Taste sensitivity and aesthetic preferences: Is taste only a metaphor?

EMPIRICAL STUDIES OF THE ARTS, Vol. 29(2) 171-189, 2011

TASTE SENSITIVITY AND AESTHETIC PREFERENCES:

IS TASTE ONLY A METAPHOR?

C. NATHAN DEWALL

University of Kentucky

PAUL J. SILVIA

University of North Carolina at Greensboro

DAVID R. SCHURTZ

University of Kentucky

JESSICA MCKENZIE

Clark University

ABSTRACT

“Taste” is often used to describe sensitivity to both foods and visual art.We examined whether a biological marker of physical taste sensitivityinfluenced aesthetic preferences. In three studies, we measured physical tastesensitivity by exposing participants to the chemical phenylthiocarbamide(PTC) and having them rate how bitter it tasted (if they tasted it at all).Across all studies, miscalibrated physical taste sensitivity (extremely highand low taste bud density) related to extreme negative responses to dis-turbing and provocative artwork. Miscalibrated physical taste sensitivitywas related specifically to avoiding (high disgust) disturbing artwork, butnot to approach-related negative affect (anger). These findings provide novelevidence regarding biological influences on aesthetic preferences.

“Taste. You cannot buy such a rare and wonderful thing. You can’tsend away for it in a catalogue. And I’m afraid it’s becoming obsolete.”(Rosalind Russell)

171

! 2011, Baywood Publishing Co., Inc.doi: 10.2190/EM.29.2.chttp://baywood.com

What do people mean when they say that a person has extreme “taste” for visualart? Is taste only a metaphor? Or, instead, is there a link between physiologicaltaste sensitivity and taste for artwork? The current work examined whether agenetic marker of physical taste sensitivity had implications for aestheticpreferences of visual art. Specifically, we tested the hypothesis that miscali-brated physical taste sensitivity, as indicated by extremely low or high densityof taste receptors on the anterior tongue, would relate to extreme aestheticpreferences for artwork.

WHAT IS TASTE?

People frequently use the word “taste” to describe individual differencesin aesthetic preferences and evaluations. Much of this work has focused onhow experts and novices differ in their evaluations of art. For the most part,experts prefer art that is complex, abstract, visually asymmetrical, and aesthet-ically important (Axelsson, 2007; Kozbelt, 2006; Locher, Smith, & Smith, 2001;McWhinnie, 1968; Parsons, 1987). Novices prefer art that is simple, familiar,representational, and easy to process (see Reber, Schwartz, & Winkielman,2004, for a review). Some consider aesthetic taste as a marker of socioeconomicstatus. According to this perspective, members of the elite social class have highlevels of taste, whereas members of the proletariat have low levels of taste(Bourdieu, 1984; Holbrook, Weiss, & Habich, 2004; Kraaykamp, 2002).

What other factors may influence taste for artwork? Psychological aestheticshas become increasingly interested in biological contributions to aestheticexperience (e.g., Skov & Vartanian, 2009), yet to date there is still not muchwork on biological markers of taste for artwork. The current work sought to fillthis gap in the literature by examining how physical taste sensitivity influencesevaluations of art.

Defining and Measuring PhysicalTaste Sensitivity

Physical taste sensitivity refers to the extremity of responsiveness to varioussubstances (Drewnowski, Henderson, Shore, & Barratt-Fornell, 1997; Kalmus,1958). Typically, researchers measure taste sensitivity by exposing people tosubstances that only certain people can taste and measuring the degree to whichtheir reaction to the substances is extreme. One widely used method, employedin our research, involves people tasting a piece of paper treated with the chemicalphenylthiocarbamide (PTC; see Bartoshuk, Duffy, & Miller, 1994, for a review).After tasting the PTC paper, people indicate whether they can detect a taste forthe chemical and, if so, they rate the bitterness of the taste. Using this method,which corresponds to studies in which actual taste bud density is measured(Tepper & Nurse, 1997), researchers group people into three categories according

172 / DEWALL ET AL.

to their taste sensitivity (Bartoshuk, 2000; Bartoshuk et al., 1994; Joiner & Perez,2004; Tepper & Nurse, 1997). Some people react strongly to PTC due to a highdensity of taste buds. These supertasters, as they are frequently called, havehigh density of taste receptors on the anterior tongue (fungiform papillae, tastebuds). People in the second group—commonly referred to as tasters—have inter-mediate taste bud density and rate PTC as only mildly bitter. A third groupconsists of people who have low taste bud density and are “blind” to the bittertaste of PTC that tasters and (especially) supertasters evaluate as negative. Peoplein this group are called non-tasters, since they tend not to taste substances thatother people easily taste. We used this valid and established method for placingpeople into groups—supertasters, tasters, and non-tasters—according to theirsensitivity to PTC.

Taste as Physical and Mental

Can physical taste sensitivity influence outcomes other than sensitivity tosubstances? The neurobiology underlying taste sensitivity involves many differentneural pathways operating simultaneously, resulting in taste sensitivity affectingseveral seemingly unrelated outcomes. Some research, for example, has shownthat PTC sensitivity can be used as a genetic marker for various disordersseemingly unrelated to taste sensitivity. For example, non-tasters, compared tosupertasters and tasters, have a higher probability of having a DSM-IV diagnosisof schizophrenia (Moberg, McGue, Kanes, Roalf, Balderston, Gur, et al., 2007)and Parkinson’s disease (Moberg, Balderston, Rick, Roalf, Weintraub, Kleiner-Fisman, et al., 2007). PTC sensitivity also relates to vulnerability for depressionand alcoholism. Supertasters, compared to non-tasters and tasters, have a lowerpercentage of first-order relatives with major depressive disorder (Joiner & Perez,2004). And supertasters exhibit lower signs of alcohol use problems than doboth non-tasters and tasters (Driscoll, Perez, Cukrowicz, Butler, & Joiner, 2006).These findings suggest that individual differences in taste sensitivity influenceoutcomes other than those directly related to sense of taste.

Taste Sensitivity and Evaluations of Artwork:The Miscalibration Hypothesis

How might physical taste sensitivity influence aesthetic preferences? We pro-pose that the effect of physical taste sensitivity on art preferences depends onwhether people deviate from the standard level of physical taste sensitivity; thatis, if they are miscalibrated. Most people are tasters (Joiner & Perez, 2004; seeStudies 1-3 below), thereby making non-tasters and supertasters non-standardfrom average levels of taste sensitivity. Therefore, people who have miscalibratedphysical taste sensitivity, such as having extremely low (non-tasters) or high(supertasters) levels, should have especially extreme reactions to artwork. Thismiscalibration hypothesis, therefore, predicts a curvilinear relationship between

IS TASTE ONLY A METAPHOR? / 173

physical taste sensitivity and art preferences, with non-tasters and supertastershaving the most extreme art preferences.

Although psychology is typically interested in typical levels of traits, the notionof miscalibrated individual difference traits has been used to predict extremeresponses in various areas of psychology. Of particular relevance to the currentinvestigation, a large body of research has investigated the effects of miscali-brated emotional sensitivity on various outcomes. When predicting extremeaggressive outbursts, for example, people who are extremely emotionally insen-sitive (e.g., people high in psychopathy) or are extremely emotionally sensitive(e.g., people high in neuroticism) tend to behave the most aggressively (Hare,2003; Hellmuth & McNulty, 2008). Similar findings have emerged from theattachment literature (Mikulincer & Shaver, 2007). People with extremely low(i.e., those with an avoidant attachment style) and extremely high (i.e., thosewith an anxious attachment style) levels of need for closeness with others aremost likely to lash out at intimate relationship partners (see Finkel & Slotter,2007, for a review), whereas people who feel secure in their need for closenesswith others—the most common attachment style (Hazan & Shaver, 1987; Vohs,Baumeister, & Ciarocco, 2005) that is marked by intermediate levels of avoidanceand anxiety (Bartholomew & Horowitz, 1991)—show intermediate levels oflashing out at intimate partners.

The implication is that individual differences, including physical taste sensi-tivity, are best understood in relation to a standard or average level, and devia-tions from that standard in some cases relate to extreme responses. Thus, thereis theoretical and empirical precedent for the miscalibration hypothesis asit relates to negative reactions to visual artwork. Non-tasters and supertastersshould have the most extreme reactions to artwork, whereas tasters should showart preferences that fall in between those two groups.

A contrasting possibility is that physical taste sensitivity will have a posi-tive linear relationship to extremity of art preferences. This linear sensitivityhypothesis predicts that supertasters, compared to tasters and non-tasters, willhave extreme evaluations of artwork. Taste sensitivity will thus correlate posi-tively with pleasing works of art and will correlate negatively with disturbingworks of art: supertasters will have the most positive evaluations of pleasingartwork and the most negative evaluations of disturbing artwork.

The Present Research

In three studies, we tested competing hypotheses—the miscalibrationhypothesis and the linear sensitivity hypothesis—by having people complete ameasure of physical taste sensitivity and then evaluate various types of artwork(pleasing, disturbing, or neutral). We predicted that people with miscalibratedphysical taste sensitivity, as indicated by extremely low or high density of tastereceptors on the anterior tongue, would have the most extreme art preferences.

174 / DEWALL ET AL.

In addition, we expected that physical taste sensitivity would influence evalu-ations of disturbing artwork more so than positive and neutral artwork. Negativelyvalenced stimuli have a stronger impact on evaluations than do positive stimuli(Baumeister, Bratslavksy, Finkenauer, & Vohs, 2001). When asked to evaluatea person, for example, negative traits about the person carry more weight thando positive traits (Fiske, 1980; Hodges, 1974; Vonk, 1993). Therefore, any effectof physical taste sensitivity on evaluations of artwork should be strongest fornegative or disturbing art compared to positive and neutral art.

We also expected that the effect of physical taste sensitivity on aestheticpreferences would have the strongest impact among people with relativelylittle expertise in artwork, presumably because people with low expertise havelittle formal knowledge on which to base their judgments (Parsons, 1987;Smith & Smith, 2006). Because experts in the arts have extensive formal knowl-edge of the arts, their judgments are probably more strongly affected by theirtraining and domain knowledge than by low-level sensory processes relatedto taste sensitivity.

STUDY 1

Method

Participants

One hundred thirty-six undergraduates participated in exchange for extra credittoward a psychology course. Following previous procedures (Joiner & Perez,2004), participants refrained from smoking, eating, or drinking for 1 hour priorto participating.

Materials and Procedure

Participants completed the study in a large classroom. After giving informedconsent, participants completed a trait curiosity scale (Kashdan, Rose, & Fincham,2004) and a brief demographic questionnaire. Participants then viewed 44images depicting four types of artwork: disturbing images (e.g., Head Surroundedby Sides of Beef by Francis Bacon and Echo of a Scream by David AlfaroSiqueiros), homoerotic male nudes (e.g., Thomas and Ken Moody by RobertMapplethorpe), black-and-white abstract images (e.g., untitled works by GustavMorin, Spencer Selby, and Andrew Topel), and conventional images with nostrong emotional tug (e.g., Element 2 and Element 10 by Maria Friberg). Mostof these images have been used in past research on aesthetic responses tophotography, visual poetry, and paintings (e.g., Cooper & Silvia, 2009; Silvia,2005a, 2005b, 2010; Silvia & Brown, 2007; Turner & Silvia, 2006). Each imagewas displayed for 5 seconds, during which participants provided their evalu-ation of the image from 1 (extremely dislike) to 7 (extremely like).


After providing ratings of the images, participants completed the positive andnegative affect schedule (PANAS; Watson, Clark, & Tellegen, 1988). Partici-pants were then given an envelope containing a 3.80 × 1.43 cm piece of papertreated with phenylthiocarbamide (PTC; Carolina Biological Supply Company,Burlington, NC). The experimenter instructed participants to moisten their tonguewith saliva and then to place the piece of paper in the middle of their tongue foras long as it took to make an evaluation of any taste that was detected. Participantsplaced the used PTC paper in the envelope and rated the intensity of the bitternessof the paper from 0 (no taste at all) to 9 (among the most bitter things I haveever tasted). This approach to assessing taste sensitivity has been used in severalprevious studies (e.g., Frank & Korchmar, 1985; Joiner & Perez, 2004).

Results and Discussion

Did miscalibrated physical taste sensitivity predict extreme aesthetic prefer-ences? Taste sensitivity can be analyzed two ways: as a continuous variableranging from 0 to 9, or as an ordered-categorical variable consisting of non-tasters(n = 35; score of 0), tasters (n = 61; scores of 1-6), and supertasters (n = 40; scoresof 7-9). We report analyses of the continuous variable; for convenience andcontinuity with other research, the tables and figures display the effects accordingto three groups. Aesthetic preference scores were averaged within each categoryof images (disturbing, homoerotic, black-and-white abstract, and conventional).

We conducted a multivariate regression model with taste sensitivity’s linearand quadratic effects as the main predictors of interest and four simultaneousoutcomes (preferences for disturbing, homoerotic, black-and-white, and con-ventional images). A multivariate model simplifies the treatment of “picturetype” (a nominal, unordered variable with four levels) and protects against theinflated Type 1 error that would otherwise result from a series of univariatemodels. Trait curiosity, positive affect (PA), negative affect (NA), and genderwere included as covariates. The analyses were conducted with Mplus 5.21,using maximum likelihood with robust standard errors.

For disturbing pictures, taste sensitivity had a nonsignificant linear effect(" = –.027, p = .74) but a significant quadratic effect (" = –.226, p = .007). Othersignificant linear effects appeared for trait curiosity (" = .218, p = .011) and PA(" = .245, p = .002). For homoerotic pictures, taste sensitivity had a nonsignificantlinear effect (" = .048, p = .57) but a significant quadratic effect (" = –.229,p = .005). Other significant effects appeared for trait curiosity (" = .356, p < .001)and NA (" = –.116, p = .071).

For black-and-white images and conventional images, taste sensitivity hadneither linear (black-and-white: " = .097, p = .251; conventional: " = –.034,p = .696) nor quadratic (black-and-white: " = –.049, p = .583; conventional:" = –.042, p = .616) effects. The only significant effect was for PA, whichpredicted higher preference for both picture types (black-and-white: " = .348,p < .001; conventional: " = .254, p = .002).

176 / DEWALL ET AL.

Figure 1 displays the pattern of means broken down by supertaster group;Table 1 reports the descriptive statistics. The figure depicts the central findingsfrom the regression model: relative to tasters, non-tasters and supertasters pre-ferred the disturbing and homoerotic images less.

These findings support the miscalibration hypothesis and disconfirm thelinear sensitivity hypothesis. People with extreme levels of taste sensitivity—low and high—made extreme evaluations of artwork. The results were uniqueto provocative artwork, which lends support for the prediction that the effectof taste sensitivity on artwork evaluations would be strongest for negativelyvalenced stimuli.

STUDY 2

Study 2 was conducted to replicate and extend the results supporting themiscalibration hypothesis, using different artwork to provide converging evidence


Figure 1. Taste sensitivity and aesthetic preferences fordisturbing and erotic artwork: Study 1.

and more precise dependent measures. Because the effects of Study 1 were foundonly for negatively valenced artwork, Study 2 used only negatively valencedstimuli to explore this finding in more detail (thereby enabling us to replicatethe Study 1 results). Null results are inherently ambiguous, making it unneces-sary to include positively valenced images as a means of testing our hypothesis.Moreover, Study 2 measured a wider range of emotions, given that simpleliking is only one of many feelings that people experience in relation to the arts(Silvia, 2009, in press). Instead of assessing simple liking, Study 2 includedmulti-item measures of interest, anger, disgust, confusion, and pleasingness.Of these, disgust and pleasingness are the emotions most related to physicaltaste: they relate closely to the “approach and avoid” and “accept versus reject”dimensions of affect.

If the effects of Study 1 were due to supertasters and non-tasters wanting toavoid the provocative artwork, then these participants should report high levelsof disgust in response to disturbing artwork. It is also possible that supertastersand non-tasters will express low levels of pleasingness. We did not expecteffects for interest and confusion, two knowledge emotions that are more stronglypredicted by novelty and comprehensibility than by valence (Silvia, 2010; Turner& Silvia, 2006), or for anger, an approach-oriented negative emotion. Neverthe-less, including a wide range of emotions allowed us to test whether taste sensitivityhad specific or diffuse effects on aesthetic responses.

Method

Participants

One hundred nineteen undergraduates participated in this study. As inStudy 1, participants refrained from smoking, eating, or drinking for 1 hour priorto participating. Due to a clerical error, participant gender was not collected.

178 / DEWALL ET AL.

Table 1. Preferences Ratings: Study 1

Non-taster Taster Supertaster

Disturbing

Homoerotic

Black-and-White

Conventional

3.34 (.77)

3.38 (1.06)

3.67 (.92)

4.73 (.84)

3.46 (.73)

3.64 (1.07)

3.75 (.77)

4.67 (.69)

3.16 (.72)

3.13 (1.41)

3.72 (1.03)

4.52 (1.02)

Note: Standard deviations appear in parentheses.


Participants arrived at a laboratory in groups of two to four for a studyconcerning the relationship between taste sensitivity and perceptions of artwork.Participants sat in individual cubicles and were presented with 12 images fromAndres Serrano’s The Morgue series. Each image depicts a corpse that wasphotographed in a New York City morgue; the image’s title describes the causeof the person’s death (e.g., “Death by Drowning,” “Knifed to Death”). Foreach image, participants gave ratings of interest, anger, disgust, confusion, andpleasingness. Ratings were made on a scale that ranged from 1 (not at all) to7 (extremely). After rating all 12 images, participants completed the PANASand a 50-item measure of the Big Five factors taken from the InternationalPersonality Item Pool (Goldberg, Johnson, Eber, Hogan, Ashton, Cloningeret al., 2006). Participants then completed the taste sensitivity test using thesame method as in Study 1, after which they were debriefed, given their compen-sation, and dismissed.


As before, we conducted a multivariate regression model in which ratings ofthe five emotions (pleasingness, disgust, anger, interest, and confusion) weremodeled as simultaneous outcomes, and taste sensitivity was included as a con-tinuous variable. The Big Five factors were included as covariates.1 As before,the analyses were conducted with Mplus 5.21 using maximum likelihood withrobust standard errors. Figure 2 displays the pattern of effects broken downby groups of non-tasters (n = 31), tasters (n = 61), and supertasters (n = 27);Table 2 reports the descriptive statistics.

As expected, and consistent with the miscalibration hypothesis, we foundsignificant quadratic effects for disgust and pleasingness, the emotions moststrongly connected to approach and avoidance. For disgust, taste sensitivity hada nonsignificant linear effect (" = .047, p = .49) but a significant quadratic effect(" = .135, p = .049). For pleasingness, taste sensitivity again had a nonsignificantlinear effect (" = –.042, p = .48) but a significant quadratic effect (" = –.121,p = .026). Taste sensitivity had no significant linear or quadratic effects forratings of anger, interest, and confusion ("s ranged from .015 to .09).

Although included as control variables, the Big Five factors had interestingeffects in their own right. Consistent with much past work, Openness to Experi-ence predicted greater levels of interest (" = .166, p = .025). Neuroticism predicted


1 Although we measured PA and NA as well, we opted to include the Big Five factors andomit PA and NA as covariates. PA and NA overlap strongly with Extraversion and Neuroticism(Watson, 2000), which complicates estimating the multivariate regression model. Given thatStudy 1 used PA and NA as covariates, including the Big Five in Study 2 was more informative.Using PA and NA here does not appreciably change the study’s main results.

stronger negative emotions to the works—particularly higher disgust (" = .238,p < .001) and anger (" = .130, p = .024) and lower pleasingness (" = –.124,p = .017)—consistent with the higher emotional reactivity typical of neuroticism.

These findings provide additional support for the miscalibration hypothesis thatextreme taste sensitivity—low and high—has direct relations to extreme aestheticjudgments. Taste sensitivity scores predicted levels of disgust and pleasingness,the two emotions most closely related to an “approach and avoid” dimension ofbehavior and an “accept versus reject” dimension of affect. Extreme tastesensitivity—low or high—predicted high levels of disgust and low levels ofpleasingness in response to disturbing artwork.

STUDY 3

Studies 1 and 2 provided consistent support for the miscalibration hypothesis byshowing that extreme physical taste sensitivity related to more extreme judgmentsof disturbing and provocative artwork. In Study 3, we sought to replicate and

180 / DEWALL ET AL.

Figure 2. Taste sensitivity, pleasingness, and disgust: Study 2.

extend the first two studies by examining expertise in the arts as a moderator of theeffect of taste sensitivity. A large body of work in empirical aesthetics shows thatmany variables affect the aesthetic experience of novices but not of experts(Augustine & Leder, 2006; Kozbelt, 2006). Novices tend to use tacit, intuitiveexperiences as a basis for their aesthetic judgments, whereas experts tend to useformal, crystallized knowledge of the arts (Parsons, 1987; Smith & Smith, 2006).Novices, for example, like visual art that contains their favorite colors. Experts, incontrast, report that the mere presence of certain colors (as opposed to theircompositional use) is less important (Parsons, 1987). As a result, novices’aesthetic judgments are influenced by a range of low-level features, such as anobject’s angularity, color, and typicality (Silvia & Barona, 2009). We would thusexpect taste sensitivity to have a larger effect for novices than for experts.

As in Study 2, we focused on disturbing and controversial images, but weexpanded the set of images beyond a single series associated with a single artist.People viewed eight images, which had been used in past work on negativeaesthetic emotions, and they rated the images for pleasingness and disgust, whichemerged as central outcomes in the earlier studies. To rule out possible ordereffects, we also counterbalanced the order in which participants were exposed tothe PTC paper.

Method

Participants

One hundred thirty-two undergraduates participated in this study.


Participants arrived at a laboratory in groups of two to eight and sat at indi-vidual tables. The experimenters explained that the study was about people’s


Table 2. Preferences Ratings: Study 2

Non-taster Taster Supertaster

Pleasant

Disgusting

Angry

Interesting

Confusing

2.25 (.72)

2.60 (1.31)

3.73 (1.40)

3.77 (1.13)

3.69 (.94)

2.37 (.78)

2.35 (1.17)

3.47 (1.26)

4.04 (1.01)

3.71 (.78)

1.91 (.71)

3.24 (1.48)

3.67 (1.41)

4.05 (.97)

3.55 (1.15)

Note: Standard deviations appear in parentheses.

opinions about a broad range of images. The taste ratings were obtained via thesame methods as in Studies 1 and 2. The set of images consisted of eight picturesthat had been used in past work on negative aesthetic emotions (Cooper & Silvia,2009; Silvia & Brown, 2007; Turner & Silvia, 2006), such as Piss Christ byAndres Serrano, Death and Funeral of Cain by David Alfaro Siqueiros, andSaturn Devouring His Children by Francisco Goya. People received a bookletwith color reproductions of the images. After viewing each image for as long asthey wished, people rated the image for disgust (“I find this picture disgusting”),pleasingness (“This picture is pleasant”), and anger (“This pictures makes meangry”); as distracter items, we also obtained ratings of the images’ brightnessand comprehensibility. All items were completed using 1 (not at all) to 7 (yes,definitely) scales. The order was counterbalanced: half of the participants viewedand rated the artwork first, and the other half gave the taste ratings first.

After completing the taste task and rating the images, participants completedthe PANAS and a measure of expertise in the arts. Smith and Smith’s (2006)aesthetic fluency scale lists terms and figures from art history (e.g., IsamuNoguchi, Fauvism) and asks people how familiar each term or figure is to them,using a 0 to 4 scale (0 = I have never heard of this artist or term, 4 = I can talkintelligently about this artist or idea in art). Several studies have providedevidence for the validity of the scale as a measure of expertise in a range ofsamples and aesthetic contexts (Silvia, 2007; Silvia & Barona, 2009; Silvia& Berg, 2011; Silvia, Martin, & Nusbaum, 2009; Smith & Smith, 2006).


The aesthetic fluency scale had reliable scores (# = .85), so we averaged theitems for an overall expertise score and then standardized this score. Our analysesexamined whether taste sensitivity predicted aesthetic judgments, particularlyif it interacted with expertise. We tested these hypotheses with a multivariateregression model that had three outcomes (ratings of pleasingness, disgust, andanger) and that included linear and quadratic contrast codes for taste sensitivity,a main effect for expertise, and then interactions between expertise and thelinear effect and the quadratic effect. To simplify the interpretation of the effectsin light of the many outcomes and interaction terms, we used the discrete tastegroups—non-tasters (n = 37), tasters (n = 54), and supertasters (n = 41)—as thepredictor instead of the continuous taste ratings. In addition, we examined theeffect of taste group among participants who were relatively high (i.e., 1 standarddeviation above the mean) and relatively low (i.e., 1 standard deviation belowthe mean) on expertise (Aiken & West, 1991). We estimated the models inMplus 5.21, using maximum-likelihood with robust standard errors.

Our main hypothesis was that the quadratic effect of physical taste sensitivityon aesthetic judgments would be moderated by individual differences in artexpertise. For disgust, as predicted, we found an interaction between art expertise

182 / DEWALL ET AL.

and the quadratic taste sensitivity effect (" = –.143, p = .05), indicating thatthe nonlinear effect of taste sensitivity varied across levels of art expertise.We also observed a main effect of art expertise (people high in expertisereported less disgust; " = –.223, p = .002). Figure 3 depicts the estimated means.For people low in art expertise, there was a U-shaped, nonlinear effect oftaste sensitivity: people high and low in taste sensitivity reported the mostdisgust. For people high in art expertise, in contrast, taste sensitivity had noeffect on disgust.

For pleasantness, there was only a main effect of art expertise (" = .275,p < .001): people high in expertise found the pictures more pleasing. UnlikeStudies 1 and 2, pleasingness scores had unusually restricted variability: for alleight pictures, the modal response was 1 on the 1 to 7 scale, creating a floor effect.


Figure 3. Taste sensitivity, art expertise, and disgust: Study 3.

Consistent with Study 2, we found no main effects or interactions with angerratings. There was a lone main effect of expertise (" = –.244, p < .001), in whichpeople higher in art expertise reported less anger. Physical taste sensitivity thushad a specific impact on avoidance-related negative emotion instead of havinga general effect on negative emotions.

In short, we found evidence that expertise in the arts moderates the effectof miscalibrated taste sensitivity on aesthetic judgments, namely judgments ofdisgust in response to negative and controversial art. The moderating effectof art expertise is congruent with a large body of work that shows that expertsand novices experience art differently (e.g., Augustine & Leder, 2006; Kozbelt,2006; Parsons, 1987; Silvia, 2006). People lower in art expertise displayeda sharper nonlinear effect of taste sensitivity, whereas people higher in artexpertise showed no effect. Studies 1 and 2 found weaker nonlinear effects oftaste sensitivity, presumably because art expertise went unmeasured and henceunmodeled in those studies.

GENERAL DISCUSSION

Taste can refer to how people experience food or beverages, but it is also usedto describe how people experience visual art. The current research focused on onepossible biological mechanism underlying taste for visual artwork. Across threestudies, we found consistent evidence for the miscalibration hypothesis, whichpredicts that extreme physical taste sensitivity will relate to extreme responsesto visual art. Specifically, participants with extremely low or high physical tastesensitivity made relatively extreme evaluations compared to participants withaverage taste sensitivity. The effects of physical taste sensitivity on evaluationsof artwork were most pronounced for disturbing and provocative artwork, whichconfirms prior work suggesting that negatively valenced stimuli have a strongerimpact on evaluations than do positive stimuli (Baumeister et al., 2001). Inaddition, the results of Studies 2 and 3 showed that extreme taste sensitivityled participants to express responses specifically linked to avoiding disturbingimages (high levels of disgust) as opposed to expressing general negativity towardthe disturbing images. We found no differences on the negative, approach-relatedemotion of anger (Carver & Harmon-Jones, 2009).

The results of Study 3 showed that the relationship between miscalibratedphysical taste sensitivity and aesthetic judgment was found only among partici-pants who had relatively little artistic expertise. These findings suggest that whenasked to make aesthetic judgments, people who have relatively little artisticexpertise depend primarily on their tacit, intuitive responses, which are affectedby factors such as genetic predisposition for taste sensitivity. People who haverelatively high levels of artistic expertise, in contrast, have an expansiveknowledge and experiential base from which to draw when making aestheticjudgments. Expertise in the arts thus eliminates the effect of miscalibratedphysical taste sensitivity on aesthetic judgments.

184 / DEWALL ET AL.

Limitations and Future Research

Though the results across the three studies consistently supported the miscali-bration hypothesis, the work had some limitations that warrant consideration.One limitation is that we focused on one type of aesthetic behavior—the experi-ence of visual artwork—and hence it is an open question whether our resultsapply to other types of aesthetic behavior. We believe that the stimuli used inthe current work, which were all static images, may have underestimated thestrength of the link between physical taste sensitivity and art preferences. Becausedynamic stimuli produce stronger emotional responses compared to static stimuli(e.g., Weyers, Muhlberger, Hefele, & Pauli, 2006), we would expect that viewingdisturbing aesthetic behavior would produce especially strong reactions amongnon-tasters and supertasters with low levels of expertise in the aesthetic behavior.A second limitation is that we did not take into account non-genetic factorsthat may influence taste sensitivity. Chronic caffeine or nicotine use, bothof which influence taste sensitivity (Kunin, Bloch, Terada, Rogan, & Smith,2001; Palmatier & Bevins, 2001; Snedecor, Pomerleau, Mehringer, Pomerleau, &Ninowski, 2006), may suppress the effects of sensitivity to PTC on art preferences.Future work may explore how PTC sensitivity interacts with chronic behaviorsthat influence taste sensitivity to predict art preferences.

Future work may explore the potential evolutionary basis of our findings.Selection pressures posed by infectious microbes were involved in the develop-ment of the behavioral immune system, which served the function of promotingthe behavioral avoidance of any carrier that might threaten one’s ability to surviveand reproduce (Schaller, 2006; Schaller & Duncan, 2007). Crucially, the behaviorimmune system produces domain-general responses because it is more beneficialfor people to avoid objects that may threaten their well-being than to avoidobjects that may be a boon to their health. If sense of taste evolved to alert peopleto substances that may threaten their health, then a by-product of that adaptationmay involve having heightened avoidant responses to visual artwork amongpeople with miscalibrated taste sensitivity.

CONCLUSION

The present results provide novel insight into why people frequently describesensitivity to artwork in terms used to explain physical taste sensitivity. Althoughlinking taste for artwork with gustatory sensitivity may represent a simplelinguistic metaphor, the tendency to think in perceptual terms when describingcomplex concepts demonstrates a fundamental way that people relate to theirenvironment (Lakoff & Johnson, 1999; Tolaas, 1991). Indeed, a growing bodyof evidence suggests that the use of metaphor helps to explain general affec-tive experiences. Happiness is frequently described as feeling “up” (ratherthan “down”), bright objects are good (and dark objects are bad), and vertical“highness” is divine (Meier, Hauser, Robinson, Friesen, & Schjeldahl, 2007;


Meier & Robinson, 2004; Meier, Robinson, & Clore, 2004). The current investi-gation adds to this growing body of literature and suggests that taste for artwork isrooted at least partly in biological differences in physical taste sensitivity, butthat these biological differences have their effect primarily among those with lowartistic expertise. Although our results do not directly refute William Bernbach’sstatement not to “confuse good taste with the absence of taste,” they do suggestthat miscalibrated taste sensitivity, coupled with low artistic expertise, relates toextremity of judgments of disturbing and provocative art.

ACKNOWLEDGMENTS

The authors wish to thank Stephen Leeper and Camilla Sanders for theirassistance with data collection.

REFERENCES

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpretinginteractions. Thousand Oaks, CA: Sage.

Augustine, M. D., & Leder, H. (2006). Art expertise: A study of concepts and conceptualspaces. Psychology Science, 48, 135-156.

Axelsson, Ö. (2007). Individual differences in preferences for photographs. Psychologyof Aesthetics, Creativity, and the Arts, 1, 61-72.

Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults:A test of the four-category model. Journal of Personality and Social Psychology, 61,226-244.

Bartoshuk, L. M. (2000). Psychophysical advances aid the study of genetic variation intaste. Appetite, 34, 105.

Bartoshuk, L. M., Duffy, V. B., & Miller, I. J. (1994). PTC/PROP taste: Anatomy,psychophysics, and sex effects. Physiology & Behavior, 56, 1165-1171.

Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is strongerthan good. Review of General Psychology, 5, 323-370.

Bourdieu, P. (1984). Distinction: A social critique of the judgment of taste. London:Routledge and Kegan.

Carver, C. S., & Harmon-Jones, E. (2009). Anger is an approach-related affect: Evidenceand implications. Psychological Bulletin, 135, 183-204.

Cooper, J. M., & Silvia, P. J. (2009). Opposing art: Rejection as an action tendency ofhostile aesthetic emotions. Empirical Studies of the Arts, 27, 109-126.

Drewnowski, A., Henderson, S. A., Shore, A. B., & Barratt-Fornell, A. (1997). Nontasters,tasters, and supertasters of 6-n-Propylthiouracial (PROP) and hedonic response tosweet. Physiology & Behavior, 62, 649-655.

Driscoll, K., Perez, M., Cukrowicz, K., Butler, M., & Joiner, T. (2006). The associationsof phenylthiocarbamide (PTC) tasting to alcohol problems and family history ofalcoholism differ by gender. Psychiatry Research, 143, 21-27.

Finkel, E. J., & Slotter, E. B. (2007). An attachment theory perspective on the perpetrationof intimate partner violence. DePaul Law Review, 56, 895-907.

186 / DEWALL ET AL.

Fiske, S. T. (1980). Attention and weight in person perception: The impact of nega-tive and extreme behavior. Journal of Experimental Research in Personality, 22,889-906.

Frank, R. A., & Korchmar, D. L. (1985). Gustatory processing differences in PTCtasters and non-tasters: A reaction time analysis. Physiology & Behavior, 35,239-242.

Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R., Ashton, M. C., Cloninger, C. R.,et al. (2006). The international personality item pool and the future of public-domainpersonality assessment. Journal of Research in Personality, 40, 84-96.

Hare, R. D. (2003). The Hare, PCL-R (2nd ed.). Toronto, Ontario: Multi-Health Systems.Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process.

Journal of Personality and Social Psychology, 52, 511-524.Hellmuth, J. C., & McNulty, J. K. (2008). Neuroticism, marital violence and the moder-

ating role of stress and behavioral skills. Journal of Personality and Social Psychology,95, 166-180.

Hodges, B. H. (1974). Effect of valence on relative weighting in impression formation.Journal of Personality and Social Psychology, 30, 378-381.

Holbrook, M. B., Weiss, M. J., & Habich, J. (2004). Class-related distinctions in Americancultural tastes. Empirical Studies of the Arts, 22, 91-115.

Joiner, T., & Perez, M. (2004). Phenylthiocarbamide (PTC) tasting and family historyof depression, revisited: Low rates of depression in families of supertasters. PsychiatryResearch, 126, 83-87.

Kalmus, H. (1958). Improvements in the classification of the taster genotypes. Annals ofHuman Genetics, 22, 222-230.

Kashdan, T. B., Rose, P., & Fincham, F. D. (2004). Curiosity and exploration: Facili-tating positive subjective experiences and personal growth opportunities. Journal ofPersonality Assessment, 82, 291-305.

Kozbelt, A. (2006). Dynamic evaluation of Matisse’s 1935 Large Reclining Nude.Empirical Studies of the Arts, 24, 119-137.

Kraaykamp, G. (2002). Cumulative advantages and inequality in lifestyle: A Dutchdescription of distinction in taste. The Netherlands’ Journal of Social Sciences, 38,121-143.

Kunin, D., Bloch, R. T., Terada, Y., Rogan, F., & Smith, B. R. (2001). Caffeine promotesan ethanol-induced conditioned taste aversion: A dose-dependent interaction. Psycho-pharmacology, 9, 326-333.

Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh: The embodied mind andits challenge to Western thought. New York: Basic Books.

Locher, P. J., Smith, J. K., & Smith, L. F. (2001). The influence of presentation formatand viewer training in the visual arts on the perception of pictorial and aestheticqualities of paintings. Perception, 30, 449-465.

McWhinnie, H. J. (1968). A review of research on aesthetic measure. Acta Psychologica,28, 363-375.

Meier, B. P., Hauser, D. J., Robinson, M. D., Friesen, C. K., & Schjeldahl, K. (2007).What’s “up” with God? Vertical space as a representation of the divine. Journal ofPersonality and Social Psychology, 93, 699-710.

Meier, B. P., & Robinson, M. D. (2004). Why the sunny side is up: Associations betweenaffect and vertical position. Psychological Science, 15, 243-247.


Meier, B. P., Robinson, M. D., & Clore, G. L. (2004). Why good guys wear white:Automatic inferences about stimulus valence based on brightness. PsychologicalScience, 15, 82-87.

Mikulincer, M., & Shaver, P. R. (2007). Attachment in adulthood: Structure, dynamicsand change. New York: The Guilford Press.

Moberg, P. J., Balderston, C. C., Rick, J. H., Roalf, D. R., Weintraub, D., Kleiner-Fisman, G., et al. (2007). Phenylthiocarbamide (PTC) perception in Parkinson disease.Cognitive and Behavioral Neurology, 20, 145-148.

Moberg, P. J., McGue, C., Kanes, S. J., Roalf, D. R., Balderston, C. C., Gur, R. E.,et al. (2007). Phenylthiocarbamide (PTC) perception in patients with schizophreniaand first-degree family members: Relationship to clinical symptomatology andpsychophysical olfactory performance. Schizophrenia Research, 90, 221-228.

Palmatier, M. I., & Bevins, R. A. (2001). Chronic caffeine exposure in rats blocks asubsequent nicotine-conditioned taste avoidance in a one-bottle, but not a two-bottletest. Pharmacology, Biochemistry and Behavior, 70, 279-289.

Parsons, M. J. (1987). How we understand art: A cognitive developmental account ofaesthetic experience. New York: Cambridge University Press.

Reber, R., Schwartz, N., & Winkielman, P. (2004). Processing fluency and aestheticpleasure: Is beauty in the perceiver’s processing experience? Personality and SocialPsychology Review, 8, 364-382.

Schaller, M. (2006). Parasites, behavioral defenses, and the social psychologicalmechanisms through which cultures are evoked. Psychological Inquiry, 17, 96-101.

Schaller, M., & Duncan, L. A. (2007). The behavioral immune system: Its evolution andsocial psychological implications. In J. P. Forgas, M. G. Haselton, & W. von Hippel(Eds.), Evolution and the social mind: Evolutionary psychology and social cognition(pp. 293-307). New York: Psychology Press.

Silvia, P. J. (2005a). Cognitive appraisals and interest in visual art: Exploring an appraisaltheory of aesthetic emotions. Empirical Studies of the Arts, 23, 119-133.

Silvia, P. J. (2005b). What is interesting? Exploring the appraisal structure of interest.Emotion, 5, 89-102.

Silvia, P. J. (2006). Artistic training and interest in visual art: Applying the appraisal modelof aesthetic emotions. Empirical Studies of the Arts, 24, 139-161.

Silvia, P. J. (2007). Knowledge-based assessment of expertise in the arts: Exploringaesthetic fluency. Psychology of Aesthetics, Creativity, and the Arts, 1, 247-249.

Silvia, P. J. (2009). Looking past pleasure: Anger, confusion, disgust, pride, surprise,and other unusual aesthetic emotions. Psychology of Aesthetics, Creativity, and theArts, 3, 48-51.

Silvia, P. J. (2010). Confusion and interest: The role of knowledge emotions in aestheticexperience. Psychology of Aesthetics, Creativity, and the Arts, 4, 75-80.

Silvia, P. J. (in press). Human emotions and aesthetic experience: An overview ofempirical aesthetics. In A. Shimamura & S. E. Palmer (Eds.), Aesthetic science:Connecting minds, brains, and experience. New York: Oxford University Press.

Silvia, P. J., & Barona, C. M. (2009). Do people prefer curved objects? Angularity,expertise, and aesthetic preference. Empirical Studies of the Arts, 27, 25-42.

Silvia, P. J., & Berg, C. (2011). Finding movies interesting: How expertise andappraisals influence the aesthetic experience of film. Empirical Studies of the Arts, 29,73-88.

188 / DEWALL ET AL.

Silvia, P. J., & Brown, E. M. (2007). Anger, disgust, and the negative aesthetic emotions:Expanding an appraisal model of aesthetic experience. Psychology of Aesthetics,Creativity, and the Arts, 1, 100-106.

Silvia, P. J., Martin, C., Nusbaum, E. C. (2009). A snapshot of creativity: Evaluating aquick and simple method for assessing divergent thinking. Thinking Skills andCreativity, 4, 79-85.

Skov, M., & Vartanian, O. (Eds.). (2009). Neuroaesthetics. Amityville, NY: Baywood.Smith, L., & Smith, J. (2006). The nature and growth of aesthetic fluency. In P. Locher,

C. Martindale, & L. Dorfman (Eds.), New direction in aesthetics, creativity, and thearts (pp. 47-58). Amityville, NY: Baywood.

Snedecor, S. M., Pomerleau, C. S., Mehringer, A. M., Pomerleau, O. F., & Ninowski, R.(2006). Differences in smoking-related variables based on phenylthiocarbamide‘taster’ status. Addictive Behaviors, 31, 2309-2312.

Tepper, B. J., & Nurse, R. J. (1997). Fat perception is related to PROP taster status.Physiology and Behavior, 61, 949-954.

Tolaas, J. (1991). Notes on the origin of some spatialization metaphors. Metaphor andSymbolic Activity, 6, 203-218.

Turner, S. A., Jr., & Silvia, P. J. (2006). Must interesting things be pleasant? A test ofcompeting appraisal structures. Emotion, 6, 670-674.

Vohs, K. D., Baumeister, R. F., & Ciarocco, N. J. (2005). Self-regulation and self-presentation: Regulatory resource depletion impairs impression management andeffortful self-presentation depletes regulatory resources. Journal of Personality andSocial Psychology, 88, 632-657.

Vonk, R. (1993). The negativity effect in trait ratings and in open-ended descriptions ofpersons. Personality and Social Psychology Bulletin, 19, 269-278.

Watson, D. (2000). Mood and temperament. New York: Guilford.Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief

measures of positive and negative affect: The PANAS scales. Journal of Personalityand Social Psychology, 54, 1063-1070.

Weyers, P., Muhlberger, A., Hefele, C., & Pauli, P. (2006). Electromyographic responsesto static and dynamic avatar emotional facial expressions. Psychophysiology, 5,450-453.

Direct reprint requests to:

C. Nathan DeWallDepartment of PsychologyUniversity of KentuckyLexington, KY 40506-0044e-mail: [email protected]


Taste sensitivity and aesthetic preferences: Is taste only a metaphor?

Documents

Transcript of Taste sensitivity and aesthetic preferences: Is taste only a metaphor?