Effect of Ad Pacing and Optimal Level of Arousal on Attitude Toward the Ad

ABSTRACT - It was predicted that the pacing of TV advertisements would influence perceived arousal levels, and, depending on viewer's optimal level of arousal, attitudes toward the ad. Ad pacing did indeed influence arousal levels, but only for those with low optimal levels of arousal. Post hoc correlational evidence suggests that low optimals, but not high optimals, like relatively fast ads because they are emotionally arousing. Results are discussed in light of the multiply-determined nature of attitude toward the ad, and how ad pacing, through its effect of perceived levels of arousal, may increase our understanding of what makes advertisements effective.

Citation:

Mark A. Pavelchak, Meryl P. Gardner, and V. Carter Broach (1991) ,"Effect of Ad Pacing and Optimal Level of Arousal on Attitude Toward the Ad", in NA - Advances in Consumer Research Volume 18, eds. Rebecca H. Holman and Michael R. Solomon, Provo, UT : Association for Consumer Research, Pages: 94-99.

Advances in Consumer Research Volume 18, 1991      Pages 94-99

EFFECT OF AD PACING AND OPTIMAL LEVEL OF AROUSAL ON ATTITUDE TOWARD THE AD

Mark A. Pavelchak, University of Delaware

Meryl P. Gardner, University of Delaware

V. Carter Broach, University of Delaware

ABSTRACT -

It was predicted that the pacing of TV advertisements would influence perceived arousal levels, and, depending on viewer's optimal level of arousal, attitudes toward the ad. Ad pacing did indeed influence arousal levels, but only for those with low optimal levels of arousal. Post hoc correlational evidence suggests that low optimals, but not high optimals, like relatively fast ads because they are emotionally arousing. Results are discussed in light of the multiply-determined nature of attitude toward the ad, and how ad pacing, through its effect of perceived levels of arousal, may increase our understanding of what makes advertisements effective.

INTRODUCTION

The concept of optimal level of arousal suggests that every organism has a preferred level of stimulation (e.g., Hebb 1955; Leuba 1955). Departures from this optimal level are aversive, which motivates (arousal seeking or avoiding) behavior intended to restore the optimum level. The concept has a long history in the study of human behavior (e.g., Berlyne 1960; Breuer and Freud 1895/1937; Driver and Streufert 1964). In consumer research, the concept has been applied to exploratory behavior (Raju 1980; Faison 1977; Venkatesan 1973), new product purchasing (Hirschman 1980; Robertson 1971), product/service satisfaction (Hanna and Wagle 1988), and reactions to television programs (Zillman and Bryant 1985). We feel that the concept has potential for enhancing our understanding of advertising effectiveness.

Two measures typically associated with advertising effectiveness are attitude toward the brand (Abrand) and purchase intent. Because attitude toward the advertisement (Aad) mediates Abrand and purchase intent (e.g., Mitchell and Olson 1981; Shimp 1981), understanding the underlying determinants of Aad is important (MacKenzie and Lutz 1989). Therefore, this study focuses on the possible effect of optimal level of arousal on Aad.

Over the years, arousal has been measured in two primary ways: physiological indices and self-report. Each approach has its advocates and detractors, and a legitimate place in research on advertising effectiveness. As our concern is with conscious feelings of arousal and how they are influenced by advertising, we measure arousal via self report in the current study. Consistent with previous researchers, we conceptualize arousal as a basic dimension of subjective emotional experience ranging from sleep to &antic excitement (Berlyne 1967; Mehrabian and Russell 1974). The dimension of arousal is conceptually orthogonal to pleasure/displeasure, the other primary dimension of emotional experience (Mehrabian and Russell 1974). This means that feelings of arousal per se are not inherently pleasurable or displeasurable.

It is well established that TV commercials can influence perceived levels of arousal in viewers (e.g., Pavelchak 1989, Wells 1964). However, because arousal is not inherently pleasurable, the effect of ad-induces arousal on Aad probably depends on secondary factors, such as viewer's optimal level of arousal. In general, individuals who prefer high levels of stimulation should react positively to an arousal-inducing ad, while those who prefer lower levels of stimulation should react positively to an arousal-reducing ad.

As with any stimulus, a variety of aspects of TV commercials have the potential to influence levels of arousal (e.g., humor, annoying appeals, sexually suggestive images). Many such factors, however, are only relevant to a small subset of commercials. In the present study, we selected an aspect that is relevant to all commercials: its activity level or pacing. Along with evaluation and potency, activity level is one of the three fundamental dimensions of stimulus meaning (Osgood, Suci, and Tannenbaum 1957). In addition, variations in ad activity level have been shown to influence perceived levels of arousal, while variations in evaluation level do not (Pavelchak 1989). Specifically, relatively "fast" ads tend to increase levels of perceived arousal, while relatively "slow" ads tend to have the opposite effect. Relating this to the construct of optimal level of arousal leads to the following hypotheses:

H1A: Individuals with relatively high optimal levels of arousal (high optimals) should like "fast" ads (given that other ad features are controlled for) more than those with relatively low optimal levels of arousal (low optimals).

H1B: Low optimals should like "slow" ads (given that other ad features are controlled for) more than high optimals.

Jointly, confirmation of these hypotheses should result in a significant interaction between self-reported optimal level (low/high) and type of ad seen (slow/fast) on Aad.

METHODOLOGY

Overview

Data for the present study were collected through a laboratory experiment with one manipulated factor, pace of the commercial (slow/fast). Upon arrival, subjects' optimal level of arousal was measured. Afterward, subjects indicated their emotional state (including level of arousal), were shown either a fast or slow-paced TV commercial, indicated their emotional state again, and then responded to measures of advertising effectiveness (including Aad). Based on their response to the optimal level of arousal measures, subjects were divided via median split into low optimals and high optimal. Therefore, the overall design of the study is 2 (fast ad/slow ad) X 2 (low optimal/high optimal), and the primary dependent measure is Aad.

Subjects

Subjects were students enrolled in an undergraduate introduction to marketing course at a major Eastern university. Subjects participated as part of a course research requirement. Of the 114 students who participated, eight were eliminated because they reported prior exposure to the ad used in their experimental session. This left a final sample of 106 subjects, 35 males and 71 females.

Advertisement Selection Procedure

The television ads used in this experiment were professionally developed by advertising agencies. Such ads generate a more "natural" emotional response than mock ads (Mitchell 1986). To minimize familiarity, the ads were videotaped from television programming in cities other than the one in which this study was conducted. Over 100 such commercials were rated by 20-30 judges on semantic differential items known to load highly on the dimensions of evaluation, activity, and potency. Each judge rated 10-20 ads (most judges rated 20). The judges, who came from the same population as those used in the main experiment, also indicated whether they had seen the ad before and whether they had purchased the product/service depicted in the ad. Based on judges' activity ratings, four ads were selected, two rated as relatively fast (ads for a bank and a coat store), and two rated as relatively slow (ads for a carpet store and a drug store). As expected, activity ratings of the fast ads were significantly higher than ratings of the slow ads (; < .05)

Measures

Subjective emotional states, including feelings of arousal, were measured using the Affect Grid developed by Russell, Weiss and Mendolsohn (1989). It consists of a 9 X 9 matrix of squares with two dimensions labeled as displeasure/pleasure and arousal/sleepiness. Subjects place a mark at the point on the grid that best represents their feelings at the moment. Optimal level of arousal was determined by responses to two separate scales that have some precedent in consumer research: Mehrabian and Russell's (1973) Arousal-Seeking scale, and Zuckerman, Eysenck and Eysenck's (1978) Sensation-Seeking scale. Each of these forty item scales measures the degree to which individuals tend to seek or avoid arousal-inducing situations. One primary difference between the two scales is that all of the items in the Zuckerman scale are positively worded (endorsement suggests high optimal level) whereas the Mehrabian scale contains both positively and negatively-worded items. Reliability analyses were performed on each scale and the coefficient alpha values were: Mehrabian = .86; Zuckerman = .78. These levels were deemed rather low for scales with so many items, but acceptable.

Attitude toward the ad was measured using a four-item scale developed by Mitchell (1986). Reliability for this scale was considered acceptable, as the coefficient alpha was .88.

Procedure

Prior to each session, one of the four videotaped commercials was randomly selected to be shown. Subjects were run in groups of five to ten. Upon arrival, subjects were told that the study concerned "personal characteristics and reactions to commercials." Subjects then filled out the Zuckerman scale, then the Mehrabian and Russell scale. Next, the affect grid was explained and subjects indicated on one grid their feelings at that point in time. Afterwards, the videotaped commercial was shown, and subjects filled out a second affect grid to represent how the commercial made them feel. Subjects then filled out a questionnaire that included measures of Aad, Abrand, whether or not they had seen the ad before, whether or not they had purchased the product depicted in the ad before, and other items. Finally, subjects were debriefed and thanked for their participation.

RESULTS

With the lack of control inherent in a study with self-reported factors, it important to examine the data for possible-problems stemming from self-selection. Therefore, t-tests were performed on: pre-exposure arousal scores, pre-exposure pleasure scores, and Arousal and Sensation Seeking scores, with type of ad (slow/fast) as the between-subjects factor. Fortunately, none of the analyses were even close to significant (all g's > .38). Although no ad pacing manipulation check measure was included in the post-experimental questionnaire, we sought to verify that the manipulation had the intended psychological impact by examining subjects' post-exposure arousal scores. A two-way ANOVA was performed on these scores with type of ad (slow/fast) and optimal level (low/high) as between-subjects factors. As expected, postexposure arousal scores were significantly higher in the fast ad condition (ka = 5.31) than in the slow ad condition (ki = 4.59, F(1,102) = 7.01, p = .009). No other effects in this analysis were significant. Similar analyses on Aad and Abrand revealed no significant differences, indicating that there was no overall preference for the fast ads relative to the slow ads, or for the advertised brands.

TABLE 1

Aad SCORES AS A FUNCTION OF AD PACING AND OPTIMAL LEVEL OF AROUSAL

Turning now to the primary analyses, the first analysis utilized Mehrabian scale scores as a measure of optimal level. An ANOVA was performed on Aad scores with type of ad (slow/fast) and optimal level (low/high) as between-subjects factors. It was predicted that high optimals would like fast ads more than low optimals, and low optimals would like slow ads more than high optimals. The predicted interaction was not significant (E2 = .29), and surprisingly, the pattern of means was opposite to the predicted pattern (see Table 1A). The picture was even more dramatic when the Zuckerman scale was used as a measure of optimal level, as the Optimal Level X Condition interaction was statistically significant (! i(l,102) = 4.85, g = .03) (see Table 1B).

As a first step in understanding these results, the self-reported arousal data were re-examined. This time, arousal change scores, which of course take pre-exposure scores into account, were examined. It was discovered that, while the ad pacing manipulation was effective overall, the effect was due exclusively to low optimal level subjects who were shown a fast ad (See Table 2). The contrast of their arousal change scores versus the remaining subjects was statistically significant (i(104) = 2.12, p. < .03).

Given that relatively fast-paced ads influenced perceived levels of arousal in low optimals but not in high optimals, it remains to be seen whether there was a subsequent effect of arousal on Aad (in the fast ad condition). As evidence that there was such an effect, the correlation between Aad and postexposure arousal scores was +.49 f; < .001), and between Aad and arousal change scores, the correlation was +.31 fp = .01). On the surface, however, the evidence that these relationships are due to arousal seeking is weak: the correlation between Aad and arousal seeking is -.23 fp < .05) using the Zuckerman scale, and only -.05 (n.s.) using the Mehrabian scale.

A closer look at the data, however, is more encouraging. Responses to the Zuckerman scale were decomposed into its four sub-scales (Thrill Seeking, Experience Seeking, Disinhibition, and Boredom Avoidance). Correlations regarding the Experience Seeking scale, which basically is a measure of sensuality orientation, were intriguing: Aad and Experience seeking correlated -.43 (g = .001), while arousal change scores and Experience seeking correlated -.44 (I2 < .001). (Similar, although weaker results were obtained when an experience seeking scale derived from the Mehrabian scale was utilized).

Ad-induced arousal may have had an effect on Aad in the fast ad condition, but not in the slow ad condition. Neither high nor low optimals' perceived levels of arousal were influenced by the slow ads (both mean arousal change scores were close to zero; see Table 2). Nevertheless, high and low optimals differed in their liking for these ads, with high Optimals (based on the Zuckerman scale) being more favorable (t(49) = 2.06, p = .044).

TABLE 2

AROUSAL CHANGE SCORES AS A FUNCTION OF AD PACING AND OPTIMAL LEVEL OF AROUSAL

DISCUSSION

Ad Pacing and Arousal

One implicit objective of this study was to provide additional empirical evidence that ad pacing can influence perceived levels of arousal. To achieve this objective, subjects were shown ads that varied in activity level. We realize that the ads undoubtedly varied along many other dimensions, and that such differences may be responsible for our results. We tried to partially address this issue by using two versions of each type of ad. No doubt our results would have been more clear-cut if additional factors (such as the type of product) were held constant across the ads. In reality, however, unless time compression methodology is used, it may be impossible to vary only ad pacing. Our results should be interpreted in this light.

While a significant effect of ad pacing on post-exposure arousal scores was observed, a closer look at the data revealed that only the arousal levels of low optimals who saw a fast ad were influenced (increased). In contrast, the arousal levels of high optimals were unaffected by fast ads. One possible explanation for this result is that the fast ads were not fast enough. In our pretesting procedure, we encountered ads that were rated as having much higher activity levels than the ones used in the present study. Unfortunately, those ads could not be used because of an attempt to equate type of product/service across the four ads. If significantly faster ads were used, the arousal levels of high optimals may have been increased, although perhaps not as much as those of low optimals.

In the slow ad condition, arousal levels of both low and high optimals were virtually unaffected. It may be that 30-second television ads can be used to increase, but not decrease, levels of perceived arousal, especially when subjects have not seen the ad before. An ad's slow pacing may be counteracted by its novelty, which tends to increase i levels of attention and perhaps arousal (Berlyne f 1967). In addition, emotional states with low levels of arousal (boredom, sleepiness), may simply require a stimulus duration of longer than 30 seconds to be activated.

This study has demonstrated the usefulness of the Russell et al (1989) Affect Grid in studies of advertising and emotion. Subjects reported little difficulty using the grid, and its simplicity permits ; the relatively unobtrusive measurement of emotional states at multiple points during an experimental session. One flaw in our use of the scale was our failure to ask subjects to indicate their optimal level ; on the grid. Had this been done, a closer correspondence between arousal seeking and momentary perceived levels of arousal could have been made

Arousal and Attitude Toward the Ad

The pattern of correlations in the fast ad condition among arousal change scores, Aad, and arousal (experience) seeking scores suggests that arousal change might have mediated the effect of experience seeking on Aad. The direction of the relationships suggest that low experience seekers appear to like relatively fast ads more than high experience seekers, and do so because they are emotionally arousing. High experience seekers, on the other hand, did not find the relatively fast ads emotionally arousing, and were less favorable toward them. Thus, it seems as though ad-induced arousal can in fact influence attitude toward the ad, although in a manner contrary to our initial expectations. These results are admittedly post hoc, but do suggest that Aad may be determined in part by ad-induced feelings of arousal sometimes, and that such an effect may depend on the arousal seeking or avoiding nature of the viewer.

In the slow ad condition, in contrast, the arousal levels of neither high nor low optimals were influenced by the ads, and yet the two groups differed in terms of Aad (high optimals were more favorable). This result reminds us that Aad is multiply determined; that ad-induced arousal is only one of many factors that contribute to such judgments.

It is interesting to speculate about the effect of ad pacing on Aad if "really fast" ads were used. We assume that such ads would increase the arousal levels of both low and high optimals, not just low optimals. Yet would extremely fast pacing be "too fast" for the so-called arousal-avoiding low optimals, and would their relatively positive reaction toward fast ads reverse? Would extremely fast pacing be to the liking of high optimals, as we originally hypothesized? If so, the pattern of means for Aad would reverse, and our hypothesis would receive support.

Ad Pacing, Arousal, and Attitude Toward the Ad

This study has shown that ad pacing may be a promising factor to manipulate in attempts to enhance advertising effectiveness. It also shows that ad pacing, or at least the small difference in ad pacing used in this study, does not influence perceived levels of arousal in all viewers. It remains to be seen what features of ads will increase levels of arousal in high optimals. It may be extremely fast pacing, or a completely different factor. Given the present results, practitioners may find it useful to develop a taxonomy of ad features that are arousing to specific segments, and manipulate those features in ads targeted toward those segments.

Directions for Future Research

Additional work is needed to understand the effects of an ad's pace or activity level, and the conditions under which pacing affects arousal. Further research is also needed to explore the relationship between arousal and attitude toward the ad. Findings indicate that attitude toward the ad is multiply determined and that arousal needs to be considered in concert with other factors. In addition, research should address the conditions under which ad pacing affects attitude toward the ad.

REFERENCES

Berlyne, Donald E. (1960), Conflict, Arousal, and Curiosity, New York: McGraw-Hill Book Co.

Berlyne, Donald E. (1967), "Arousal and Reinforcement," in Nebraska Symposium on Motivation, ed., D. Levine, Lincoln: University of Nebraska Press.

Breuer, J., and Sigmund Freud (1937), Studies in Hysteria (A.A. Brill, trans.), New York: Nervous and Mental Disease Publishing Company (Originally published 1895).

Driver, Michael J., and Siegfried Streufert (1965), "The 'General Incongruity Adaptation Level' (GIAL) Hypothesis: An Analysis and Integration of Cognitive Approaches to Motivation," Working Paper No. 114, Institute for Research in the Behavioral, Economic, and Management Sciences, Purdue University, Lafayette, IN.

Faison, Edmund W. (1977), "The Neglected Variety Drive: A Useful Concept for Consumer Behavior, Journal of Consumer Research, 4 (December), 172-175.

Hanna, Nessim and John S. Wagle (1988), "Who is Your Satisfied Customer?" Journal of Services Marketing, 2 (Summer), 5-13.

Hebb, D. O. (1955), "Drives and the CNS (Conceptual Nervous System)." Psychological Review, 62, 243-254.

Hirschman, Elizabeth C. (1980), "Innovativeness, Novelty Seeking, and Consumer Creativity," Journal of Consumer Research, 7 (December), 283-295.

Leuba, Clarence (1955),."Toward Some Integration of Learning Theories: The Concept of Optimal Stimulation," Psychological Reports, 1, 27-33.

MacKenzie, Scott B. and Richard J. Lutz (1989), "An Empirical Examination of the Structural Antecedents of Attitude Toward the Ad in an Advertising Pretesting Context," Journal of Marketing, 53 (April), 48-65.

Mehrabian, Albert and James A. Russell (1973), "A Measure of Arousal Seeking Tendency," Environment and Behavior,5, 315-333.

Mehrabian, Albert (1974), An Approach to Environmental Psychology, Cambridge: M.I.T. Press.

Mitchell, Andrew A. (1986), "Theoretical and Methodological Issues in Developing and Individual Level Model of Advertising Effects," in Advertising and. Consumer Psychology, eds. Jerry C. Olson and Keith Sentis, New York: Praeger Publishing Co., 172-196.

Mitchell, Andrew A. and Jerry C. Olson (1981), "Are Product Attribute Beliefs the Only Mediator of Advertising Effects on Brand Attitude?" Journal of Marketing Research, 18, 318-332.

Osgood, Charles E., G. J. Suci, and Percy H. Tannenbaum (1957), The Measurement of Meaning, Urbana: University of Illinois Press.

Pavelchak, Mark A. (1989), "Dimensions of Stimulus Meaning: A Basis for Predicting Emotional Reactions to Commercials," paper presented at the meeting of the Society for Consumer Research (APA), New Orleans, LA.

Raju, P. S. (1980), "Optimum Stimulation Level: Its Relationship to Personality, Demographics, and Exploratory Behavior," Journal of Consumer Research, 7 (December), 272-282.

Robertson, Thomas S. (1971), "Innovative Behavior and Communication, New York: Holt, Rinehart and Winston.

Russell, James A., Anna Weiss, and Gerald A. Mendolsohn (1989), "Affect Grid: A Single Item Scale of Pleasure and Arousal," Journal of Personality and Social Psychology, 57 (March), 493-502.

Shimp, Terence A. (1981), "Attitude Toward the Ad as a Mediator of Consumer Brand Choice," Journal of Advertising, 10 (2), 9-15.

Venkatesan M. (1973), "Cognitive Consistency and Novelty Seeking," in Consumer Behavior: Theoretical Sources, eds., S. Ward and T.S. Robertson, Englewood Cliffs, N.J. pp. 354-384.

Wells, William D. (1964), "EQ, Son of EQ, and the Reaction Profile," Journal of Marketing, 28 (October), 45-52.

Zillman, Dolf and Jennings Bryant (1985), "Affect, Mood, and Emotions as Determinants of Selective Exposure," in Selective Exposure to Communications, eds., Dolf Zillman and Jennings Bryant, Hillsdale, NJ: Erlbaum, 157190.

Zuckerman, Miron, S. Eysenck, and M. Eysenck (1978), "Sensation Seeking in England and America: Cross-Cultural, Age and Sex Comparisons," Journal of Consulting and Clinical Psychology, 46, 139-149.

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