Advances in Consumer Research Volume 21, 1994 Pages 514-518
THE INFLUENCE OF MOOD AND GENDER ON CONSUMERS' TIME PERCEPTIONS
James J. Kellaris, University of Cincinnati
Susan Powell Mantel, University of Cincinnati
This study explores the subjective experience of time passage as a function of mood and gender. Because consumers often hear music used as a "time filler," we manipulated instrumental background music in a lab experiment to induce mood states in male and female listeners. Gender was found to moderate the influence of mood on event duration estimates. Specifically, event durations were underestimated to a greater extent by females in less (versus more) positive mood states. We discuss potential commercial applications of our findings and develop directions for future consumer research on time perception.
INTRODUCTION & BACKGROUND
The experience of time pervades every aspect of human life (Bergadaa 1990), including consumptive activities. Although time is measured objectively, its passage is experienced subjectively (Hornik 1984). Under some circumstances, the disparity between objective "clock time" and perceived time can be wide. As we will discuss later, this subjective distortion can have interesting theoretic implications and important commercial consequences.
Time perception may be influenced by internal states induced by external stimuli (Levin and Zakay 1989). For example, drugs that increase (or decrease) mental activity have been shown to cause an overestimation (or underestimation) of elapsed time (Fraisse 1963). In addition, research has shown that time perceptions can be influenced by individual characteristics that moderate the impact of internal states (Block 1990).
Psychological research has suggested two processes that may underlie time perception: amount of information processed and allocation of attentional resources. Several researchers (Block 1990; Block 1989; Levin and Zakay 1989; Ornstein 1969) have explored time perception from an information processing perspective. This stream of research suggests that variation in the amount of information processed, stored, and subsequently retrieved from a time period can influence the estimation of the interval's duration. If a subject can recall more things happening during a time interval, s/he will infer more elapsed time. Zakay (1989) suggests that subjects who allocate more attention to processing stimulus events will have a reduced capacity to attend to their "internal clock;" thus, their estimates of lapsed time will be less accurate. As a subject becomes more involved in a processing task, less attention is paid to the passage of time. Hence time estimation will rely more on inferences based on the amount of information encountered/processed/ stored/retrieved from the time interval, and less on a cognitive timer. Internal states such as mood should influence attention and information processing, and thus time perceptions.
There are two predominant theories concerning the influence of mood (or emotion) on memory. Leventhal (1981) advocates a theory of emotion that holds that the human mental structure uses two separate, parallel systems to process information: one emotional and one cognitive. Thus, emotionally charged information may be stored twice within the mental structure; once within the emotional system and once within the cognitive system. Ger (1989), on the other hand, represents the storage of emotional and cognitive information onto separate "nodes" within one mental structure. In either case, research suggests that emotion can influence the organization, storage, and retrieval of cognitive information (Isen 1989; Ger 1989; Matlin 1989) and thus, should affect time perceptions.
The "Pollyanna Principle" states that pleasant information is typically processed, stored, and recalled more efficiently than less pleasant information (Matlin 1989). In general, mood states tend to influence behavior, judgment, and recall; however, negative mood states tend to have more multifarious effects than positive moods (Gardner 1985). According to Gardner's review, one might expect positive affect generated by external stimuli to enhance recall of information, while less positive affect might decrease coding and storage of information. Therefore, with less information stored during a time interval in which less positive affect is experienced, a shorter duration estimate can be expected (Kellaris and Kent 1992).
Several studies have investigated gender differences in time perception. Two studies found that men estimate short time intervals more accurately than women (e.g., Rammsayer and Lustnauer 1989; Krishnan and Saxena 1984). In addition, Krishnan and Saxena's results indicate that females tend to underestimate the time interval as compared to males. The observed effect was explained from a socialization perspective: males tend to learn time-consciousness as a result of their social and work experiences, which historically have involved more structured scheduling and greater time-pressures.
Gender has also been found to interact with mood to influence behavior (Maccoby and Jacklin 1974; Cox and Kellaris 1988) and memory (Clark and Teasdale 1985). Specifically, Clark and Teasdale found that women are more likely than men to recall pleasant words when in a happy mood and less likely to recall pleasant information when in a depressed mood. Several explanations for the interaction of mood and gender have been suggested. Clark and Teasdale use an associative network model to explain the gender difference and suggest that women may have stronger links between mood and cognitive nodes in memory, either because they are used more frequently or more intensely. An alternative explanation of mood-gender interaction is proposed by Rothkopf and Blaney (1991). The authors suggest that mood congruent recall requires at least some awareness of one's own mood state. Further, women are purportedly more likely to be aware of (and effected by) their own mood state in recall of information. In general, the psychological literature suggests that gender differences may have both a main effect on perceived time and a moderating effect on mood's influence.
The exploratory study reported in this paper manipulates mood via background music in ads (Alpert and Alpert 1989) and measures gender to assess the main and interactive effects of these two variables on time perception. Because mood effects are often ambiguous or inconsistent across studies, we cannot state a priori expectations concerning a main effect of mood on time perception. We can, however, anticipate a main effect of gender on the estimation of event durations based on the psychological literature. Specifically, females should underestimate time intervals to a greater extent than males. In addition, because gender and mood have been shown to exert an interactive influence on recall (which may be antecedent to time inferences), it is expected that females will attribute longer durations to stimulus events that elicit a more positive mood, and shorter durations under less positive mood. Further, since males are less likely to be aware of (and effected by) their own mood states, it is expected that their time estimates will remain relatively invariant across mood states.
VARIANCE ANALYSIS (ANOVA) ON DURATION ESTIMATES
Our experiment explored the influence of music-induced mood and gender of listener on time perceptions using a 2 (less/more positive mood) by 2 (male/female) between-subjects design. Subjects were exposed to simulated radio ads that contained mood-inducing background music and provided a deceptive premise for the experimental task. Perceived time passage, measured retrospectively as estimates of ad durations, was the dependent variable. The procedure involved randomly assigning subjects to treatments, exposing them individually to test ads via headsets, and asking them to complete a brief, self-administered questionnaire.
One hundred fourteen undergraduate students were recruited from a subject pool at a large midwestern university. Subjects were naive to the purpose of the study. Extra course credit was offered as an incentive to participate. The gender distribution was about 66% male, 34% female. Ages ranged from 18 to 42 years, with a median age of 21. Complete responses were obtained from all but four of the subjects for a final sample size of one hundred ten (N=110).
A set of simulated radio ads were created to manipulate mood. Listening to ads also provided a deceptive premise for the mood-induction procedure. To create the stimuli, we wrote ad copy for two fictitious restaurants: a Chinese restaurant, and an elegant British-American restaurant. The copy for each restaurant ad was as similar as possible, containing about one hundred words, the name of the restaurant followed by a slogan, and other information designed to be affectively neutral. Each ad was read by the same announcer, recorded on the same equipment under identical conditions, and lasted for exactly 32 seconds.
Brief excerpts of instrumental music were selected to manipulate mood. The music was "mixed" with ad copy in a sound studio to create a set of twelve test ads. Twenty-five student subjects listened to and rated the mood of the ads in a pretest. We retained the highest and lowest rated ads for each restaurant. This stimulus construction procedure resulted in a set of multiple test ads for each level of mood. Multiple exemplars were used within each cell to increase construct validity (Cook and Campbell 1979).
Subjects were processed individually in a listening lab. The purpose of the study was not disclosed. Subjects were told only that they would be hearing a radio ad about which they would be asked some "opinion questions." Subjects did not anticipate the time estimate measure. Printed instructions reminded subjects that they would be hearing a radio ad, that they should not look at the questions until the ad has stopped playing, and that they should begin answering the questions when the ad is over.
The dependent variable was perceived duration of the ad, measured retrospectively. Instructions preceding the time estimation measure read as follows: "Radio ads typically last anywhere from 10 seconds to over 45 seconds. Since many ads are read 'live' on the air, they often last for some odd duration (as opposed to TV ads which normally last for exactly 30 seconds). How long did the ad that you just heard seem to last? Please be as precise as possible, even if you are not certain." The estimate item was "I estimate that the ad lasted for ______ seconds." This type of open-ended item is commonly used in time perception research in psychology (e.g., Block 1990; Fraisse 1984; Levin and Zakay 1989) and, recently, in consumer research (Kellaris and Altsech 1992; Kellaris and Kent 1992).
The questionnaire also contained demographic items (i.e., sex, age), and a multi-item seven-point semantic differential scale to facilitate post-test manipulation and confounding checks. The mood scale was preceded by the prompt "The music made me feel." The scale items were "good/bad," "pleased/ displeased," "happy/sad," and "positive/negative." A composite scale was formed by summing and averaging the items (Cronbach's alpha=.86).
Variance analysis (ANOVA) examined the impact of the treatment variables on the manipulation check measure. We found a significant main effect of the mood treatment on the mood manipulation check scale (F(1,109)=7.4, p<.01), with no main or interactive effect involving gender. This provides evidence of a successful manipulation that is unconfounded with gender (Perdue and Summers 1986).
Effects on Perceived Time
Time estimates ranged from 5 to 60 seconds, with an average estimate of about 27 seconds (median=25 sec.) across all conditions. This is consistent with previous research that found duration estimates to vary widely about a slightly underestimated central tendency (e.g., Fraisse 1984). ANOVA results are summarized in Table 1.
INTERACTIVE EFFECT OF MOOD AND GENDER ON DURATION ESTIMATES
Variance analysis found a significant main effect of gender on perceived duration (F(1,109)=4.29, p<.05). The group mean duration estimate is 28.4 seconds for males, and 24.1 seconds for females. The magnitude of the effect was estimated as w2=.03 (see Keppel 1982, p. 92 for a discussion of effect size interpretation).
The two-way interaction of mood and gender was also statistically significant (F(1,109)=9.16, p<.003), with an effect size of w2=.07. This interaction is illustrated in Figure 1.
The duration estimates of males and females differed statistically under the less positive mood condition (t=3.54, p<.001), but not under the more positive mood condition (t=-.35, n.s.). A significant difference was observed between mood groups among females (t=-2.94, p<.006), but not among males (t=1.02, n.s.).
Comparisons with Clock Time
Comparing subjects' duration estimates to actual clock time, we found a general tendency to underestimate retrospectively the passage of time (t=-3.55, p<.001). This tendency was observed across mood conditions for both male (t=-2.95, p<.01) and female (t=-4.43, p<.001) subjects, and under both less (t=-3.55, p<.001) and more (t=-3.48, p<.001) positive mood states across genders.
However, comparing mean duration estimates separately for each cell of our 2 X 2 design with clock time, we found statistical differences between perceived and real time in only two cells: among females in the less positive mood group (t=-6.76, p<.001), and among males in the more positive mood group (t=-3.29, p<.01). The duration estimates of females in more positive moods and males in less positive moods did not differ statistically from clock time.
Our experimental findings suggest that both gender and its interaction with mood can influence consumers' time perceptions. The data supported our expectation that males would produce more accurate time estimates, whereas females tended to perceive shorter, less accurate durations. Further, while both genders tended to underestimate the duration of a stimulus event (test ad) in relation to clock time, the pattern of time estimates varied systematically across mood conditions, although mood had no apparent influence on time perceptions independently of gender.
Females' time estimates were significantly different for the less versus more positive mood conditions while the males' responses were relatively invariant across mood conditions. Perhaps, as suggested by Rothkopf and Blaney (1991), the female subjects were more affected by their mood state than their male counterparts due to greater self-awareness. This greater self-awareness would not necessarily be detected in our post-test manipulation check of mood state, which should have induced self-awareness post hoc in both males and females by asking them to recall their feelings at the time of exposure. We must commend testing of the self-awareness explanation to future research.
It is interesting to note that although the males' responses were statistically invariant across mood conditions, the direction of mean duration estimates was negative (i.e., males in good moods estimated the event duration to be about two seconds shorter). Based on the "mood preservation hypothesis" (people in good moods avoid activities that would result in loss of good mood), we speculate that the males in less (versus more) positive mood states may have been slightly more motivated (or less demotivated) to process the information they encountered (Isen 1989).
For the women, the shorter time estimates associated with the less positive mood condition suggests that time does not necessarily fly when they are having fun. This result, although inconsistent with popular wisdom, is consistent with prior mood/music research in consumer behavior (Kellaris and Kent 1992). This finding is also consistent with at least two process explanations. Women may be more likely to store a larger quantity of information about positively valenced stimulus events (Rothkopf and Blaney 1991). Alternatively, women may have stronger links between mood and cognitive nodes in memory and thus can retrieve more stored information when they are in positive moods (Clark and Teasdale 1985). No matter the underlying process, our data suggest that mood has a greater influence on the temporal experience of female (versus male) subjects.
Although these results should be considered preliminary, with reliability subject to future replication, we believe that they provide an interesting, initial demonstration of principles that eventually may be used in practical contexts.
There are many circumstances under which marketers may wish to diminish consumers' perceptions of time passage. For example, consumers may grow impatient and annoyed during long waits in check-out lines or other queues, while waiting for tables at a busy restaurant, while placed on "hold" when phoning a business, or while the viewing of a program is interupted with a commercial. Dissatisfaction may be mitigated by "shrinking" the perceived duration of the delay. Our study provides a preliminary glimpse into how such diminution of time can be achieved by manipulating moods with music.
Specifically, our findings suggest that music designed to elevate customers' moods may be counter-productive among females. Elevating moods may actually increase the accuracy of their time perceptions, such that they experience the full duration of a delay. Music that provides a distraction without inducing an elevated mood should be used when time diminution is a goal.
There are also circumstances under which marketers may wish to augment the perceived passage of time. For example, once restaurant patrons are seated at a table, "expanding" time perceptions may lead to quicker table turn-over without making customers feel rushed. Although we did not observe systematic over-estimation of time under the conditions in this study, our findings suggest the general possibility that music can be used to alter temporal perception. Future research should investigate conditions that lead to time augmentation.
Limitations & Future Research
The findings of this study should be interpreted cautiously given the preliminary nature of the study. Although the results are consistent with previous research, the use of student subjects and a forced exposure condition may have influenced responses in ways that are difficult to evaluate. Additional research is needed to assess the extent to which these lab findings generalize to other groups of consumers and natural settings.
Future exploration of time perception should be of ample interest to consumer researchers and marketing practitioners alike. Of particular interest are the process mechanisms underlying temporal perception. Because our findings are consistent with several explanations, we must commend the investigation of underlying psychological processes to future research.
In sum, this study has examined the influence of music-induced mood and gender on time perception. We found potentially consequential gender differences, and developed tentative commercial applications. We believe that a better understanding of factors that influence consumers' time perceptions eventually may allow marketers and service providers to mitigate negative consequences of time passage (e.g., delays in service) by engineering environmental stimuli such as music to diminish the perception of time passage.
Alpert, Judy I. and Mark I. Alpert (1989), "Background Music as an Influence in Consumer Mood and Advertising Responses," Advances in Consumer Research, 16, 485-491.
Bergadaa, Michelle M. (1990), "The Role of Time in the Action of the Consumer," Journal of Consumer Research, 17, 289-302.
Block, Richard A. (Ed.) (1990), Cognitive Models of Psychological Time, Hillsdale, NJ: Erlbaum.
Block, Richard A. (1989), "Experiencing and Remembering Time: Affordances, Context, and Cognition," in I. Levin and D. Zakay (Eds.), Time and Human Cognition: A life Span Prospective, Amsterdam: North-Holland, 333-363.
Cook, Thomas D. and Donald T. Campbell (1979), Quasi-Experimentation: Design and Analysis Issues for Field Settings, Chicago, IL: Rand McNally.
Cox, Dena S., and James J. Kellaris (1988), "The Effect of Mood and Gender on Perceived Likelihood of Making Unplanned Purchases," in L. Alwitt, ed., Proceedings of the Division of Consumer Psychology, Washington, D.C., American Psychological Associaiton, 14-16.
Clark, David M. and John D. Teasdale (1985), "Constraints on the Effects of Mood on Memory," Journal of Personality and Social Psychology, 48:6, 1595-1608.
Fraisse, Paul (1963), The Psychology of Time, Greenwood Press Publishers, Westport, Conneticut.
Fraisse, Paul (1984), "Perception and Estimation of Time," Annual Review of Psychology, 35, 1-36.
Gardner, Meryl Paula (1985), "Mood States and Consumer Behavior: A Critical Review," Journal of Consumer Research, 12(December), 281-300.
Ger, Guliz (1989), "Nature of Effects of Affect on Judgment: Theoretical and Methodological Issues," in Cognitive and Affective Responses to Advertising, ed. Patricia Cafferata and Alice M. Tybout, Lexington, Massachusetts: D.C. Heath Company, 263-275.
Hornik, Jacob (1984), "Subjective vs. Objective Time Measures: A Note on the Perception of Time in Consumer Behavior," Journal of Consumer Research, 11, 615-618.
Isen, Alice M. (1989), "Some Ways in Which Affect Influences Cognitive Processes: Implications for Advertising and Consumer Behavior," in Cognitive and Affective Responses to Advertising, ed. Patricia Cafferata and Alice M. Tybout, Lexington Massachusetts: D.C. Heath and Company, 91-117.
Kellaris, James J., and Moses B. Altsech (1992), "The Experience of Time as a Function of Musical Loudness and Gender of Listener," in J. Sherry and B. Sternthal, eds., Advances in Consumer Research, Vol. 19, Provo, UT: Association for Consumer Research, 725-729.
Kellaris, James J., and Robert J. Kent (1992), "The Influence of Music on Consumers' Temporal Perceptions: Does Time Fly When You're Having Fun?" Journal of Consumer Psychology, 1, 4, 365-376.
Keppel, Geoffrey (1982), Design & Analysis C A Researcher's Handbook, 2nd ed., Englewood Cliffs, NJ: Prentice-Hall.
Krishnan, Lila and Neelaksh Kumar Saxena (1984), "Perceived Time: Its Rela-tionship with Locus of Control, Filled Versus Unfilled Time Intervals, and Perceiver's Sex," The Journal of General Psychology, 110, 275-281.
Leventhal, Howard (1982), "The Integration of Emotion and Cognition: A View from the Perceptual-Motor Theory of Emotion," in Affect and Cognition: The Seventeenth Annual Carnegie Symposium on Cognition, eds. Margaret Sydnor Clark and Susan T. Fisk, Hillsdale, NJ: Lawrence Eribaum, 121-156.
Levin, Iris and Daniel Zakay (Eds.) (1989), Time and Human Cognition, Advances in Psychology series # 59, Elsevier Science Publishers, North-Holland.
Maccoby, Eleanor Emmons and Carol Nagy Jacklin (1974), The Psychology of Sex Differences, Stanford, CA: Stanford University Press.
Matlin, Margaret W. (1989), Cognition, Chicago: Holt, Rinehart, and Winston, Inc.
Ornstein, Robert E. (1969), On the Experience of Time, New York, NY: Penguin Books.
Perdue, Barbara C., and O. John Summers (1986), "Checking the Success of Manipulations in Marketing Experiments," Journal of Marketing Research, 23, 317-326.
Rammsayer, Thomas and Sebastian Lustnauer (1989), "Sex Differences in Time Perception," Perceptual and Motor Skills, 68, 195-198.
Rothkopf, Jeffrey S. and Paul H. Blaney (1991), "Mood Congruent Memory: The Role of Affective Focus and Gender," Cognition and Emotion, 5:1, 53-64.
Zakay, Daniel (1989), "Subjective Time and Attentional Resource Allocation : An Integrated Model of Time Estimation," in I. Levin and D. Zakay (Eds.), Time and Human Cognition: A Life Span Prospective, Amsterdam: North-Holland, 365-398.