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 concerning the important question of whether or not food color influences taste and flavor perception in humans. Although a superficial reading of the literature on this topic would appear to give a somewhat mixed answer, we argue that this is, at least in part, due to the fact that many researchers have failed to distinguish between two qualita- tively distinct research questions. The first concerns the role that food coloring plays in the perception of the intensity of a particular flavor (e.g., strawberry, banana, etc.) or taste 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 question is currently rather ambiguous. While some researchers have reported a significant crossmodal effect of changing the intensity of a food or drink���s coloring on people���s judgments of taste or flavor intensity, many others have failed to demonstrate any such effect. By contrast, the research findings concerning the second question clearly support the view that people���s judgments of flavor identity are often affected by the changing of a food or drink���s color (be it appropriate, inappropriate, or absent). We discuss the possible mechanisms underlying these crossmodal effects and suggest some of the key directions for future research in order to move our understanding in this area forward. Keywords Flavor . Taste . Color . Perception . Crossmodal . Multisensory. Expectancy. Attention Introduction Does food coloring influence taste and flavor perception in humans? Although researchers have been investigating this important (both on a theoretical and practical level) 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 reached. That, at least, would seem to be the conclusion drawn by the majority of researchers in the field. Take, for example, Lavin and Lawless���s (1998, p. 284) claim that ���The literature on the effects of color on taste and flavor judgments is consistent in its inconsistency��� or Koch and 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.��� Meanwhile, Bayarri et al. (2001, p. 399) have also suggested that ������the possible influence of color on flavor perception is under discussion and no clear conclusions have been attained yet.��� In the present article, we argue that one important reason why our understanding of the nature of any crossmodal effects of food coloring on taste and flavor perception in humans has progressed so slowly relates to the fact that many researchers have failed to C. Spence : M. U. Shankar Department of Experimental Psychology, University of Oxford, Oxford, UK C. A. Levitan Cognitive Science Program, Occidental College, 1600 Campus Road, Los Angeles, CA 90041, USA M. Zampini Center for Mind/Brain Sciences & Department of Cognitive Sciences and Education, University of Trento, Rovereto (TN), Italy C. Spence (*) Crossmodal Research Laboratory, Department of Experimental Psychology, University of Oxford, Oxford OX1 3UD, UK e-mail: charles.spence@psy.ox.ac.uk Chem. Percept. (2010) 3:68���84 DOI 10.1007/s12078-010-9067-z

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

yellow on perceptual thresholds for four of the basic tastes (salty, sour, sweet, and bitter). Note that each of the basic tastes was tested in a separate part of the experiment, meaning that the participants were presumably never uncertain with regard to the identity of the tastant whose presence they were trying to detect. In many cases, Maga observed that the concentration of the tastant had to be increased in order for his participants to be able to detect its presence in the colored (as compared to the uncolored) solutions. So, for example, the addition of green coloring to a sweet solution significantly increased taste sensitivity, while yellow color decreased taste sensitivity (see Table 2). Interestingly, red coloring had no significant effect on sensitivity to sweet taste. With respect to sour taste sensitivity, both the yellow and green coloring of solutions decreased participants��� sensitivity, with red coloring again having no effect. Coloring a clear solution red decreased bitter taste sensitivity, while the addition of yellow and green coloring had no such effect. Finally, adding color had no effect on taste detection thresholds for salt solutions. Johnson and Clydesdale (1982), in an oft-cited study, demonstrated an effect of food coloring on taste perception in sweetened solutions (that sometimes contained a cherry flavoring). The participants in their study had to perform both a threshold task and a magnitude estimation task. In the threshold task, Johnson and Clydesdale found that on average, when odorless solutions were colored red, partic- ipants could more easily detect the presence of sucrose than when they were uncolored (cf. Maga 1974), though the intensity of the color did not have a significant effect on their performance. In the magnitude estimation task, however, Johnson and Clydesdale found that changing the level of food coloring had a significant effect on partic- ipants��� perception of the sweetness of both odorless and cherry-flavored solutions, with the darker-colored solutions being rated as 2���10% sweeter than the lighter-colored Table 1 Summary of the studies that have been published to date that have investigated the effect of varying the presence vs. absence, the appropriateness/inappropriateness, or the intensity, of the color added to a solution on participants��� taste and/or flavor (Fl) intensity ratings. The table highlights the fact that the majority of research in this area has focused on the influence of color on sweetness perception. The table also highlights the inconsistency in the pattern of results that has been reported to date across the various studies that have been reported Study Tastant Flavor Result Pangborn (1960) Sw, So Sig Pangborn et al. (1963) Sw Sig Romeu and De Vicente (1968) Fl Sig Maga (1974) Sw, Sa, So, Bi Sig Kostyla (1978) Sw, So Fl Sig McCullough et al. (1978) Sw Complex DuBose et al. (1980) Fl Complex Johnson and Clydesdale (1982) Sw Sig Johnson et al. (1982) Sw Sig Hyman (1983) Fl Sig Johnson et al. (1983) Sw Sig Gifford and Clydesdale (1986) Sa n.s. Gifford et al. (1987) Sa n.s. Roth et al. (1988) Sw Sig Frank et al. (1989) Sw n.s. Fletcher et al. (1991) Sw Complex Philipsen et al. (1995) Sw Sig Chan and Kane-Martinelli (1997) Sa n.s. Strugnell (1997) Sw Sig Alley and Alley (1998) Sw n.s. Lavin and Lawless (1998) Sw Complex Bayarri et al. (2001) Sw Fl Complex Zampini et al. (2007) Fl Complex Zampini et al. (2008) Sw, So Fl Complex Sw sweet, Sa salt, So sour, Sig significant result, n.s. non-significant result, Complex typically a mixture of significant and non-significant results Table 1 Summary of the studies that have been published to date that have investigated the effect of varying the presence vs. absence, the appropriateness/inappropriateness, or the intensity, of the color added to a solution on participants��� taste and/or flavor (Fl) intensity ratings. The table highlights the fact that the majority of research in this area has focused on the influence of color on sweetness perception. The table also highlights the inconsistency in the pattern of results that has been reported to date across the various studies that have been reported 70 Chem. Percept. (2010) 3:68���84