Listen to the Music: Its Impact on Affect, Perceived Time Passage, and Applause

ABSTRACT - This paper integrates a series of interesting findings related to how one responds to a musical stimulus. In particular, this paper focuses on one=s affect toward a musical stimulus (Affectms) and uses information theory to examine the complexity of a music stimulus. Two models are then developed which provide a general framework for understanding applause duration and for understanding the perceived time duration of a musical stimulus.

Citation:

Brian Wansink (1992) ,"Listen to the Music: Its Impact on Affect, Perceived Time Passage, and Applause", in NA - Advances in Consumer Research Volume 19, eds. John F. Sherry, Jr. and Brian Sternthal, Provo, UT : Association for Consumer Research, Pages: 715-718.

Advances in Consumer Research Volume 19, 1992      Pages 715-718

LISTEN TO THE MUSIC: ITS IMPACT ON AFFECT, PERCEIVED TIME PASSAGE, AND APPLAUSE

Brian Wansink, Dartmouth College

ABSTRACT -

This paper integrates a series of interesting findings related to how one responds to a musical stimulus. In particular, this paper focuses on one=s affect toward a musical stimulus (Affect_ms) and uses information theory to examine the complexity of a music stimulus. Two models are then developed which provide a general framework for understanding applause duration and for understanding the perceived time duration of a musical stimulus.

Few topics in marketing are of wider appeal than those which examine the impact of music on processing, preference, and behavior. Though much of the work done in this area is either inconclusive or methodologically flawed, the three papers described below provide solid methodological advances toward better understanding of music on culture (Blair and Hatala 1992), on time perception (Kellaris and Altsech 1992), and on affect (Kellaris 1992). In further examining the contributions of the issues raised by these papers, this paper will raise additional insights related to the results of the studies in the latter two papers.

ESTIMATING THE TIME DURATION OF MUSIC

Music Volume and Gender Effects

The research of Kellaris and Altsech (1992) suggest that the estimated duration of a song is systematically underestimated by females if the song is played at a soft volume (60 dB). The differential impact of music on gender is well documented. Not only are females more sensitive to higher frequencies, and to increases in musical amplitude, but it also appears that they may generate less processing of auditory information in the left hemisphere of their brain. Though Kellaris and Altsech hypothesized that increases in the volume of music would cause females to overestimate the length of the music, in this study it resulted in them accurately estimating the time along with both the males who listened to loud music and those who listened to soft music.

The most interesting result of this research is the degree to which females underestimated the time duration of soft music. The critical issue that this result raises, however, may have less to do with gender differences and volume than it does with one=s affect toward the musical stimulus (Affect_ms) and volume. The music used by Kellaris and Altsech (1992) was an instrumental piece described as "light pop-rock" or what might be considered "easy listening." In general, let us assume that females are more attracted to this type of music than males (who tend to instead "prefer loud >Heavy Metal= music"). This is a reasonable assumption, given that informal interviews confirm that this gender bias is reflected both in radio listenership and in the music purchases of "light pop-rock."

The Interaction of Affect_ms and Comfort

It is worth positing that the critical mediating variable when assessing time duration estimates would be Affect_ms: The more favorable one is toward the music, the shorter its perceived duration will be. In effect, the time spent listening to the music is more enjoyable, it does not seem to "drag," and it is accompanied by anticipation, not unlike what occurs when one views a favorite advertisement (Rossiter and Percy 1987). Therefore, given that females might favor "light pop-rock" more than males, we might expect a much different pattern of results if all the subjects were instead exposed to heavy metal music. In this case, males would tend to underestimate the duration of the music.

Though such a rationale would predict a main effect in the difference of the estimates between males and females, it does not explain the results of why we find that soft music is perceived by females as having less duration than loud music.

Rather than focusing on loudness as a variable, it may be that loudness should be viewed more in terms of the specific impact it has on an individual. Since females are more sensitive to loud music than males, at some decibel level loud music will result in greater annoyance or even discomfort. Under these conditions, one=s general comfort-level might suggest an even more robust model. If "comfort" is used as a moderating variable, it would result in gender interacting with volume to impact on "comfort," which in turn influences one=s perception of time duration. Figure 1 provides a model of these variables. This interpretation of the interaction of gender and volume on time duration is consistent with the results reported by Kellaris and Altsech (1992).

Given this interpretation of time duration, the type of music listened to can either increase or decrease one=s Affect_ms and in turn affect perceptions of time duration. Affect_ms would have an impact on one=s perception of a musical stimulus=s duration both directly and indirectly through it=s interaction with one=s comfort level. (In general, one=s "comfort" is thought to be influenced by such factors as physical comfort and psychological comfort -- Thompson 1975. As antecedents to comfort, these two variables encompass such disparate factors such as music loudness and physical crowding). To the degree that loudness can affect one=s comfort, it can affect one's perceived duration of music. In addition, as the results of the study suggest, if a high level of comfort (e.g., the music is soft) is coupled with high Affect_ms, we would expect one to underestimate the duration of the music stimulus. In summary, when one is experiencing a low level of comfort, or has low Affect_ms, we would expect a higher estimate of the musical stimuli=s duration than if one is "comfortable" and "likes the music." Such a model makes the results much more general and less driven by gender issues.

FIGURE 1

BASIC DETERMINANTS OF PERCEIVED MUSIC DURATION

Insofar as gender can impact both Affect_ms and one=s comfort level (through greater sensitivity to loud noises, etc.), we would expect the results found in the study.

The primary contribution of Kellaris and Altsech (1992) is that its findings suggest a new paradigm (Figure 1) that is rich for consideration. Fruitful future research in this area would test the validity of the model proposed in Figure 1 and would further investigate this issue of an audience member=s "comfort."

THE IMPACT OF STIMULUS COMPLEXITY ON AFFECT_MS AND APPLAUSE

Kellaris (1992) presents a clever and rigorous field-study which uses a balanced experimental design that carefully accounts for confounds, and unobtrusively obtains a behavioral measure of Affect_ms by observing the duration of audience applause. The results indicated that the impact of a song=s tempo on a crowd=s applause follows a Wendt curve. Though this is hypothesis is consistent with past research, the results indicated that the downward-sloping effect past the asymptote was highly significant for songs in minor chords but not those played in major chords.

Information Theory and Stimulus Complexity

To make the basic Wendt model more generalizable to account for these results, it is useful to relate it with information theory (Shannon and Weaver 1949). In doing so, let us view a musical stimulus as providing a certain amount of information (Wansink 1984). For instance, a musical stimulus that is highly organized or structured (e.g., it has a melody that is repeated, is constructed with familiar musical elements, etc.) contains less information because it is more predictable than a less organized musical stimulus, the extreme case would be represented by random sounds at random volumes).

Given this orientation of information theory, it is reasonable to look at this distinction between major and minor chords in terms of their complexity as stimuli. In general, one can say these two differ on a dimension we could label as objective stimulus complexity. Because major chords are more commonly encountered and are more familiar (or predictable) to us, they can be seen as containing less information and as being less complex than minor chords. This variable of objective stimulus complexity could be measured using an information theory approach (Wansink 1984).

To restate a preceding point, since minor chords are less common and therefore less predictable, they contain more information than major chords and can therefore be viewed as objectively more complex than major chords. The results of Kellaris (1992) support this. If we assume that the songs played in major chords are at the top of an Affect_ms asymptote (as represented by a Wendt curve with objective stimulus complexity on the x-axis and Affect_ms on the y-axis), those songs played in minor chords would be on the downside of this curve because they are objectively more complex. This is consistent with Kellaris' quadratic results: Songs in minor chords exhibited this nonmonotonic relationship with applause duration while major chords did not.

FIGURE 2

PROPOSED DETERMINANTS OF AFFECT TOWARD A MUSICAL STIMULUS

Stimulus Factors and Audience Factors Affecting Complexity

Because of the limitations of even such well-executed field studies as Kellaris=, the data are not available to examine what audience-related factors might affect this measure of applause. Familiarity, fatigue, involvement, and arousal would all have effects not only on Affect_ms but also on Perceived Stimulus Complexity (see Figure 2.). From an information theory perspective, PSC can also be increased by any stimulus factors that increase the amount of uncertainty associated with a stimulus, such as unfamiliar tempos or chords, or irregular tempos and chord changes. Up to a point (see Figure 3 -- point x'), increases in objective stimulus complexity (such as those that are associated with chord changes and tempo changes) have a positive impact on Affect_ms. After this point, however, additional changes will have a negative impact, unless accompanied by increases in a relevant audience state variable (such as arousal, familiarity, etc.) which would represent a shift in Figure 3 from State A to State B.

Given the detailed coding of Kellaris= (1982) study, a number of general conclusions could be reached by analyzing the indirect impact of these different stimulus-related variables (tempo, duration, meter, modality, presence of lyrics, instrumentation, type, etc.). In addition, this data could be further analyzed by examining the presentation order of songs and by running a regression which includes a lagged-term of applause (e.g., applause length on the previous song) to assess the impact of arousal and also of fatigue. In effect, the more aroused a crowd (see Figure 3), the more they are likely to respond toward a song that is more complex than a song at point x (but less complex than one at point x").

The impact of these factors will undoubtedly vary across people. As suggested by Anand and Holbrook (1986), the more aroused an individual is, the more likely it is that he or she will desire a stimulus that is objectively more complex (to point x" in Figure 3). This shift from State A to State B has different implications for how favorably one will be toward a complex stimulus (Affect_ms). It is not unreasonable that such shifts would also occur as a person=s familiarity with the music changes or as his or her comfort level changes (Kellaris and Altsech 1992).

SUMMARY

Though music represents a complex gestalt that one "pulls apart . . . only at one=s peril" (Anand and Holbrook 1986), a number of fascinating findings exist that can be best understood by broadening and generalizing the context in which they are observed. This paper integrates a series of interesting findings by providing an information theory approach to Affect_ms and by suggesting how this interpretation can influence Affect_ms, which in turn affects other variables such as perceived time duration.

FIGURE 3

AUDIENCE FACTORS AND THEIR RELATIONSHIP BETWEEN OBJECTIVE STIMULUS COMPLEXITY & AFFECT TOWARD A MUSICAL STIMULUS

REFERENCES

Anand, Punam and Morris B. Holbrook (1986), "Chasing the Wundt Curve: An Adventure in Consumer Esthetics," In Richard Lutz, Advances in Consumer Research (Volume 13), Provo, UT: Association for Consumer Research.

Blair, M. Elizabeth (1992), "The Use of Rap Music in Children=s Advertising," In John Sherry and Brian Sternthal, Advances in Consumer Research (Volume 19), Provo, UT: Association for Consumer Research.

Kellaris, James J. (1992), "Consumer Esthetics Outside the Lab: Preliminary Report on a Musical Field Study," In John Sherry and Brian Sternthal, Advances in Consumer Research (Volume 19), Provo, UT: Association for Consumer Research.

Kellaris, James J. and Moses B. Altsech (1992), "The Experience of Time as a Function of Musical Loudness and Gender of Listener," In John Sherry and Brian Sternthal, Advances in Consumer Research (Volume 19), Provo, UT: Association for Consumer Research.

Kellaris, James J. and Robert J. Kent (1991), "Exploring Tempo and Modality Effects on Consumer Responses to Music," In Rebecca Holman and Michael Solomon, Advances in Consumer Research (Volume 18), Provo, UT: Association for Consumer Research, 243-248.

Rossiter, John R. Rossiter and Larry Percy (1987), Advertising and Promotion Management, New York: McGraw-Hill.

Shannon, Claude E. and Warren Weaver (1949), The Mathematical Theory of Communication, Urbana: University of Illinois Press.

Thompson, Richard F. (1975), Introduction to Physiological Psychology, New York: Harper and Row.

Wansink, Brian Charles (1984), Information Theory as a Communication Paradigm and its Application to Advertising, unpublished thesis, Drake University, Des Moines, Iowa. Ann Arbor, MI: University Microfilms, Inc.

----------------------------------------