Chasing the Wundt Curve: an Adventure in Consumer Esthetics

Punam Anand, New York University
Morris B. Holbrook, Columbia University
ABSTRACT - This paper provides a brief progress report on research investigating the effects of musical tempo on perceived activity and affect. After introducing some key issues, we describe our experimental procedures and report preliminary support for (1) a logarithmic or power function relating tempo to perceived activity, (2) a nonmonotonic Wundt curve relating perceived activity to affect, and (3) a shift in this Wundt curve in response to a change in arousal. We conclude with a discussion of remaining issues that deserve investigation.
[ to cite ]:
Punam Anand and Morris B. Holbrook (1986) ,"Chasing the Wundt Curve: an Adventure in Consumer Esthetics", in NA - Advances in Consumer Research Volume 13, eds. Richard J. Lutz, Provo, UT : Association for Consumer Research, Pages: 655-657.

Advances in Consumer Research Volume 13, 1986      Pages 655-657


Punam Anand, New York University

Morris B. Holbrook, Columbia University

[The second author gratefully acknowledges the support of the Columbia Business School's Faculty Research Fund.]


This paper provides a brief progress report on research investigating the effects of musical tempo on perceived activity and affect. After introducing some key issues, we describe our experimental procedures and report preliminary support for (1) a logarithmic or power function relating tempo to perceived activity, (2) a nonmonotonic Wundt curve relating perceived activity to affect, and (3) a shift in this Wundt curve in response to a change in arousal. We conclude with a discussion of remaining issues that deserve investigation.


Many marketers and designers of advertisements have traditionally employed music in their programs to encourage consumer purchasing and to enhance persuasiveness. In particular, music has been used to improve store image (Milliman 1982) and to increase message acceptance (Galizio and Hendrick 1972). Despite the widespread use of music, research documenting its effects has remained limited. One reason for this gap in the literature is that the study of music as a form of communication poses several difficulties. The term "music" denotes a wide range of stimuli (for example, pitch, frequency, density, modality, and tempo) that hang together in an extremely complex gestalt. One pulls apart this gestalt only at one's peril. Nevertheless, through the isolation of individual variables, some provocative findings have begun to emerge. Our general impression is that such findings have appeared most frequently and reliably in work exploring the effects of tempo on perceived activity and affect.

Zimny and Weidenfeller (1963) measured subjects' heart rate and skin resistance to three pieces of music, found that faster music decreased skin resistance, but detected no such changes for neutral or calming music. They interpreted their data as indicating that more exciting music produces an increase in physiological arousal. Similarly, Seidman (1981) found that fast music tends to elicit happiness and excitement, whereas slow music tends to elicit more somber moods such as calmness and sadness. However, there is some evidence that fast tempos may also have debilitating effects. Wakshlag, Reitz, and Zillmann (1982) found that, although faster background music increased selective exposure to an educational program, slower music yielded less decrease in visual attention and information acquisition. In an applied setting, Milliman (1982) found that slower background music resulted in longer in-store traffic flow and increased sales volume.

In the field of consumer research, several studies have shown effects of musical tempo on perceived activity (e.g., Holbrook 1982, 1983; Holbrook and Huber 1983, Huber and Holbrook 1979) and of tempo or perceived activity on affect (e.g., Holbrook 1981, 1983; Holbrook and Corfman 1985; Huber and Holbrook 1980). However, most of the effects reported in the consumer literature have been only directional (e.g., perceptual responses to two levels of tempo) or linear (e.g., affective responses to a range of perceived activity). Yet these linear monotonic effects inadequately represent the underlying theory from conventional psychophysics positing a logarithmic relation between tempo and perceived activity (Fechner 1860, Stevens 1975). Moreover, linear or monotonic relationships may seriously distort the nonmonotonic impact of perceived activity on affect (Wundt 1874) described by psychobiologists (e.g., Berlyne 1971, 1974). Both these aspects of the work on musical tempo in consumer esthetics therefore require repair. In addition, further aspects of the effects of tempo deserve exploration.

Tempo and Perceived Activity

Briefly, psychophysics is concerned with the quantitative measurement of sensation. Fechner's Law (1860), modeling the relationship between the strength of an external stimulus and subjective expressions of intensity, is an extension of Weber's Law (1846). Weber proposed that a fixed percentage must be added to a stimulus in order to produce a just noticeable difference (JND). Fechner assumed that each time the stimulus is increased by one JND, the sensation increases by a constant increment. These assumptions imply that perceived activity is a linear function of the logarithm of tempo. By contrast, Stevens (1975) has argued vehemently that magnitude estimates are related to stimulus intensity by a power law by which equal percentage increases in the stimulus produce equal percentage increases in the response so that (for our purposes) a linear relationship would exist between the logarithm of tempo and the logarithm of perceived activity. Baird and Noma (1978) show that, in accord with the debate between Fechner and Stevens, the relative appropriateness of the logarithmic and power laws depends on whether one measures sensation or perception on an equal-interval or a ratio scale. Hence, when scale properties remain in doubt (as in the present case), the choice of models becomes at least partly an empirical issue. The difference between the logarithmic and power functions appears in Figure 1.



Perceived Activity and Affect

Abundant evidence connects affect to fluctuations in arousal (Berlyne 1967, 1971, 1974). Stimulus patterns that give rise to very low or very high levels of arousal will be unpleasant, aversive, and disliked, whereas those that lead to moderate arousal levels will be more pleasurable. Given a direct connection between perceived activity and arousal, the resulting relationship should follow the Wundt curve presented in Figure 2.



The Moderating Effect of General Arousal

In addition, esthetic theory (Berlyne 1971) also suggests that the nomonotonic Wundt curve should shift toward the right as general arousal increases (due to changes in the situational context). This hypothesized relationship appears in Figure 3.




To test these propositions, we varied the tempo of a piece programmed for a Casio keyboard. This instrument was selected because it permits a carefully controlled and precisely replicable range of speed settings. The fourteen tempo levels ranged from 56.6 beats/minute (the Casio's slowest speed) to 348.3 beats/minute (judged to be faster than any tempo likely to be encountered in everyday experience). In the spirit of Weber's Law, each tempo level was determined by adding a constant percentage increment (15%) to the preceding level. Subjects heard all fourteen different levels of tempo in randomly determined orders (cf. Stevens 1975).

General arousal was manipulated by assigning different distraction tasks to different subjects. One group of subjects sat still with their eyes closed while the experimenter changed tape cassettes. The other group worked on four anagram puzzles (related to musical terms and intended to be mildly arousing) between tempo presentations. Thus, in the manner typical of work in psychophysics, tempo varied within subjects (cf. Stevens 1975). By contrast, general arousal was manipulated between treatment groups. Forty-four subjects participated in the experiment. After subjects listened to each tempo, they filled out two instruments intended to measure perceived activity and affect. The affect measure consisted of six bipolar adjectival scales (e.g., unpleasant/ pleasant). The perceived activity measure contained eight bipolar adjective scales (e.R., slow/fast). After listening to all the tapes, subjects also provided information on demographics, familiarity with the music, and past musical training.


Extensive data analyses have generated preliminary results that appear strongly to support the three hypothesized relationships. Specifically, (1) perceived activity varies directly with the logarithm of tempo; (2) affect responds nonmonotonically to perceived activity (and to the log of tempo) with a peak at moderate levels of tempo; and (3) this peak shifts from a lower to a higher position on the perceived activity continuum with increases in general arousal. By contrast, demographic variables, familiarity, and past musical training have shown no impact on perceived activity, affect, or the moderating effect of arousal.


Amidst these promising preliminary findings, some important issues remain for further analysis. Three appear to deserve special attention.

First, the superiority of the logarithmic versus the power law should depend on whether our index of perceived activity behaves as a category (equal interval) scale or as a scale of numerical magnitude (ratio) estimation (Baird and Noma 1978, Stevens 1975). We can muster arguments for either interpretation of the activity index and therefore feel that the issue should be decided by a careful comparison of the fits obtained by logarithmic and power functions. This analysis still awaits completion.

Second, though most psychophysicists work with group averages based on within-subject designs (Stevens 1975), we shall be more comfortable with our results for the moderating effect of general arousal on the relationship between tempo and affect if these findings retain their significance when tested by the appropriate repeated-measures design (cf. Latour and Miniard 1983).

Third, theory and previous studies suggest the likelihood of sequence effects on both perceived activity (Stevens 1975) and affect (McClelland, Atkinson, Clark and Lowell 1976). Specifically, one's impression of speed or liking for a piece of music should depend in part on the tempo of the last piece heard. The results of preliminary regression analyses suggest that such order effects do indeed occur, but our statistical procedures need further refinement.

In sum, these and other issues still await investigation before we can report our results in full. More generally, the glimpse provided here appears to argue for he more widespread and systematic exploration of the psychophysical aspects of consumption experiences. With rare exceptions (Buchanan and Morrison 1984; Moskowitz, Jacobs, and Firtle 1980), psychophysics has remained a neglected area in consumer research. We hope this study encourages a broader adoption of psychophysical methods to increase our understanding of the relationship between different aspects of music and sensations in particular, the nature of consumer esthetics in general, and ultimately the full spectrum of experience involved in the phenomena of consumption.


Baird, John C. and Elliot Noma (1978) Fundamentals of Scaling and Psychophysics, New York: Wiley.

Berlyne, D.E. (1967), Arousal and Reinforcement in, Nebraska Symposium on Motivation. e.d., D. Levine, Lincoln, Nebraska: University of Nebraska Press. Appleton-Century-Crofts.

Berlyne, D. E. ed. (1974), Studies in the New Experimental Aesthetics, New York: Wiley.

Buchanan, Bruce S. and Donald G. Morrison (1984), "Taste Tests: Psychophysical Issues in Comparative Test Design," Psychology and Marketing, 1 (1), 69-91.

Fechner, G.T. (1860), Elemente der Psvchophvsik, Vol.l, available in English translation as Elements of PsschoDhvsics, New York: Holt, Rinehart and Winston, 1966.

Galizio, M. and Clyde Hendrick (1972), "Effect of Musical Accompaniment on Attitude: The Guitar as a Prop for Persuasion," Journal of Applied Psychology, 2, 4, 350-354

Holbrook, Morris B (1981), "Integrating Compositional and Decompositional Analyses to Represent the Intervening Role of Perceptions in Evaluative Judgments," Journal of Marketing Research, 18 (February), 13-28.

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), 247-252.

Holbrook, Morris B. and Kim P. Corfman (1985), "Quality and Value in the Consumption Experience: Phaedrus Rides Again," in Perceived Quality: How Consumers View Stores and Merchandise, ed. Jacob Jacoby and Jerry C. Olson, Lexington, MA: Lexington Books, 31-57.

Holbrook, Morris B. and Joel Huber (1983), "Detecting the Differences in Jazz: A Comparison of Methods for Assessing Perceptual Veridicality in Applied Aesthetics," Empirical Studies of the Arts, 1 (1), 35-53.

Huber, Joel and Morris B. Holbrook (1983), "Using Attribute Ratings for Product Positioning: Some Distinctions Among Compositional Approaches," Journal of Marketing Research, 16 (November), 507-516.

Huber, Joel and Morris B. Holbrook (1980), "The Determinants of Esthetic Value and Growth, n in Advances in Consumer Research, Vol. 7, ed. Jerry C. Olson, Ann Arbor: Association for Consumer Research, 121-126.

Latour, Stephen A. and Paul W. Miniard (1983), "The Misuse of Repeated Measures Analysis in Marketing Research, n Journal of Marketing Research, 20 (February), 45-57.

McClelland, D.C., J. W. Atkinson, R. A. Clark, and Edgar L. Lowell (1976), The Achievement Motive, New York: Irvington Publishers, Inc.

Milliman, R. E.(1980), "Using Background Music to Affect the Behavior of Supermarket Shoppers, n Journal of Marketing, 46 (Summer), 86-91.

Moskowitz, H. R., 3. Jacobs, and N. Firtle (1980), "Discrimination Testing and Product Decisions," Journal of Marketing Research, 17 (February), 84-90.

Seidman, S. (1981), "On the Contributions of Music to Media Productions", Educational Communication and Technology Journal, 19, 49-61.

Stevens, S. S. (1975), Psychophysics: Introduction to its Perceptual, Neural, and Social Aspects, et. Geraldine Stevens, New York: John Wiley and Sons.

Wakshlag, J. J., R. J. Reitz, and Dolf Zillmann (1982), "Selective Exposure to and Acquisition of Information from Educational Television Programs as a Function of Appeal and Tempo of Background Music", Journal of Educational Psychology, 14, 666-677.

Weber, E. H. (1846), "Der Tastsinn und das Gemeinfuhl," in Handworterbuch der Phvsiologie, Vol. 3, ed. R. Wagner, vieweg, Braunschweix, 481-588.

Wundt, W.M. (1874), Grundzuge der Phvsiologischen Psychologie, Leipzig: Engelmann.

Zimny, G. and E. Weidenfeller (1963), "Effects of music on GSR and Heart Rate," American Journal of Psychology, 76, 311-314.