correctly identify the flavor of the drink in 60% of task demands or any simple form of decisional bias (cf. Shankar et al. Even though Zampini et al. then. yellow. Taken together. or else left uncolored. drinks that they were tasting were not in any way Here. with the exact proportion varying from one study that had been colored yellow. this claim—i. they were “the literature on the effects of color on taste and flavour informed that each of the 124 drinks that they had to judgments is consistent in its inconsistency” is incorrect for sample was just as likely to appear as a colorless solution. Oram et al. 284) claim that of the drinks that they had to evaluate. The orange-flavored solution was identified significantly Zampini et al. experiment 2) were also explicitly informed ipants’ flavor identification responses. their study. (1995) reported that for adults tested in Delwiche 2004. That is. The into the colors of the drinks.g.’s appears clear-cut: food coloring most certainly does participants were clearly aware that the colors of the influence people’s flavor identification responses. So. Perceptual vs. The results have provided empirical support for the claim that the color (see Fig.8 Uncoloured unequivocally) on the question of whether these crossmodal 0. Kanig 1955. (sometimes known as ‘capture’). researchers typically report a visual et al.’s (2007. gray. .. 1995. many studies over the last 70 years or so yellowish orange. (2010) 3:68–84 75 beverage to taste and identify and were informed that its Interim Summary color was independent of the flavor. orange. drink was colored red than when it was colored green. Meanwhile. 2008) studies therefore add weight on 40% of the trials and by their flavor on 28% of the trials. Instead. we would that there was no relationship between the color and flavor argue that Lavin and Lawless’s (1998. the colors of the drinks one does not tend to see complete visual dominance nevertheless still exerted a significant influence on parti. p. In summary. orange. Decisional Contributions to Color’s Influence on Flavor Identification * 1 * Proportion correct None of the studies reported thus far bear directly (and 0. Percept. trials. or some Green 0. We discuss possible reasons for this discrepancy below. blue. though. and flavorless solutions the trials. 2008). Stillman 1993. However.. participants’ cipants’ flavor identification responses (see Fig. the answer regardless of its actual flavor. Oram et al. given that the numerical trends horizontal lines indicate those comparisons between conditions that were in the appropriate direction. Hall 1958. 2009. or else left to the next.’s (2007. orange. we are not aware more accurately when the drink was colored orange. green. (2007. Moir 1936. red.2 0 6 Raspberry Orange The one exception to this claim comes from one of the conditions in Flavour of the solution Zampini et al. green.6 As such. Levitan et al.

Green. green for lime. Percept. influence on participants’ flavor identification responses Green and Swets 1966) when participants are allowed to (see also O’Mahony 1992). there is no obvious explanation specific flavor they had in mind) and also rate the flavor intensity. Red. for this particular result. One problem here in biases that may be present (see Chen and Spence 2010. color appeared to have less of an impact on the identification indicated the “other” option they were prompted to suggest the of the strawberry solutions. make unconstrained flavor identification responses. Red. number of correct flavor identification responses for the strawberry- tion. terms of trying to answer this question is that there are no Intraub 1984). and orange for orange) led to a modest increase in the (experiment 2) documenting the effect of color on flavor identifica. Orange. orange. and flavorless drinks were presented and orange-flavored drinks (relative to the performance seen when either colored green.g. (2010) 3:68–84 a) 100 LIME FLAVOR Correct responses (%) 75 50 25 0 Green. Coloring the solutions congruently (e. strawberry. no one has as yet used them in order to try simple means of assessing perceptual sensitivity (such as and isolate any genuinely perceptual contribution to color’s the d′ measure that is afforded by signal detection theory. Orange. red. At present. Orange.(when colored was added to the solutions. While it is certainly true that techniques do published to date (no matter whether using a free respond- exist to try and counteract the contribution of any guessing ing or forced choice discrimination task) have shown a . Red. or colorless in a fully counter. 3 Summary of the results of the study of Zampini et al (2007) for strawberry.76 Chem. Lime. Colorless standard double standard double standard double Color of the solutions Fig. participants judged the colorless solutions). Incongruent coloring balanced design. Red. red coloring added to the various solutions unknown combination of the two. flavor of each drink from a list of 22 alternatives (if the participants Overall. all that can be said with any confidence at the happens when one is trying to identify a food’s flavor in present time is that all of the studies that have been natural settings. Green. The participants had to try and identify the green or red) and lime-flavored solution (when colored orange or red). Either a standard or double intensity of the colorant impaired the participants’ ability to identify the orange. Colorless standard double standard double standard double Color of the solutions b) ORANGE FLAVOR 100 Correct responses (%) 75 50 25 0 Green.. Note also that it is difficult to discern any They were given 124 samples to evaluate in total using a sip-and-spit clear pattern of results from the doubling of the intensity of the food method. orange. as often As such. Orange.

2009. changing a food or drink’s color (cf. cation responses. 2006.g. Red. Österbauer et al. Colorless standard double standard double standard double Color of the solutions Fig. Green.Chem.g. level may be assisted by the evidence emerging from cognitive neuroscience research. Such combination of the two effects will only be resolved by effects. Percept.. Whether this reflects a genuine perceptual McClure et al. Small 2004. 2006. Red. González et al. 2005. Given that such top-down effects have recently to the auditory modulation of visual brightness judgments. Green. been observed on the basis of verbal food/taste descriptors see also Lau et al. we are answer to the question of whether or not color’s crossmodal optimistic that they might also be found here (that is.g.. 2003 for such an approach being applied perception. Red. 3 (continued) crossmodal effect of color on participants’ flavor identifi. . component to the crossmodal effect of color on flavor see Odgaard et al. 2004. Orange. together with any other contextual cues (e.. that color. Progress here will likely come through the consistent with the existence of at least some perceptual use of more sophisticated psychophysical paradigms (e. (2010) 3:68–84 77 c) 100 STRAWBERRY FLAVOR Correct responses (%) 75 50 25 0 Green. would certainly be future research. effect. or some unknown Veldhuizen et al. It would. Red. Colorless standard double standard double standard double Color of the solutions d) FLAVORLESS 100 Correct responses (%) 75 50 25 0 Green. Nitschke et al. 2006). Orange. 1995). Orange. Orange. for the influence on flavor identification occurs at a perceptual case of visual color cues). It is also possible that a definitive (e. be interesting to know whether the neural activity in primary Expectancy-Based Effects of Food Coloring (or possibly secondary) gustatory cortex and perhaps more interestingly in orbitofrontal cortex elicited by a gustatory/ It seems likely that whenever we see a food of a certain olfactory/flavor stimulus can be changed simply by color. Sarinopoulos et al. should they be observed. Verhagen and Engelen 2006). for example. a predominantly decisional effect.

a sip-and-spit method was used Urbányi 1982. see also Scanlon 1985). 1984a. Morrot et al. (2007) (experiment 1) assessing the flavor expectations generated by a group of UK etc. 2001.e. yellow. or left colorless Koch 2003. pear (5%). Here. Misidentification at this (cognitive) level will Correct responses (%) presumably also provide access to a wealth of semantic 75 information about the “misidentified” flavor (see Engen 1972. Zellner and Durlach 2003). describes as “visual appearance” participants on being presented with clear plastic beakers containing cues). cream soda (5%) simply report (without tasting) what flavor they would Gray Blackcurrant (53%). 2008. Fig. it is interesting to note that red. red. Plassmann et al. Koch and Koch 2003. 2005. e. also happens to be one of the colors that are most often reported as having an effect Correct responses (%) 75 in terms of influencing flavor. aniseed (15%). red. Stillman 1993. raspberry (9%). Williams et al. 25 which in turn might reasonably be expected to influence participants’ judgments/responses.g. Koch and gray. 1995. melon (11%) participants (all from the UK and of UK origin) a range of Orange Orange (91%). Morrot et al. b. 2009. Regarding the expectancy-based account of color’s 25 influence on multisensory flavor perception. Once again. cherry (27%) regarding a drink’s likely flavor. Levitan et al. 2005). Zellner et al. also be noted that different colors will likely lead to the generation of different taste/flavor expecta- 0 Yellow Grey Orange Red Colorless tions by different people as a function of their background Color of the solution and/or culture (Shankar et al. aniseed (5%). Pangborn et al. the likely flavor of that food (see Cardello 2007. This misidentification 0 Yellow Grey Orange Red Colorless may then have perceptual (i. Stevenson and that they expected a particular colored drink to taste of a certain flavor . raspberry (27%).. (2008) expertise/experience with tasting a particular foodstuff highlighting the crossmodal effect of color on participants’ flavor (e. Table 4 Results of the study of Zampini et al. Smith 2007. 2006. and flavor (e. Skrandies and Reuther 2008. toffee (4%) colored drinks (green. melon (2%). 2009.78 Chem. Hyman 50 1983. Oram et al. 4 Summary of the results of the study of Zampini et al. orange. expect it to have. 2002. Parr et al. 1963. 1996.g. cream soda (16%). 1984a. apple (20%).. Percept. (2007) (experiment 1) recently investigated the nature of these crossmodal associations by showing Green Lime (69%). as well as decisional) Color of the solution consequences (see also Williams et al. taste. see also Demattè et al. yellow. licorice (40%). Jönsson et al. identification responses. pear (2%) color-induced flavor expectations (what Hutchings 1977 called “anticipatory” effects) may lead to the “misidentification” of The table shows the percentage of trials in which participants reported flavors when a drink is subsequently tasted (cf. transparent or opaque. 2001. Color of the solution odor. 2).. It is our belief that such Colorless Flavorless (51%). 50 2008. however. and odor perception (in terms of both identity and intensity judgments. b). will lead us to generate specific expectations regarding transparent drinks that had been colored green. and Yellow Lemon (89%). drink expectation) Zampini et al. blue. hot or cold. one of the colors that have often been shown 0 Yellow Grey Orange Red Colorless to generate strong expectations concerning a food’s taste. perceptual) effects (de Craen et al. 2003. Johnson and Clydesdale 1982. Revonsuo 1999. O’Mahony 100 Flavorless 1983. Parr et al. spearmint (2%). gray. Shankar et Color of Expected flavor (% of participants with that al. It 25 should. orange. see Fig. 2006).. cherry (4%). blue.g. what Francis 1977. 1991). The results (see Table 4) showed that all aniseed (4%) of the colored drinks generated systematic expectations Red Strawberry (46%). (2010) 3:68–84 100 Orange Oaten 2008). 2008. though see also is it a food or drink item. Gottfried and Dolan 2003. apple (4%). 50 Marques 2006. Shiv et al.. Lee et al. vanilla (15%).e. Maga 1974. melon (2%) colorless) and asking them to look at each drink in turn and Blue Spearmint (86%). one 100 Blackcurrant need only think of the placebo effect where a person’s beliefs Correct responses (%) (be they induced by the color of a pill or verbal labeling) 75 have been shown to have surprisingly low-level (i. Yeomans et al.

1) draws a clear distinction between cognitive and textural. The two groups of participants in their study. Nitschke et al. Importantly... multisensory integration. others appear to view cognitive/expecta. found to generate consistent flavor expectations based on Marks and Wheeler 1998. 2004. a critical a change in d′). not stimulus or stimulus attribute can enhance the perceptual all) of the colored drinks (see Table 5). (2009) recently investigated the nature of results therefore raise the possibility that color-induced any cross-cultural differences in color–flavor expectancy expectancy effects might also influence a person’s percep- effects. tion in humans (automatically) in a bottom-up manner via level (e.. identification) by directing a each colored drink to taste of. Fig.e. down cognitive influence (in the form of expectancy While some researchers appear to believe that such effects). others clearly think that flavor cues in a food or drink with an atypical visual color color’s influence on flavor is perceptual in nature (e. McClure et al. and/or branding as tion. representation of that stimulus or stimulus attribute (i. and visual cues in the orbitofrontal cortex of both perceptual effects. Numerous neurophysiological studies have different things. the flavor 1995). see cue would be expected to lead to a form of sensory (i.e. Shiv et al. Österbauer tion effects as being perceptual (as opposed to physical. then adding a 1992).g. 1976. green. However.. Taken together. et al. or hedonic. Rolls et al.. gustatory. see Cardello 1996. 2005. Sarinopoulos et al. Ashkenazi and Marks 2004. and age (Oram et al.. Wilson and Klaaren (see Table 5. Engen 1972. 2004.g. see also Lavin and Lawless 1998). 1995).g.. food or drink’s flavor or aroma. Cardello 1994. 1995. For example. no one has. though. pricing. Ernst and Bülthoff 2004. 1). (Fig. simple. physiological/perceptual) consequences (e. Marks 2002. Posner et al. were shown seven response that they happen to make. Rolls and Baylis 1994. orange. McCullough et al. 2006.g. Percept. one tual representation of flavor information and not just the from the UK the other from Taiwan. 2007. Color-induced expectancy effects might influence flavor and red) and had to report on the flavor that they expected perception (in particular. 2008. Fig. 2006. 1). the participants in each group were stimulus (e. Plassmann et al. cf. one might expect this approach to understanding forms of expectancy effect. Research elsewhere in psychology has shown that expectations generated by the two groups of participants the focusing of a participant’s attention on a particular were significantly different for several (but importantly. e.g. Given the individual differences in the flavor expect. multisensory perception to have an influence here as well Hutchings 1977. however. food coloring can also influence flavor identifica- cognitive effects of color on flavor will not have any low. monkey and man (see Gottfried and Dolan 2003..Chem. by influencing people’s criteria for making individual do indeed lead to the misidentification of a various judgments. Oram et al. It is that the Bayesian approach is currently proving so therefore important to note that expectancy effects (here successful elsewhere in the study of multisensory integra- one can also think of labeling. Discussion here though is visual) dominance over flavor perception (see Ernst and complicated by the fact that different researchers appear to Banks 2002. 1995. Cardello 2007. Philipsen et al.. Fig. Such Shankar et al. Marshall et al. 1). Our prediction is that if 1978). while Delwiche (2004) already demonstrated the integration of olfactory. the research reviewed in this section brown color to a drink should result in different flavor therefore supports the view that “higher level” (or top- (aroma) identification responses for these two individu. can have an effect on perceptual sensitivity as measured by ations generated by viewing a food’s coloring. Such decisional effects are likely to be especially one person expects a brown drink to taste of cola while prominent when the stimulus itself is ambiguous (as flavor another expects a drink of the same color to taste of grape stimuli often are. humans (e. It is. (2007).g.. terms of trying to separate top-down (i. 2006). use the same terms (e. measured their participants’ flavor (or for that matter taste or odor) expectations and then followed through to investigate whether those The Multisensory Integration of Color/Flavor Cues expectations do indeed predict the subsequent pattern of flavor (taste or odor) misidentification that results when One of the most important issues for future research that color is added to a foodstuff (cf.e. though see Davis 1981) soon.. Just as in the study of participant’s attention to a salient component of a flavor Zampini et al. 1996). While this experiment sounds (and indeed is) very low-level (early) effects.e. as yet. Delwiche 2004.g. cf. The multisensory integration of see Keast et al. the colors of the drinks. submitted). clear.g. the problem that one soon runs into in have been shown in neuroimaging studies to have surpris. Given sensory. Chaiken 1980. 2005.. “perceptual”) to mean very Spence 2008). concerns the question of whether in addition to any top- Shankar et al. see also Oram et al. down) cognitive beliefs or expectations can (and do) have als. yellow. (2010) 3:68–84 79 Teerling 1992). stimulus-driven) influences is that many of .. cognitive) and ingly early effects on sensory information processing in bottom-up (i. blue. important to bear in mind that experiment that has yet to be conducted would involve color cues can (and do) sometimes lead to decisional effects investigating whether the flavor expectations held by an as well (e. colored drinks (brown.

yellow. blue. be ideal to utilize the articulatory Isolating the Bottom-Up Effects of Color on Flavor suppression technique in order to assess whether one can Perception eliminate (or at least modulate) color’s crossmodal effect on flavor identification. it speaking out loud. “automatic”). however. red. or else left colorless. it may be possible to isolate any to that reported by Stevenson and Oaten (2008). It would. automatic) multisen- stance. it would bottom-up effects of color on flavor perception by utilizing provide evidence that color’s influence on flavor perception articulatory suppression to minimize the effect of top-down is mediated primarily by cognitive (or top-down) expectan- cognitive factors on participants’ performance. for example. perception as a function of whether an odor is delivered Stevenson and Oaten demonstrated that the crossmodal orthonasally (as in Stevenson and Oaten’s study) or influence of color (either appropriate or inappropriate) on retronasally (as when actually consuming a food or drink). For in. of course. . The cherry and important to note that Koza et al. see Navarra et al. it is orthonasal olfaction is automatic or not. (2010) assessing the flavor expectations elicited by differently colored drinks in two groups of participants.. 2009. Stevenson and Oaten (2008) recently used this sory integration.7 If one were to obtain a similar result In the future. one may also be able to dissociate when their participants engaged in articulatory suppression (saying the word “the” out loud repeatedly) while evaluat- ing the odors.. odor discrimination performance was significantly reduced Alternatively. Percept. and orange) that elicited significantly different flavor expectations from the two groups of participants the factors that are likely to promote top-down cognitive research in other areas of psychology has shown that effects are also likely to increase the likelihood of bottom. However. one from the UK. One might think such a result likely given technique in order to investigate whether color’s effect on Stevenson and Oaten’s use of food odors.e. multisensory integration tends to be relatively automatic up multisensory integration as well (see Spence 2007 on (i. This result suggests that the crossmodal 7 Though.g. One could though instead ask the participant to read out a list of words silently in their head or else to engage in some provides some evidence against color influencing odor other highly attention-demanding task. (2005) have shown that strawberry olfactants used in this study were dissolved in color can have qualitatively different effects on olfactory liquids that were colored green. see Santangelo and Spence 2008). As such. there may be some practical challenges effect of color on odor perception is not mandatory (or.80 Chem. (2010) 3:68–84 Table 5 Results of the cross-cultural study of Shankar et al. The shaded rows indicate the colored drinks (brown. cy factors and not by bottom-up (i. the other from Taiwan The table highlights the three most commonly expected flavor responses (chosen by more than one participant). though. unaffected by the performance of a secondary task.. of course. monitoring a perception through a process of bottom-up multisensory rapid serial visual presentation stream for occasionally presented integration (perhaps leading to sensory dominance) since targets (e. this point). in associated with trying to taste a colored solution while at the time Stevenson and Oaten’s words. Santangelo and Spence 2008 for reviews).e.

Chem Senses 31:531–538 Desor J. MacKinney G (eds) Color down manner via cognitive expectations and in a bottom-up in foods: A symposium sponsored by the Quartermaster Food manner via the sensory dominance resulting from the and Container Institute for the Armed Forces Quartermaster automatic (i. Roos PJ. (such as red being associated with sweetness in ripe fruits In: Meiselman HL. 2006). Army Quartermaster the various unimodal sensory signals. Food Sci integration tends to be more sensitive to spatiotemporal Technol Int 7:399–404 Bertelson P. Peterson MS. Polet IA. the question of how best to the dog: semantically-congruent sounds modulate the identifica- unravel the various mechanism(s). flavor and texture represents an important and challenging area for study in judgments of foods. Maller O (1980) Effects of colorants and Ashkenazi A. might. Leonard de Vries A. Bottom-up perception of sweetness and fruit flavor of fruit drinks.Chem. flavor intensity and acceptance of foods by young adults and Having demonstrated that food coloring does indeed elderly adults. Nutr Food Sci 5:24–28 food coloring on flavor identification operates in both a top. New York. smell (orthonasal or retronasal). 4–16 further here though is the fact that the relative contribution Cain WS (1979) To know with the nose: keys to odor identification. Physiol and responses toward. Complicating matters Corps. Pangborn 1960). Spence C (2008) The multisensory perception of flavor. 1415 . Beauchamp G (1974) The human capacity to transmit olfactory information. pp 223–261 recent introduction of the association between blue coloring Chaiken S (1980) Heuristic vs systematic information processing and the use of source versus message cues in persuasion. J Am Diet Assoc 97:657–659 Chen Y-C. Downloaded from http://www. of top-down and bottom-up factors might also vary as a Science 203:467–470 function of the specific combinations of color (and other Cardello AV (1994) Consumer expectations and their role in food visual appearance cues. Levitan et al. Marks LE (2004) Effect of endogenous attention on flavorants on identification. Thomson DMH (eds) Measurement of food preferences. By product development. J Pers Soc and raspberry flavoring (see Garber et al. Alley TR (1998) The influence of physical state and color 02dolnick. Duran L (2001) Influence of color on color and the flavor (see also Lee et al. of course. and flavor being investigated. Sanabria D. pp.. London. or directly on the and of their effectiveness. McFie HJH (eds) Food choice. Food Qual Prefer 15:137–146 core attribute of flavor perception. 2008) acceptance. Part of the challenge here comes from food intensity in young and elderly adults. What is. Calvo C. Spence C (2006) Cross-modal associations between odors and colors. for example. Perception 14:755–762 the fact that (as shown in Fig. Wagner JR. Blackwell L (1995) Visual clues and their effects on odour It may. Spence C (2010) When hearing the bark helps to identify influence flavor identification.nytimes. In: MacFie HJH (ed) Consumer-led food contrast. and texture. Shankar et Psychol 29:752–766 al. though.National Research Council: Washington. Psychon Bull Rev 5:482–489 cognitive effects (see Spence 2007). Cardello AV. perceived flavor intensity. (2010) 3:68–84 81 bottom-up multisensory integration effects from top-down Auvray M. Blackie Academic & Professional. acceptance and green with tartness/sourness in unripe fruits. pp 1–82 1974. Aschersleben G (1998) Automatic visual bias of coincidence of the constituent signals than are top-down perceived auditory location. Kleijnen J (1996) Effect of colour of drugs: systematic review of perceived effect of drugs on taste. odor. In: MacFie HJH. Francis 1977. Engelen L et al (2004) Amount of ingested time is that color plays an integral part in our experience of. In Farrell KT. visual cues do not constitute a flavor. underlying this crossmodal effect clearly Christensen C (1983) Effects of color on aroma. the more pp 253–297 that a color is associated with a particular taste or flavor Cardello AV (1996) The role of the human senses in food acceptance. certain at the present de Wijk RA. September 2. the product development.e. National Academy of Sciences . Blackie Academic & Professional. Br Med J 313:1624–1626 flavor percept itself. J Food Psychophys 66:596–608 Sci 45:1393–1399. 2008. 2010. and detection of weak gustatory and olfactory flavors. On the one hand. the stronger one might expect that Cardello AV (2007) Measuring consumer expectations to improve food both the bottom-up and top-down effects should be. Woodhead Publishing. be Chan MM. at least initially. J Food Sci 48:787–790 Christensen C (1985) Effect of color on judgments of food aroma and the years to come. food and drink even if. Cognition 114:389–404 neural level. effects resulting from cognitive expectancies by playing Conscious Cogn 17:1016–1031 Bartoshuk LM (1975) Taste mixtures: is mixture suppression related with the parameters of stimulus presentation. Demattè ML.html?_r=1&scp=1&sq=chocolate%20strawberry% on perceived sweetness. S. Kane-Martinelli C (1997) The effect of color on perceived expected to have a primarily cognitive effect. 1) it is currently unclear Davis RG (1981) The role of nonolfactory context cues in odor whether color’s crossmodal influence on flavor perception identification. Cambridge. custard dessert as affected by its color. such as by to compression? Physiol Behav 14:643–649 varying the timing and/or spatial relationship between the Bayarri S. Brice BA (1954) Color and its relationship to food investigations. as suggested Behav 82:397–403 Delwiche J (2004) The impact of perceptual interactions on perceived by the ISO (1992. Maga and consumption. Percept hedonic quality of fruit-flavored beverages and cake. Costell E. J Psychology: Interdisciplinary Applied 20yogurt&st=cse on 07/07/09 132:561–568 DuBose Alley RL. the consequences of marketing. turn out that the crossmodal effect of assessment. Percept Psychophys 16:551–556 References Dolnick E (2008) Fish or foul? The New York Times. at both a behavioral and tion of masked pictures. see also Triplett 1994). Percept Psychophys 30:83–89 in any given situation is being driven by color’s influence de Craen AJ. Percept. 2008). involuntary) multisensory integration of Research and Development Command U.

Janhoj T (2007) Understanding creaminess. for multiple semantics linked to sensory modalities. Blackwell Scientific. Dalton PH. Laing DG. J Consum Res 33:490–498 weak taste stimuli. beer. Clydesdale FM. Bull Psychon Soc 24:203–206 Massachusetts at Amherst Fletcher L. Clydesdale FM (1982) Perceived sweetness and redness in masking techniques on the intensity rating of sweetness of colored sucrose solutions. Food Qual Prefer 9:283– Gottfried JA. Cain WS. Food Field Frost MB. In: A. Koch EC (2003) Preconceptions of taste based on color. Hyatt EM. Little Inc. Ducheny K. Ariely D (2006) Try it. J Food Technol 34 Masurovsky BI (1939) How to obtain the right food color. Perception 31:617–635 relationship between color and salt concentrations in chicken Lau K-N. Clydesdale FM (1986) The psychophysical relationship intensity. you’ll like it: the influence sics. Hekkert Koza BJ. Byram J (1988) Taste–smell interactions are tastant and relationships between perceived sweetness and redness in odorant dependent. the appearance of nondurable consumer goods and their University of Massachusetts. Reinhold. Starr RG Jr (2003) Measuring consumer Psychol: Interdisciplinary Applied 137:233–242 response to food products. J Exp Psychol Learn Mem Cogn Engen T (1972) The effect of expectation on judgments of odour. Hyatt EM. (ed) Flavor research and food Levitan C. Zellner DA (2005) Color enhances P (eds) Product experience.D. Hutchinson I (2002) The limited systems. Nature 415:429–433 ISO (2008) Standard 5492: terms relating to sensory analysis. Tchekhova A. Clydesdale FM (1982) Francis FJ (1977) Color and appearance as dominating sensory Psychophysical relationships between perceived sweetness and properties of food. but not red Jönsson FU. Bull Psychon Soc 21:145–148 Chem Senses 31:539–545 Institute of Food Technologists (1980) Food colors. Food Qual Prefer 14:3–15 Kostyla AS (1978) The psychophysical relationships between color Garber LL Jr. Applied Science. Cilmi A. Meseguer V. Mem Cogn Parker KJ (eds) Sensory properties of foods. Dzendolet E. 14:371–377 Chem Senses 30:353–365 Frick RW (1995) Accepting the null hypothesis. Chem Senses 23:19–29 Hutchings JB (1977) The importance of visual appearance of foods to Marques JF (2006) Specialization and semantic organization: evidence the food processor and the consumer. Ellersieck M (1991) Effects of visual Johnson J. Damon RA Jr (1987) The psychophysical and taste–odour mixtures. Heymann H. Ph. Spence C (2008) Assessing the role of acceptance: a survey of the scope of flavor and associated color cues and people’s beliefs about color–flavor associations on research. Applied Science 34:60–67 Publishers. tests. Barking. Amherst packaging on consumer behavior. J Food Prot 46:21–25. Belloch V et al (2006) Reading cinnamon activates Lavin J. Dolan RJ (2003) The nose smells what the eye sees: 289 crossmodal visual facilitation of human olfactory perception. Food Eng (7):77–84 11:13. consumption. Mem Cogn 23:132–138 Kanig JL (1955) Mental impact of colors in foods studied. J Sens Stud 2:137–147 distinguish perception bias from discrimination ability in taste González J. Alba JW (2007a) Linguistic framing of sensory experience: Maga JA (1974) Influence of color on taste thresholds. In: Birch GG. Lawless H (1998) Effects of color and odor on judgments of olfactory brain regions. Wiley. Starr RG Jr (2001) Placing food color sensory level. Chem Senses 13:445–455 strawberry-flavored beverages. Oxford. Frederick S. Breslin PAS (2004) Flavour interactions at the Garber LL Jr. Ernst MO. Wheeler ME (1998) Attention and the detectability of tongue. Sawyer M. J Sens Stud 6:179–191 Johnson JL. Roberts DD (eds) Flavour experimentation into a valid consumer context. D. Kagan A (1995) Using economic incentives to flavored broths. enhances the sweetness of sucrose solutions. NeuroImage 32:906–912 sweetness among children and adults. thesis. Barros-Loscertales A. Swets JA (1966) Signal detection theory and psychophy. New York of expectation. Olsson MJ (2005) A color. J Food Sci 47:747–752 gelatins and lemonades. University of aptitude. Mize S (1989) Strawberry odor. Alba JW (2007b) Taste perception: more than meets the Marks LE. Damon R. Chem Senses 30:643–649 between color and sodium chloride concentrations in model Laing DG. J Garber LL Jr. pp 224–240 Senses 33:415–423 Hoegg J. London. PhD thesis. Banks MS (2002) Humans integrate visual and haptic International Organization for Standardization information in a statistically optimal fashion. Marks LE (2002) The role of attention in chemosensation. Am J Psychol 52:255–265 events on memory for pictures. Dolese M. Lee L. Bülthoff HH (2004) Merging the senses into a robust International Organization for Standardization percept. Methods Psychological Research 6:1–32 Green DM. Link C. Lönner P. Jinks AL. Trends Cogn Sci 8:162–169 Johnson J (1982) Psychophysical relationships between colour and Eskenazi B. Zampini M. Lecoutre B. In: Taylor AJ. Percept. pp 45–47 Johnson JL. Acta 10:115–125 Psychol 36:450–458 ISO (1992) Standard 5492: terms relating to sensory analysis. Hutchinson I (2006) The capacity of Hyman A (1983) The influence of color on the taste perception of humans to identify components in complex odor–taste mixtures. Psychol Sci 17:1054–1058 Inc. Chem given in 1956–1957. Parker KJ (eds) color in cherry-flavored beverages. Jinks AL. Elsevier. J Food Prot 45:601–606 Sensory properties of foods. J Food Prot 49:977–982 capacity of humans to identify the components of taste mixtures Gifford SR. New York. compiled from papers presented in a series of symposia the discrimination of the flavor of sugar-coated chocolates. London. Brennan JG. Dzendolet E. Chem Senses metamemory perspective on odor naming and identification. Derzko G (2001) Asserting the smallness of effects in Neuron 39:375–386 ANOVA. Food Qual Erlbaum. Ernst MO. carbonated water preparations. 28 Frank R. Chem Senses there is some accounting for taste. Mahwah. and revelation on preferences for Hall RL (1958) Flavor study approaches at McCormick and Company. pp 581–602 orthonasal olfactory intensity and reduces retronasal olfactory Gifford SR. J Food Prod perception. Post G. (2010) 3:68–84 Duncker K (1939) The influence of past experience upon perceptual Intraub H (1984) Conceptual masking: the effects of subsequent visual properties. pp 3–21 Prefer 14:147–155 Hoegg J. 55–56 . Pulvermüller F. Hyatt EM. In: Lowrey TM (ed) Flavor 1:115–119 Psycholinguistic phenomena in marketing communications. Li R.82 Chem. Brennan JG. pp 45–57 Marshall K. Boya ÜÖ (2008) The mediating effects of and flavor of some fruit flavored beverages. Clydesdale FM (1983) Psychophysical Frank RA. J Mark Res 32:140–151 Sanjuán A. pp 228–255 Mark 7(3):3–24 Koch C. In: Schifferstein HNJ. In: Birch GG. Int Dairy J Reporter 23:57 17:1298–1311 Keast RSJ. Friend K (1986) Exploration of olfactory sweetness in fruit flavored solutions.

J Food Sci 58:810–812 Partan S. Heatherbell D. Spence C (2008) Is the exogenous orienting of spatial Fusion 11:4–11 attention truly automatic? Evidence from unimodal and multi- Nitschke JB. Spence C (2009) Assessing the Poster presented at ISOT 2004 role of attention in the audiovisual integration of speech. Am J Psychol inappropriate stimulus color on odor discrimination. Stern P (1995) Consumer Teerling A (1992) The colour of taste. J Mark Res XLII:383–393 by children and adults. Shankar MU. Conscious Cogn 17:989–1015 EE et al (2006) Altering expectancy dampens neural response to Sarinopoulos I. Marler P (1999) Communication goes multimodal. Hansen B (1963) The influence of color on Stevenson RJ. Klein RM (1976) Visual dominance: an Urbányi G (1982) Investigation into the interaction of different information-processing account of its origins and significance. Chem Senses 17:886 age affects response to sensory characteristics of a cherry Tom G. Bloomsbury. pp 484–485 discrimination of sweetness and sourness in pear-nectar. Inf Santangelo V. Dev Psychobiol 28:239–246 Skrandies W. J Food Sci 60:364–368 the role of salient cues in consumer perception. Freeman M. Carmon Z. Dixon GE. Acoust Sci Pangborn RM (1960) Influence of color on the discrimination of Technol 28:61–70 sweetness. Conscious Cogn 8:173–185 Valentin D. Brochet F. The effect of Psychol Rev 83:157–171 colour upon the evaluation of taste in fruit and vegetable Revonsuo A (1999) Binding and the phenomenal unity of conscious. Am J Stevenson RJ (2009) The psychology of flavour. Trends Neurosci 27:120–123 PC. Sarinopoulos I. Short SJ. Clydesdale FM. 283:1272–1273 Appetite 28:85 Philipsen DH. Critchley HD. and visual conver. Perception 12:627–633 Percept Mot Skills 61:314 O’Mahony M (1992) Understanding discrimination tests: a user. Rangel A (2008) Marketing (2):23–27 actions can modulate neural representations of experienced Triplett T (1994) Consumers show little taste for clear beverages. Radle LJ. Science Strugnell C (1997) Colour and its role in sweetness perception. Montague Rolls ET. Percept. Baylis LL (1994) Gustatory. Martinsen CS. Nissen MJ. Montague LM. Med Sci Res 15:329–330 Brain Behav Immun 20:120–132 O’Mahony M (1983) Gustatory responses to nongustatory stimuli. Rev Interam Psicologia 2:143–157 of stimuli with known attribute structures. Matthews PM. Woodhead Publishing Limited. Dixon GE. Oaten M (2008) The effect of appropriate and discrimination of sweetness in dry table-wine. Moinpour R (1978) Application of Romeu JV. Levitan CA. knowledge mediate the interpretation of multisensory cues? Percept Psychophys 65:123–132 Conscious Cogn (in press) Oram N. Nat Neurosci 9:435–442 Brain mechanisms of expectation associated with insula and Norton WE. J Exp Psychol Learn Mem Cogn 26:423–440 Parr WV. Wakeman EA (1996) Orbitofrontal PR (2004) Neural correlates of behavioral preference for cortex neurons: role in olfactory and visual association learning. olfactory. Clydesdale FM (1988) Psychophysical 63:103–109 relationships between perceived sweetness and color in lemon- Moir HC (1936) Some observations on the appreciation of flavour in and lime-flavored drinks. Cohen JD. Chem Senses 27:747–755 unique. Prescott J. Navarra J. perceptual skill and semantic memory in expert of conditioned odor perceptions: evaluative conditioning is not and novice wine judges. Am J Psychol 73:229–238 Spence C (2008) Multisensory perception. Calvert GA (2005) Color of scents: chromatic stimuli Smith BC (ed) (2007) Questions of taste: the philosophy of wine. White KG (2002) Demystifying wine expertise: Stevenson RJ. Acta Aliment 11:233–243 ness. Lew W. Chrea C.Chem. J Soc Chem Ind: Chemistry Industry Review 14:145–148 Rouanet H (1996) Bayesian methods for assessing importance of Morrot G. Dubourdieu D (2001) The color of odors. properties in the course of sensory evaluation. Johnson FN (1987) The influence of intensity of colour amygdala response to aversive taste: implications for placebo. Neuron 44:379–387 Neurophysiol 75:1970–1981 McCullough JM. Cypert KS. sweet taste perception. Scanlon BA (1985) Race differences in selection of cheese color. Orne MT (1962) On the social psychology of the psychological and color is reflected by electrical activity in the human brain. Beckmann CF. J Consum Mark 4 Plassmann H. Owen J. 11 Posner MI. J Sens Stud 7:1–47 sensory Perception 2:53–58 Odgaard EC. Griffin RW. J Wine Res 14:79–101 flavored beverages. pp 14:5437–5452 66–84 . pleasantness. Rose G. Newell Shiv B. Shiv B. Nitschke JB (2006) aversive taste in primary taste cortex. O’Doherty J. Soto-Faraco S. In: Spillane W (ed) Optimising sweet gence within the primate orbitofrontal cortex. J Neurophysiol Oxford University Press. Ariely D (2005) Placebo effects of marketing actions: G (1995) The influence of flavor and color on drink identification consumers may get what they pay for. Li J. Marks LE (2003) Cross-modal enhancement of Shankar MU. Mason R. Spence C (2009) The influence friendly treatment of response bias. Cambridge. London. odor. Arieh Y. J Food Sci 53(1116–1119):1162 foodstuffs. Short SJ. Oxford Pangborn RM. rating and ranking R-index of color and label information on flavor perception. Jenkinson M. Brain effects. Oxford University Psychol 76:315–317 Press. Tomlin D. In: Blumenthal H (ed) The Pangborn RM. J Appl Psychol Roth HA. Nguyen DH (2006) Taste–odour interactions in Rolls ET. Wilson JP (2000) Resistance to extinction olfactory threshold. Reuther N (2008) Match and mismatch of taste. White KG. Boakes RA. products. Gifford SR. Hansen B (1963) The influence of color on big fat duck cook book. Barnett T. modulate odor responses in the human brain. Hansen senses. Heatherbell D (2003) The nose knows: influence Stillman J (1993) Color influences flavor identification in fruit- of colour on perception of wine aroma. Alsius A. Smith sensory studies. Hutchinson I. I. Laing DG. J culturally familiar drinks. Berg HW. (2010) 3:68–84 83 McClure SM. Chemo- texts and their relationship to signal detection. De Vicente MLS (1968) Influyen los colores en el sabor? multidimensional scaling to the analysis of sensory evaluations [Do colours influence flavour?]. Am Psychol 17:776–783 Small DM (2004) Crossmodal integration—insights from the chemical Österbauer RA. Psychol Bull 119:149–158 Lang 79:309–320 Sakai N (2004) The effect of visual stimuli on perception of flavor. J experiment: with particular reference to demand characteristics Psychophysiol 22:175–184 and their implications. Percept 76:492–495 Psychophys 70:640–646 Parr WV. Proc Natl Acad Sci USA 105:1050–1054 Marketing News 28(11):2. on perceived flavour characteristics. Spence C (2010) Does higher level perceived brightness: sensory interaction versus response bias. Levitan CA. Selmants J (1987) Cueing the consumer: flavoured beverage. Oxford 93:3434–3441 Spence C (2007) Audiovisual multisensory integration. J Neurosci taste in foods. Davidson RJ.

84 Chem. Food Qual assessors. Percept. Whitten LA (1999) The effect of color intensity and behavior. Phillips N. intensity. (2010) 3:68–84 Veldhuizen MG. 18:975–984 behav Rev 30:613–650 Zampini M. Am J Psychol 104:547–561 . Wantling E. Food Qual Prefer 19:565–573 identification and liking ratings. and liking of beverages. Neurosci Bio. Eckard R (1991) Influence of color on odor smoked salmon ice-cream. Blumenthal H. Bakker J (1984b) Effect of Zellner DA. Blake A (2008) The role of Psychol 112:585–604 expectancy in sensory and hedonic evaluation: the case of Zellner DA. affective expectations on affective experiences. Am J 19:659–671 Psychol 116:633–647 Wilson T. pp 1–31 appropriateness on color-induced odor enhancement. Food Qual Prefer multimodal food perception: sensory integration. Sanabria D. In: Clear MS (ed) J Exp Psychol Hum Percept Perform 16:391–397 Review of personality and social psychology: emotion and social Zellner DA. Bartoli AM. Timberlake CF. Klaaren K (1992) Expectation whirls me round: the role of Zellner DA. Engelen L (2006) The neurocognitive bases of human color cues on flavor discrimination responses. J Inst Brew 90:250–253 Prefer 19:335–343 Williams AA. Int J Food Sci Technol experienced refreshment. Spence C (2008) Multisensory Williams AA. Nachtigal DJ. Sage. Spence C (2007) The contributes to odor quality coding. Chambers L. Am J Yeomans M. Chem Senses 34:A8 multisensory perception of flavor: assessing the influence of Verhagen JV. Small DM (2009) Taste cortex Zampini M. Durlach P (2003) Effect of color on expected and colour on the assessment of ports. Phillips N. Newbury Park. Kautz MA (1990) Color affects perceived odor intensity. Langren SP. Langron SP. Noble AC (1984a) Influence of appearance flavor perception: assessing the influence of fruit acids and color on the assessment of aroma in Bordeaux wines by trained cues on the perception of fruit-flavored beverages.