Psychology Essay

Psychology Essay

Music, Mood, and Marketing Author(s): Gordon C. Bruner II Source: Journal of Marketing , Oct., 1990, Vol. 54, No. 4 (Oct., 1990), pp. 94-104 Published by: Sage Publications, Inc. on behalf of American Marketing Association Stable URL: https://www.jstor.org/stable/1251762 REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/1251762?seq=1&cid=pdfreference#references_tab_contents You may need to log in to JSTOR to access the linked references. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms American Marketing Association and Sage Publications, Inc. are collaborating with JSTOR to digitize, preserve and extend access to Journal of Marketing This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms Gordon C. Bruner II Music, Mood, and Marketing That music affects human beings in various ways has probably been presumed as long as people have played music. Many marketing practitioners already accept this notion, given that music is increasingly used as a stimulus in the retail environment as well as in radio and television advertising. Yet, fewer than 20 published empirical studies in marketing have music as their focus. The author reviews the small body of marketing literature, surveys relevant literature outside marketing, and provides research propositions to guide future studies. M USIC has long been considered an efficient and effective means for triggering moods and communicating nonverbally. It is therefore not surprising that music has become a major component of consumer marketing, both at the point of purchase and in advertising. There is now even a trade magazine devoted exclusively to these uses of music (Marketing Through Music), and at least one ad agency has a vice president for music (Garfield 1988). Yet, despite being a prominent promotional tool, music is not well understood or controlled by marketers. Consequently, marketers are precariously dependent on musicians for their insight into the selection or composition of the “right” music for particular situations. The purpose of this article is to examine the be- havioral effects of music, with special emphasis on music’s emotional expressionism and role as a mood influencer. The practical implications of what is currently known as well as propositions for future research are discussed.’ ‘No new theory of music’s expressiveness or debate about whether such expressiveness exists is presented here. Excellent discussions of such issues are given by Hevner (1935b), Meyer (1967), Clynes (1977; Clynes and Nettheim 1982), and Langer (1976). Budd (1985) pro- vides a review. Further, several terms have been used in the literature to refer to the affective dimension of human behavior, such as “emo- tion” (Holbrook and Batra 1987), “mood” (Gardner 1985), and “sentic state” (Clynes 1977). Such terms as these are used here synony- mously. Music and Mood That music is an especially powerful stimulus for fecting moods is no revelation; it is attested throughout history by poets, playwrights, compos and, in the last two centuries, researchers. Butler (19 provides a bibliography of nearly 900 entries in se languages, all pertaining in some way to the stud music psychology in the nineteenth century. More cently, Seidman (1981) reviewed the contribution music to media productions (movies and education films), concluding that cognitive and affective co prehension of stimuli can be influenced. This con sion seems to have been long held by persons in movie industry, as is evident in the development elaborate suggestions for marrying music to video (Ze 1973, ch. 15). Further, the research of Manfred Cl (1975, 1977, 1980; Clynes and Nettheim 1982), a n scholar in both music and neurophysiology, indic that appropriately structured music acts on the n vous system like a key on a lock, activating brain cesses with corresponding emotional reactions. Music is not simply a generic sonic mass, but rath a complex chemistry of controllable elements. fortunately, no definitive taxonomy of music elem has been developed. Definitions vary and what primary component in one taxonomy is a subc ponent in another. Time- and pitch-related charac Gordon C. Bruner II is Associate Professor of Marketing, Southern Illinois University. The author acknowledges the helpful comments by the Editor and three anonymous JM reviewers, as well as Clifton Anderson, Paul Hensel, and James Kellaris. A more detailed version of this article and a bibliography can be obtained from the author. istics appear on almost all lists and also have s empirical confirmation (Henkin 1955, 1957; Nie and Cesarec 1982). Though less clear, evidence has been found for a third factor, musical texture purposes of this review and discussion, time, pitc 94 / Journal of Marketing, October 1990 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms and texture are the three main structural factors on which music is based. (Definitions for each of these factors and for relevant “subcomponents” are given in the Appendix.) dence that the relationship is not a simple monotonic one. An inverted-U preference function seems to be more likely, with 70 to 110 BPM (beats per minute) being the range of favored tempo (Dowling and Table 1 is a list of some propositions about the Harwood 1986; Fraisse 1982; Holbrook and Anand affective meaning conveyed by certain music characteristics. They illustrate the sort of beliefs held by musicians who compose, perform, and conduct music that evokes particular feelings (e.g., Cooke 1962; Zettl 1973). Many of these observations have been tested empirically, as summarized in Table 2.2 Principal re- 1988). This range of preference is likely to vary with the context (Dowling and Harwood 1986; Holbrook and Anand 1988; Konecni 1982) and may not be as important to determining preferences as musical genre (Cupchik, Rickert, and Mendelson 1982). The rhythm aspect of the time component also has been examined in several studies. Hevner (1936) observed that firm rhythms were judged to be more saTime-Related Findings cred, serious, and/or robust; smooth-flowing rhythms Several researchers over the years have studied tempo were felt to be more happy, playful, and/or dreamy. and arrived at the same general conclusion: all other Gundlach (1935) noted that music with smooth rhythm things being equal, fast music is considered to be more was characterized as brilliant or animated; music with happy and/or pleasant than slow music (Gundlach uneven rhythm was perceived to express dignity or sults of these studies are discussed next. 1935; Rigg 1940a; Scherer and Oshinsky 1977; exaltation.3 Swanwick 1973; Watson 1942; Wedin 1972). More For another aspect of the time factor, Wedin (1972) specifically, Hevner (1937) found that slow tempi found that most of the variance in the “activity” di- tended to evoke tranquil, sentimental, and/or solemn mension of the music she examined was explained by sorts of descriptions; fast tempi elicited responses re- “phrasing.” Specifically, staccato-note-filled music lating to exhilarating and/or joyous sorts of feelings. gave the impression of liveliness, energy, and/or agThough these studies suggest a positive relation itation, especially if performed with great intensity between affect and musical pace, there is strong evi- 2The number of studies using music as a stimulus variable is likely 3Though these results are reasonably similar to Hevner’s, other studies of rhythm have produced a confounding array of both rhythm deto be in the hundreds. Many fewer have music itself as the focus. scriptors and emotional expressions that make them unamenable to This smaller group is discussed here and the most relevant studies are comparison. For additional studies, see Gabrielsson (1973), Scherer highlighted in Table 2. and Oshinsky (1977), and Watson (1942). TABLE 1 Emotional Expressions Ascribed to Various Components of Music’ Time-Related Expressions 1. Duple rhythms produce a rigid and controlled expression in comparison with triple rhythm, which is more relaxed or abandoned. 2. The faster the tempo, the more animation and happiness is expressed. 3. Even, rhythmic movement can represent the unimpeded flow of some feeling; dotted, jerky, unev produce more complex expressions. 4. Firm rhythms suggest a serious mood whereas smooth-flowing rhythms are more playful. 5. Staccato notes give more emphasis to a passage than legato notes. Pitch-Related Expressions 1. “Up” and “down” in pitch not only correspond to up and down in the physical world, but can also “out-and-in” as well as “away-and-back,” respectively. 2. Rising and falling pitch can convey a growing or diminishing intensity in a given emotional conte 3. Songs in higher keys are generally considered to be happier than songs in lower keys. 4. Music in the major mode expresses more animated and positive feelings than music in the minor 5. Complex harmonies are more agitated and sad than simple harmonies, which are more serene an Texture-Related Expressions 1. Loudness can suggest animation or proximity whereas low volume implies tranquility or distance. 2. Cresendo (soft to loud) expresses an increase in force whereas diminuendo (loud to soft) suggest decrease in power. 3. The timbre of brass instruments conveys a feeling of cold, hard force whereas reed instruments produce a lonely, melancholy expression. ‘Based on information given by Cooke (1962) and Zettl (1973). Music, Mood, and Marketing / 95 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms TABLE 2 Summary of Relevant Research Involving Music Independent Variables’ Dependent Author (date) Sample and Setting Time Pitch Texture Other Variables Marketing-Related Studies Alpert and Alpert 48 college students Tempo and rhythm Harmony varied Dynamic (1986, 1988) in classroom varied “mood” intention experiment Gorn (1982) 244 and 122 college Appeal varied Choice behavior students in 2-part classroom experiment Holbrook and 44 college students Tempo varied Constant Constant Perceived activity Anand (1988) in “lab” and appeal experiment Hunt (1988) 78 college students Music presence Recall in “lab” and program experiment content Kellaris and Cox 302 college Tempo similar Modality similar Instrumentation Appeal varied Choice behavior (1987) students classroom in similar experiment Milliman (1982) 216 shoppers in Tempo varied Volume constant Shopping behaviors supermarket experiment Milliman (1986) 1392 customer Tempo varied Volume constant In-restaurant groups restaurant in behaviors experiment Park and Young 120 women in Presence of music Brand attitude (1986) “lab” experiment and involvement varied Sewall and Sarel Analysis of 832 Presence of music Recall (1986) radio ads using an average of 200 mall shoppers Simkins and Smith 40 college students Appeal varied Audio message (1974) in “lab” experiment evaluation Smith and Curnow 1100 shoppers in Volume varied Music appeal and (1966) supermarket experiment shopping behaviors Stewart and Furse Analysis of 1059 TV Presence and form Comprehension, (1986) ads of music’s use persuasion, and recall Stout and Leckenby Analysis of 50 TV Tempo examined Mode and Volume examined Presence and form Cognitive and (1988) ads using average distinctiveness of music’s use affective of 30 mall examined responses shoppers Wintle (1978) 77, 96, and 120 Compatibility of Affective ratings college students music and ads in 3-experiment varied classroom study Yalch and 86 shoppers in Background and Music appeal and Spangenberg department store foreground shopping (1988) Other Studies experiment music behaviors Clynes and 189 college Frequency Amplitude Emotion expressed Nettheim (1982) students and staff modulated modulated in unknown setting Gundlach (1935) 112 college Tempo and rhythm Pitch and range Volume and Emotion expressed students in examined examined orchestral range unspecified setting examined Hevner (1935a) 205 college Intrapair constancy Major vs. minor Intrapair constancy Emotion expressed students classroom in mode experiment Hevner (1936) 450 college Rhythm varied Melodic line and Intrapair constancy Emotion expr students in harmony varied classroom experiment Hevner (1937) 142 and ? college Fast vs. slow tempo High vs. low pitch Intrapair constancy Emotion expres students in 2 classroom experiments Heyduk (1975) 120 college Syncopation varied Number and variety Preferred and students in of chords varied perceived unspecified experimental complxity setting 96 / Journal of Marketing, October 1990 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms TABLE 2 (continued) Summary of Relevant Research Involving Music Independent Variables’ D dt Author (date) Sample and Setting Time Pitch Texture Other Variables Kinnear (1959; Van 25 college students Tempo constant Melody constant Orchestration varied Emotion expressed Stone 1960 in classroom experiment Rigg (1940a) 88 college students Tempo varied Intragroup Intragroup Emotion expressed in unspecified constancy constancy experimental setting Rigg (1940b) 84 college students Intragroup Pitch varied Intragroup Emotion expressed in unspecified constancy constancy experimental setting Scherer and 48 college students Tempo and rhythm Mode and pitch Harmonics and Emotion expressed Oshinsky (1977) in “lab” setting varied varied amplitude varied Smith and Cuddy 36 college students Constant Melodic complexity Constant Pleasantness (1986) in “lab” setting varied Swanwich (1973) 400 students of Tempo varied Direction and Emotion expressed various ages in melodic line unspecified varied setting Watson (1942) 20 musicians and Tempo and rhythm Pitch examined Volume and overall Emotion expressed 600 students of examined “sound” various ages in examined unspecified settings Wedin (1972) 5 experiments of Tempo and rhythm, Mode, harmony, Volume examined Type and style of Emotion expressed varying sample and phrasing and singability music examined size examined examined Vinovich (1975) 30, 30, and 100 Tempo and rhythm college students examined harmony in 3-experiment examined classroom study aBlank spaces in the chart indicate that variables were not reported as being controlled, man (volume). At the other extreme, legato music (partiction of melodic line and note ularly with softer performances) was perceived asascendin found a tendency for having a more peaceful, gentle, or dreamy character. viewed as more dignified or sole scending ones were more exhila Pitch-Related Findings findings on note range indicate t The effect of pitch has been investigated Gundlach the greatest by range (more than a (1935), Hevner (1937), Rigg (1940b), andbrilliant Watsonthan th ceived as more (1942). The findings are relatively consistent sugoctave range of and notes, which m gest a strong association between pitch (Gundlach and perceived mournful 1935). happiness: music with high pitch is more exciting or Texture-Related Findings happy than low pitched music, which is perceived as sad. Musical texture has been studied At least two studies have been conducted to test time-related characteristics. How conventional thinking amongfocusing musicians Western onin orchestration showe culture that the major mode makes dynamic and posments carried the melody in s itive expressions whereas thetriumphant minor mode produces and/or grotesque, w the opposite feelings. The results seem to support the awkward and/or mournful feelin conventional wisdom: the minor mode has plaintive, were perceived as brilliant and/o angry, or mysterious qualitiessounds in contrast the more with pi wereto associated happy, bright, or playful expressions of the major mode glad (Gundlach 1935). Kinnear (1 (Hevner 1935a; Scherer and reports Oshinsky 1977). that woodwind instrumen The expressive tendencies of harmony been whereas b judged to behave whimsical studied by Hevner (1936), Watson (1942), and Wedin considered serious and/or majest (1972). In general, consonant harmonies can be de- music being performed. String scribed as playful, happy, or not serene. Dissonant har- any one p seem to represent monies seem to be perceived as more agitating, om- effective gory, but were equally inous, or sad. moods. Two other pitch-related characteristics aretextural direcVolume is the only other element that has Music, Mood, and Marketing / 97 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms been examined. Gundlach (1935) found the loudest pieces in his study were described as triumphant and/ or animated, whereas the softest pieces were judged to be delicate and/or tranquil. Watson (1942) also studied volume, finding it to be loudest for songs characterized as very exciting or very happy and softest for pieces perceived as peaceful or serious. evance factor. Further, she demonstrated that supporting music seems to intensify whereas counteracting music reduces the intensity of the dimension being characterized positively in the ad. Simpkins and Smith (1974) were concerned with the effect of background music in commercial mes- sages on the evaluation of a message’s source. Their data supported the notion that the compatibility of background music with the preference of the audience Other Relevant Findings Considerable though not universal support for an optimal complexity model as it pertains to music is noted in a literature review by Walker (1981). This model suggests that when someone expresses a degree of liking for a range of music compositions, including some more complex and others less complex than optimal, an inverted-U function tends to be found. Further, this function is likely to shift to the right over time as a piece becomes more familiar. Support for this notion can be found in the theories of Berlyne (1960, 1971) and Meyer (1967) and empirical support in a musical context is found in a variety of studies (e.g., Holbrook and Anand 1988; Krugman 1943; Smith and Cuddy 1986). Apparently, repetition can lead to perceived changes in complexity with resulting influences on affect when patterns in the music become predictable. In a series of experiments, Vinovich (1975) in- vestigated the communicative relationship between the information imparted by the video portion of a television/cinema “drama” and its musical score. Results indicated that different musical moods produced different interpretations of the same video stimulus. The conclusion was that viewers tend to create a predictable cognitive interpretation of an ambiguous video drama that logically corresponds to the “feeling” of the music in order to justify their affective reactions. Marketing-Related Studies Little music-related research has been performed in marketing and it has examined issues other than the influence of music components on mood. The following discussion first reviews the few studies on nonbehavioral outcomes of music in marketing contexts (e.g., changes in affect, purchase intention, and recall), then reviews studies examining behavioral issues (e.g., sales volume, product selection, shopping time). significantly affects the evaluation of a message’s source. Also, persons for whom the music was incompatible evaluated the credibility of the sponsor significantly lower than those who heard the message without music. However, little appeared to be gained in source evaluation by the use of compatible music. Another aspect of background music was studied by Park and Young (1986). Their focus was on the impact of involvement (high cognitive, high affective, or low involvement) and music (presence or absence) on brand attitude formation. Their results indicated that music had a facilitative effect on persons in the low involvement condition but a distracting effect for those in the high cognitive involvement condition. The effect on those in the high affective involvement con- dition was unclear. The effect of music on moods and purchase intentions has been studied recently by Alpert and Alpert (1986, 1988). One finding was that happy music produced happier moods in subjects, but sad music produced the highest purchase intentions. The authors concluded that structure and expressiveness of background music can evoke different moods and purchase intentions toward advertised products. Holbrook and Anand (1988; Anand and Holbrook 1985) studied the relationship between musical tempo and affect. They observed that as tempo increased, so did affect up to a point, then it decreased, producing an inverted-U preference function. Because general arousal was anticipated to vary with the context, the study also examined the moderating role of situational arousal on preferred musical pace. The findings were somewhat mixed, but the authors concluded that the inverted-U preference function shifts to the right under conditions of higher arousal. Several content analyses of commercials have been conducted to shed light on the role of music in advertising. The largest study of this kind was by Stewart and Furse (1986), who analyzed more than 1000 commercials. A music-related factor (auditory memory A series of experiments by Wintle (1978) provided device) was found to have a significant positive reempirical support for conventional thinking that music lationship with recall and comprehension, and the ascan significantly affect the emotional response to telesociation was stronger for new than for established vision commercials. Specifically, Wintle’s data indiproducts (p. 76, 82). cated that music and television commercials share threeMore detailed coding of music components was dimensions of subjective expression: an activity facused in a recent study by Stout and Leckenby (1988). tor, a pleasantness factor, and a potency/personal relThey found that mode had the most impact on reNonbehavioral Studies 98 / Journal of Marketing, October 1990 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms sponses of all music components examined. Specifically, respondents appeared to learn more from, indicate greater behavioral intent toward, and make more personal connections with ads using the major or mixed findings in his second study (1986) was that cus- tomers in the slow-music condition took more time to eat their meals and leave than those in the fast-music treatment. Likewise, there was a significantly longer waiting time for tables in the slow-music treatment. were rated as more irritating and generally scored less Customer bar bills were found to be much greater in modes. Commercials with music in the minor mode the slow-music condition. favorably. Faster tempo was linked with generally positive comments whereas ads with moderate or slow Recently, Yalch and Spangenberg (1988; Yalch 1988) conducted research that builds on the Milliman tempi received less favorable responses. In a content analysis of radio commercials, thestudies. They compared the effects of foreground mupresence of music was not found to have a particularly sic (Top 40) and background music (instrumental “easy salient impact on recall in comparison with other varilistening”) as well as a no-music treatment in a deables, such as product category and number of brand partment store setting. One of the only behavior-rementions (Sewall and Sarel 1986). Very different re-lated differences between the two music treatments was sults were observed by Hunt (1988). His findings sup-that younger shoppers (under 25) thought they had spent port the notion that music-based radio commercials more time shopping in the easy listening condition tend to achieve higher recall scores than ads using awhereas older shoppers perceived they had shopped straight announcement format, regardless of the prolonger when the Top 40 was being played. From these gramming context. findings, the researchers speculated that when shoppers encounter nontypical environmental factors (e.g., Behavior-Related Studies unfamiliar music), time appears to slow down. An early study with behavioral implications for marketing was conducted in two grocery stores where muDiscussion sic was either “soft” or “loud” (Smith and Curnow 1966). The results indicated that the volume level of the music was correlated negatively with shopping time but not associated with average sales per person or customers’ reported satisfaction with the music. The implication was that the sales per minute were significantly higher under the loud music condition. In an application of classical conditioning, Gorn (1982) examined the influence of music in advertising on product choice. He found that the simple association between a picture of a pen (conditioned stimu- lus) and liked music (unconditioned stimulus) could significantly affect a pen selection decision, particularly for subjects not aware that they would be asked to make a decision. However, these results have been questioned recently on methodological grounds.4 Milliman (1982, 1986) performed two separate experiments in which music tempo was manipulated, one in a supermarket and the other in a restaurant. The results of his first study (1982) indicated that in-store traffic flow was significantly slower with slow music (72 BPM or less) than with fast music (94 BPM or more). Similarly, sales volume was significantly higher with slow music than with fast music. Among the On the basis of the studies in both marketing and other fields, the following postulates summarize in gener terms what can be safely concluded. * Human beings nonrandomly assign emotional meanin to music. * Human beings experience nonrandom affective reactions to music. * Music used in marketing-related contexts is capable of evoking nonrandom affective and behavioral responses in consumers. These postulates lead to the conclusion that music is an important stimulus for marketers to study, un- derstand, and employ-especially when affective re- actions to marketer-controlled stimuli are a chief concern. However, if they are to be of great practical value, more specific issues must be addressed in future research. The following propositions are suggested by past studies but warrant further testing, particularly in marketing contexts. P1: The components of music are capable of having main as well as interaction effects on moods, cognitions, and behaviors of interest in marketers. Though attempts have been made over the years to 4Kellaris and Cox (1987), among others (Allen and Madden 1985), have pointed out several weaknesses in Gorn’s study. They replicated his study but with a few changes in the music and cover story to overcome the weaknesses. No significant association was found between pen selection and music heard, regardless of the cover story subjects received. In a further nonexperimental replication of Gorn’s study, Kellaris and Cox (1988) found a high rate of hypothesis guessing, which indicates that an apparent conditioning effect could be produced in the absence of conditioning stimuli. understand the main effects of music components (e.g., tempo), the interaction effects have received much less attention. Some emotions speculated as being expressed through the interaction of several components are given in Table 3 and warrant examination in future research. P2: The structural nature of music is related to its ability to achieve various purposes. Music, Mood, and Marketing / 99 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms TABLE 3 Musical Characteristics for Producing Various Emotional Expressions’ -~~~~~~~~~~Musical .Emotional Expression Element Serious Sad Sentimental Serene Humorous Happy Exciting Majestic rrightening Mode Major Minor Minor Major Major Major Major Major Minor Musical Tempo Slow Slow Slow Slow Fast Fast Fast Medium Slow Pitch Low Low Medium Medium High High Medium Medium Low Rhythm Firm Firm Flowing Flowing Flowing Flowing Uneven Firm Uneven Harmony Consonant Dissonant Consonant Consonant Consonant Consonant Dissonant Dissonant Dissonant Volume Medium Soft Soft Soft Medium Medium Loud Loud Varied aDeveloped primarily from Hevner (1937), Kinnear (1959), and Vinovich (1975) with additional information fr and Wedin (1972). As discussed by Hecker (1984), music can play many congruent” thoughts more accessible in memory (e.g., roles of interest to marketers. However, the music ef- Gardner 1985). fectively used to serve one purpose may be inapproP5: Several variables moderate the relationships between priate for another. To explore these various purposes, music and consumer responses. marketers might attempt to match the emotion of some Among the most important moderators to evaluate are music to that expressed in the visual and/or verbal portion of an ad (e.g., Vinovich 1975; Wintle 1978); the familiarity of the music to listeners (e.g., Fontaine counteract, intensify, or change the prevailing mood and Schwalm 1979; Hilliard and Tonlin 1979), music of consumers (Parrott 1982; Shatin 1970; Wheeler enjoyment (e.g., Gorn 1982; Wheeler 1985), and prior 1985); or stimulate retrieval of mood-congruent cog- mood (e.g., Eagle 1971; Goldberg and Gorn 1987). nitions (e.g., Gardner 1985). P6: Some responses to music are learned whereas others P3: The emotions perceived to be expressed in musical are inherent. stimuli are capable of evoking corresponding affective Responses to music are not likely to be determined solely by the learning environment or by human naThere is little debate that music can be perceived as ture, but rather by some interaction of the two. For expressing emotion-like qualities. It is more radical to instance, evidence of cross-cultural similarity in musuggest that properly structured music can generate sic expressiveness has been found (e.g., Clynes 1977; particular emotions. Yet, theory and supporting evi- Gundlach 1932). The key word, however, is “similarity,” because wide ranges of perceived expressions dence indicate that there are dynamic forms of expression specific to each emotion, and the ability of a are found even within samples from the same culture. stimulus to evoke an emotion depends heavily on the Therefore, greater understanding is needed of which purity of the expression (e.g., Clynes 1977, 1982). structural elements we respond to similarly as a result of our nature and which ones we respond to differP4: The influence of music on persuasiveness is greatest ently as a result of learning. under conditions of peripheral route processing and low reactions in listeners. cognitive involvement. This proposition is suggested by the elaboration like- Methodological Recommendations lihood model (e.g., Cacioppo and Petty 1989; Petty With an increasing concern in marketing about affecand Cacioppo 1986) and more specifically by such tive behavior, music-related research is likely to bestudies as Gorn’s (1982) and Park and Young’s (1986). come more common in the future. If so, several changes Though music’s role is likely to be greatest under con- in methodology are recommended to improve the reditions of high affective involvement and low cogni- liability and validity of the findings. tive involvement, it might also have a role even under First, efforts must be made to raise the level of conditions of high cognitive involvement. Specifi- experimental sophistication to account for the comcally, the direction and amount of issue-relevant plex workings of the musical stimulus. Significant new thinking could be influenced if music makes “mood knowledge will not be acquired until individual components of music are manipulated, examined, and/or controlled. Music has been treated too generally in most past marketing studies, with interest merely in 5The music itself is assumed not to be the product or an integral its presence or absence in some treatment. Even in part of it (as in records, concerts, and plays). If music is part of the studies in which a structural element was manipuproduct, for many people it would represent high affective involvement and central route processing because what they feel about the lated, control of other musical components was rarely music would be one of the main issues being evaluated (Cacioppo and adequate. Petty 1989, p. 81, 82). This observation does not mean a component should 100 / Journal of Marketing, October 1990 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms be studied independently of its normal context, but that greater use should be made of experimental designs that can discern the main and interaction effects of the various structural components of music (p,).6 A simple but variation-prone application of this approach is to make separate recordings of a song played repeatedly on some instrument, manipulating the variable of interest from one playing to the next (e.g., Hevner 1935b; Holbrook and Bertges 1981). Better yet, music hardware and software (synthesizers, sam- ments have been made in recent studies (Aaker, Stayman, and Hagerty 1986; Holbrook and Batra 1988; Madden, Allen, and Twible 1988). Yet, some re- searchers argue that these measures are still too verbally oriented and are unable to capture adequately the distinct nature of affective responses (Batra 1986; Stewart and Hecker 1988). Managerial Implications increasing the tempo of playback would not automat- A primary decision faced by marketing managers is whether or not to use music in such settings as the retail environment and in commercials. Music is likely to have its greatest effect when consumers have high affective and/or low cognitive involvement with the product. Product categories fitting this description for most consumers include jewelry, sportswear, cosmetics, and beer. Music would tend to have less effect ically increase the pitch). Already some simple uses of this technology have been reported (Holbrook and when consumers experience high cognitive involvement, such as when buying cars, appliances, PCs, plers, and sequencers) now allow complex orchestrated music to be programmed and numerous “treat- ments” to be produced that vary solely in some predetermined manner. Further, unlike conventional recording techniques, these methods allow the music components to be manipulated independently (e.g., Anand 1988; Holbrook and Corfman 1985). Second, future research should control for the ef- fect of music familiarity. If the influence of familiarity is of interest, it could be measured and treated as a predictor or moderator variable. If there is no interest in familiarity, its potential influence could be reduced by using either original compositions (e.g., Heyduk 1975; Holbrook et al. 1985) or pieces that pretesting indicates should be unfamiliar to listeners (e.g., Holbrook and Bertges 1981). Third, the influence of liking music should be ac- counted for in research designs. One approach is to use a nonrecursive structural model of halo effects (Holbrook 1983). Such a model is capable of accounting for the influence of individual musical components on perceptions and affect while estimating the distortion due to liking alone. “Affective overtones” may not always be a significant source of distortion, but such a model should be used to test for such bias. Fourth, methodology should be robust enough to account for nonmonotonic relationships between music components and affect. Evidence is now sufficient to lead one to expect a priori an inverted-U function between several music variables and affect. Given this expectation, the testing of only two levels of a music component may not adequately capture the true relationship between the variables of interest if it is non- cameras, and insurance. If the decision is made to use music, a choice must be made between developing music specifically for the occasion, using unfamiliar but previously written music, or employing well-known hits. At least in the advertising context, the choice has increasingly been to use hit songs (Rosenbluth 1988). In many in- stances, using hits is undoubtedly an effective way to draw attention and evoke positive responses. For at least two reasons, however, caution is justified before rushing to use well-known music. First, as discussed before, research on repetition suggests that what has been considered pleasant at one point in time can become much less enjoyable if repetition makes it too familiar. Indeed, recent empirical work shows this ef- fect occurs with contemporary pop songs (Russell 1987). Second, musicians are increasingly sensitive and vocal about the perceived misuse of their songs in commercials (Magiera 1988; Stroud 1988). If the decision is made to use previously written though unfamiliar music, information like that summarized in Table 3 can be helpful. Such information gives the marketer objective criteria that can be useful in selecting music most likely to express a particular mood. This process has been employed successfully under research conditions (e.g., Alpert and Alpert 1988; Vinovich 1975) and some companies profess to be providing this service on a commercial basis (e.g., The growing attention to the affective realm of Buddy and McCormick 1987; Rosenfeld 1985). consumer behavior brings with it a corresponding need When developing original music, marketers will monotonic. for assessment of and improvement in the quality ofdraw upon professional musicians’ intuition and trainthe measures employed. Some promising develop- ing. However, marketers need not be as dependent on musicians as they have been in the past. Computeri- zation and synthesizers eventually will enable even 6Similar attention to greater methodological sophistication also is nonmusicians to compose “free from the considerneeded for content analyses because a lack of detailed codes for music characteristics can distort statistical relationships and lead to ques- ations of musical notation and yet not without guiding tionable findings (e.g., Haley, Richardson, and Baldwin 1984). order” (Clynes and Nettheim 1982). The work of Music, Mood, and Marketing / 101 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms Clynes and others (e.g., Stipp 1988) indicates that in the future, distinction between a musician and a nonmusician may have more to do with one’s profession than with the ability to make expressive music. Though it has been long accepted that music affects human beings in various ways, only recently have researchers attempted to explore the presumed relationships empirically in marketing contexts. At this time the relevant body of literature is still meager. Be- cause the desire to understand human moods and their role in consumer behavior is likely to become ever more important, music, as a powerful emotional stimulus, should be studied more thoroughly. This review and the suggestions offered are hoped to encourage research on this subject. Appendix Many terms commonly are used to describe music. The following definitions are given to aid in understanding which variables have been linked with each of three basic factors (time, pitch, and texture). Pitch Melody is the succession of notes occurring over time through- out a song. Changes in melody can be either ascending (up in pitch) or descending (down in pitch). Leaps in pitch are possible, as are repetitions of single notes. Melodies can be played in a variety of keys, which are referred to by one of the first seven letters of the alphabet plus an indication of sharp or flat. Mode refers to the series of notes, arranged in a scale of ascending pitch, which provides the tonal substance of a song (Apel 1969, p. 535). In any given key, several modes are possible; the two best known in contemporary Western culture are the major and the minor modes. If melody is viewed as being “horizontal” because it occurs over time, harmony should be viewed as “vertical” because it refers to notes played simultaneously. Harmonies can be consonant or dissonant; the former refers to notes or chords producing an agreeable subjective reaction whereas the latter refers to sounds evoking an unpleasant reaction. Texture Though the time- and pitch-related components are likely to be the most essential features of music, it is texture that provides the “color” and aesthetic richness. Timbre is part of mu- sic’s texture and refers to the distinctiveness in tone that makes Time one instrument sound different from another even if they both Rhythm is the pattern of accents given to beats or notes in a the art of weaving together the unique sonic properties of multiple instruments to produce the complex textural fabric of a song. Tempo (plural tempi) is the speed or rate at which a play the same melody. Orchestration (or instrumentation) is rhythm progresses. Phrasing is the length of time a note sounds musical work. in comparison with the rhythmic period it occupies. For example, a staccato note sounds for only a small part of a bar whereas a legato note may be sustained until the next note Finally, volume also contributes to the texture of music. It can be used to make one note louder than others around it, to sounds. make a passage of notes louder than other passages (dynamics), or to make a whole song louder than others. REFERENCES Aaker, David A., Douglas M. Stayman, and Michael R. Hag-William R. Wilson, eds. Lexington, MA: D.C. Heath and erty (1986), “Warmth in Advertising: Measurement, Im-Company, 53-85. pact, and Sequence Effects.” Journal of Consumer ReBerlyne, D. E. (1960), Conflict, Arousal, and Curiosity. New search, 12 (March), 365-81. York: McGraw-Hill Book Company. Allen, Chris T. and Thomas J. Madden (1985), “A Closer (1971), Aesthetics and Psychobiology. New York: Look at Classical Conditioning,” Journal of Consumer Re-Appleton-Century-Crofts. search, 12 (December), 301-15. Budd, Malcolm (1985), Music and the Emotions. London: Alpert, Judy I. and Mark I. Alpert (1986), “The Effects of Routledge and Kegan-Paul. Music in Advertising on Mood and Purchase Intentions,” Buddy, Pamela and Moira McCormick (1987), “New Clip unpublished working paper 85/86, 5-4, University of Texas. Service for Bars, Restaurants,” Billboard, 99 (November and (1988), “Background Music as an 7), 62. Influence in Consumer Mood and Advertising Responses,” Butler, David M. (1973), “An Historical Investigation and in Advances in Consumer Research, Vol. 16, Thomas Bibliography K. of Nineteenth Century Music Psychology LitScrull, ed. Honolulu, HI: Association for Consumer Reerature,” unpublished doctoral dissertation, Ohio State Unisearch, 485-91. versity. Anand, Punam and Morris B. Holbrook (1985), “Chasing the Wundt Curve: An Adventure in Consumer Esthetics,” in Advances in Consumer Research, Vol. 12, Elizabeth Hirschman and Morris Holbrook, eds. Ann Arbor, MI: Association for Consumer Research, 655-7. Apel, Willi, ed. (1969), Harvard Dictionary of Music. Cambridge, MA: The Belknap Press of Harvard University Press. Batra, Rajeev (1986), “Affective Advertising: Role, Pro- cesses, and Measurement,” in The Role of Affect in Consumer Behavior, Robert A. Peterson, Wayne D. Hoyer, and Cacioppo, John T. and Richard E. Petty (1989), “The Elaboration Likelihood Model: The Role of Affect and AffectLaden Information Processing in Persuasion,” Cognitive and Affective Responses to Advertising, Patricia Cafferata and Alice M. Tybout, eds. Lexington, MA: Lexington Books. Clynes, Manfred (1975), “Communication and Generation of Emotion Through Essentic Form,” in Emotions-Their Parameters and Measurement, Lennart Levi, ed. New York: Raven Press, 561-601. (1977), Sentics: The Touch of Emotions. Garden City, 102 /Journal of Marketing, October 1990 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms NY: Anchor Press/Doubleday. (1980), “The Communication of Emotion: Theory of Sentics,” in Emotion, Theory, Research, and Experience, R. Plutchik and H. Kellerman, eds. New York: Academic Press, Inc., 271-300. (1982), “Specific Human Emotions Are Psycho- biologic Entities: Psychobiologic Coherence Between Emotion and Its Dynamic Expression,” Behavioral and Brain Sciences, 5 (September), 424, 425. and Nigel Nettheim (1982), “The Living Quality of Music,” in Music, Mind, and Brain, M. Clynes, ed. New York: Plenum Press, 47-82. Cooke, Deryck (1962), The Language of Music. London: Oxford University Press. Cupchik, Gerald C., Martin Rickert, and Julie Mendelson (1982), “Similarity and Preference Judgements of Musical Stimuli,” Scandinavian Journal of Psychology, 23 (4), 27382. Dowling, W. Jay and Dane L. Harwood (1986), Music Cognition. Orlando, FL: Academic Press, Inc. Eagle, Charles Thomas (1971), “Effects of Existing Mood and Order of Presentation of Vocal and Instrumental Music on Rated Mood Responses to That Music,” unpublished doc- toral dissertation, University of Kansas. Fontaine, Craig W. and Norman D. Schwalm (1979), “Effects of Familiarity of Music on Vigilant Performance,” Perceptual and Motor Skills, 49 (August), 71-4. Fraisse, Paul (1982), “Rhythm and Tempo,” in The Psychology of Music, Diana Deutsch, ed. New York: Academic Press, Inc., 149-80. Gabrielsson, Alf (1973), “Adjective Ratings and Dimension Analyses of Auditory Rhythm Patterns,” Scandinavian Journal of Psychology, 14 (4), 244-60. Gardner, Meryl P. (1985), “Mood States and Consumer Be- havior: A Critical Review,” Journal of Consumer Re- search, 12 (December), 281-300. Garfield, Bob (1988), “Too Much Ad Music Leaves Room for Hitting the Right Note,” Advertising Age, 59 (January 4), 46. Goldberg, Marvin E. and Gerald J. Gor (1987), “Happy and Sad TV Programs: How They Affect Reactions to Commercials,” Journal of Consumer Research, 14 (December), (1935b), “Expression in Music: A Discussion of Experimental Studies and Theories,” Psychological Review, 42 (March), 186-203. (1936), “Experimental Studies of the Elements of Expression in Music,” American Journal of Psychology, 48 (April), 246-68. (1937), “The Affective Value of Pitch and Tempo in Music,” American Journal of Psychology, 49 (October), 621-30. Heyduck, Ronald G. (1975), “Rated Preference for Musical Compositions as It Relates to Complexity and Exposure Frequency,” Perception and Psychophysics, 17 (January), 84-91. Hilliard, O. Mark and Philip Tonlin (1979), “Effect of Familiarity With Background Music on Performance of Simple and Difficult Reading Comprehension Tasks,” Perceptual and Motor Skills, 49 (December), 713-14. Holbrook, Morris B. (1983), “Using a Structural Model of Halo Effect to Assess Perceptual Distortion Due to Affective Overtones,” Journal of Consumer Research, 10 (September), 247-52. and Punam Anand (1988), “Aims, Concepts, and Methods in Marketing Research on Consumer Esthetics: The Effects of Tempo on Perceptual and Affective Responses to Music,” unpublished working paper. and Rajeev Batra (1987), “Assessing the Role of Emotions as Mediators of Consumer Responses to Advertising,” Journal of Consumer Research, 14 (December), 404- 20. and (1988), “Towards a Standardized Emotional Profile (SEP) Useful in Measuring Respons the Nonverbal Components of Advertising,” in Nonv Communication in Advertising, Sidney Hecker and D W. Stewart, eds. Lexington, MA: Lexington Books, 109. and Stephen A. Bertges (1981), “Perceptual Ve idicality in Esthetic Communication,” Communicatio search, 8 (October), 387-424. and Kim P. Corfman (1985), “Quality and Valu in the Consumption Experience: Phaedrus Rides Again Perceived Quality: How Consumers View Stores and M chandise, Jacob Jacoby and Jerry C. Olson, eds. Lex ton, MA: Lexington Books, 31-57. Gorn, Gerald J. (1982), “The Effects of Music in Advertising , William L. Moore, Gary N. Dodgen, and Willia on Choice Behavior: A Classical Conditioning Approach,” J. Havlena (1985), “Nonisomorphism, Shadow Feature Journal of Marketing, 46 (Winter), 94-101. Imputed Preferences,” Marketing Science, 4 (Summer), 33. Gundlach, Ralph H. (1932), “A Quantitative Analysis of Indian Music,” American Journal of Psychology, 44 (JanuHunt, James B. (1988), “An Experimental Study of the E ary), 133-45. of Music on Radio Commercial Performance,” in Proc (1935), “Factors Determining the Characterization ings of the Southern Marketing Association, 37-40. of Musical Phrases,” American Journal of Psychology, 47 Kellaris, James J. and Anthony D. Cox (1987), “The Ef (October), 624-43. of Background Music in Advertising: A Replication Haley, Russell I., Jack Richardson, and Beth M. Baldwin Extension,” in AMA Educators’ Conference Proceedi (1984), “The Effects of Nonverbal Communications in M. R. Solomon et al., eds. Chicago: American Marke Television Advertising,” Journal of Advertising Research, Association, 283. 24 (August/September), 11-18. and (1989), “The Effects of Backgrou Hecker, Sidney (1984), “Music for Advertising Effect,” PsyMusic in Advertising: A Reassessment,” Journal of C chology and Marketing, 1 (Fall/Winter), 3-8. sumer Research, 16 (June), 113-18. Henkin, Robert I. (1955), “A Factorial Study of the CompoKinnear, Joan K. (1959), “The Effects of Instrumental 387-403. nents of Music,” Journal of Psychology, 39 (January), 161Quality Upon Mood Response to Music,” unpubli 81. master’s thesis, Florida State University. (1957), “A Reevaluation of Factorial Study of Konecni, the Vladmimir J. (1982), “Social Interaction and Components of Music,” Journal of Psychology, 43 (Janusical Preference,” in The Psychology of Music, D ary), 301-6. Deutsch, ed. New York: Academic Press, Inc., 497-51 Hevner, Kate (1935a), “The Affective Character of the Krugman, Major Hebert, E. (1943), “Affective Response to M and Minor Modes in Music,” American Journal of Psyas a Function of Familiarity,” Journal of Abnormal S chology, 47 (January), 103-18. Psychology, 38 (April), 388-92. Music, Mood, and Marketing / 103 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms Langer, Susanne K. (1976), Philosophy in a New Key. Cambridge, MA: Harvard University Press. Madden, Thomas J., Chris T. Allen, and Jacquelyn L. Twible ness of Melodic Sequences: Contrasting Effects of Repetition and Rule Familiarity,” Psychology of Music, 14 (1), 17-32. Smith, Patricia and Ross Curow (1966), “‘Arousal Hypothesis’ and the Effects of Music on Purchasing Behavior,” verse Measurement Indices Under Different Processing Journal of Applied Psychology, 50 (June), 255-6. ‘Sets’,” Journal of Marketing Research, 25 (August), 242- (1988), “Attitude Toward the Ad: An Assessment of Di- Stewart, David W. and David H. Furse (1986), Effective Tele- 52. vision Advertising. Lexington, MA: Lexington Books. Magiera, Marcy (1988), “Nike to Keep ‘Revolution’ Despite and Sidney Hecker (1988), “The Future of ReSuit,” Advertising Age, 58 (August 3), 3, 55. search on Communication in Advertising,” in Nonverbal Meyer, Leonard B. (1967), Music, the Arts, and Ideas. ChiCommunication in Advertising, Sidney Hecker and David cago: University of Chicago Press. W. Stewart, eds. Lexington, MA: Lexington Books, 256Milliman, Ronald E. (1982), “Using Background Music to Af64. fect the Behavior of Supermarket Shoppers,” Journal of Marketing, 46 (Summer), 86-91. Stipp, David (1988), “Music-Composition Programs May Bring (1986), “The Influence of Background Music on the Bach to the Future,” Wall Street Journal, 69 (September 20), 28. Behavior of Restaurant Patrons,” Journal of Consumer Research, 13 (September), 286-9. Stout, Patricia and John D. Leckenby (1988), “Let the Music Nielzen, Soren and Zvaimir Cesarec (1982), “Emotional Play: Music as a Nonverbal Element in Television ComExpressions of Music as a Function of Musical Structure,” mercials,” in Nonverbal Communication in Advertising, Psychology of Music, 10 (2), 7-17. Sidney Hecker and David W. Stewart, eds. Lexington, MA: Park, C. Whan and S. Mark Young (1986), “Consumer ReLexington Books, 207-33. sponse to Television Commercials: The Impact of Involve- Stroud, Ruth (1988), “Mimicked Mess: Midler Case May Curb ment and Background Music on Brand Attitude Forma’Sound-Alikes’,” Advertising Age, 59 (July 11), 72. tion,” Journal of Marketing Research, 23 (February), 11- Swanwick, Keith (1973), “Musical Cognition and Aesthetic 24. Response,” Psychology of Music, 1 (June), 7-13. Parrott, A. C. (1982), “Effects of Paintings and Music, Both Van Stone, Joan K. (1960). “The Effects of Instrumental Tone Alone and in Combination, on Emotional Judgement,” PerQuality Upon Mood Response to Music,” in Music Therapy 1959, Erwin H. Schneider, ed. Lawrence, KS: Allen Press. ceptual and Motor Skills, 54 (April), 635-41. Petty, Richard E. and John T. Cacioppo (1986), “The Elab- Vinovich, George S. (1975), “The Communicative Signifioration Likelihood Model of Persuasion,” in Advances in cance of Musical Affect in Eliciting Differential PercepExperimental Social Psychology, Vol. 19, L. Berkowitz, tion, Cognition, and Emotion in Sound-Motion Media Mesed. New York: Academic Press, Inc. sages,” unpublished doctoral dissertation, University of Southern California. Rigg, Melvin G. (1940a), “Speed as a Determiner of Musical Mood,” Journal of Experimental Psychology, 27 (Novem- Walker, Edward L. (1981), “The Quest for the Inverted U,” ber), 566-71. in Advances in Intrinsic Motivation and Aesthetics, Hy I. (1940b), “The Effect of Register and Tonality Upon Day, ed. New York: Plenum Press, 39-70. Musical Mood,” Journal of Musicology, 2 (September), 49- Watson, K. Brantley (1942), “The Nature and Measurement 61. of Musical Meanings,” Psychological Monographs, 54 (2), 1-43. Rosenbluth, Jean (1988), “Singles Click as Madison Ave. Jingles,” Billboard, 35 (August 27), 1, 77. Wedin, Lage (1972), “A Multidimensional Study of PercepRosenfeld, Anne H. (1985), “Music, the Beautiful Disturber,” tual-Emotional Qualities in Music,” Scandinavian Journal Psychology Today, 19 (December), 48-71. of Psychology, 13 (4), 241-57. Russell, Philip A. (1987), “Effects of Repetition on the Fa- Wheeler, Barbara L. (1985), “Relationship of Personal Charmiliarity and Likeability of Popular Music Recordings,” acteristics to Mood and Enjoyment After Hearing Live and Psychology of Music, 15 (2), 187-97. Recorded Music and to Musical Taste,” Psychology of MuScherer, Klaus R. and James S. Oshinsky (1977), “Cue Utisic, 13 (2), 81-92. lization in Emotion Attribution From Auditory Stimuli,” Wintle, Regie R. (1978), “Emotional Impact of Music on Motivation and Emotion, 1 (December), 331-46. Television Commercials,” unpublished doctoral dissertaSeidman, Steven A. (1981), “On the Contributions of Music tion, University of Nebraska. to Media Productions,” Educational Communication and Yalch, Richard F. (1988), “Effects of Store Music on ShopTechnology Journal, 29 (Spring), 49-61. ping Behavior,” The Channel of Communication, 4 (SumSewall, Murphy A. and Dan Sarel (1986), “Characteristics of mer), 7-9. Radio Commercials and Their Recall Effectiveness,” Jourand Eric Spangenberg (1988), “An Environmental nal of Marketing, 50 (January), 52-60. Psychological Study of Foreground and Background Music Shatin, Leo (1970), “Alternation of Mood Via Music: A Study as Retail Atmospheric Factors,” in AMA Educators’ Conof the Vectoring Effect,” Journal of Psychology, 75 (May), ference Proceedings, Alf W. Walle, ed. Chicago: Ameri81-6. can Marketing Association, 106-10. Simpkins, John D. and Jack A. Smith (1974), “Effects of Mu-Herbert (1973), Sight, Sound, and Motion: Applied MeZettl, sic on Source Evaluations,” Journal of Broadcasting, 18 dia Aesthetics. Belmont, CA: Wadsworth Publishing Co., (Summer), 361-7. Inc. Smith, Karen C. and Lola L. Cuddy (1986), “The Pleasing- 104 /Journal of Marketing, October 1990 This content downloaded from 132.174.251.44 on Thu, 30 Sep 2021 12:32:50 UTC All use subject to https://about.jstor.org/terms Reprint No. JM544107