Effects of Involvement, Arousal, and Pleasure on the Recognition of Sponsorship Stimuli

ABSTRACT - This paper examines the effects of involvement with a sports event, and arousal and pleasure in reaction to this event, on the recognition of embedded sponsorship stimuli. A study shows that involvement with a soccer game has a curvilinear effect on the recognition of embedded billboards. Arousal in reaction to the game has a negative effect on this recognition. These effects are distinct from each other. Pleasure did not have the expected positive effect. It was also observed that the recognition of the sponsorship stimuli was significantly lower among females than among males.

Citation:

M. Tuan Pham (1992) ,"Effects of Involvement, Arousal, and Pleasure on the Recognition of Sponsorship Stimuli", in NA - Advances in Consumer Research Volume 19, eds. John F. Sherry, Jr. and Brian Sternthal, Provo, UT : Association for Consumer Research, Pages: 85-93.

Advances in Consumer Research Volume 19, 1992      Pages 85-93

EFFECTS OF INVOLVEMENT, AROUSAL, AND PLEASURE ON THE RECOGNITION OF SPONSORSHIP STIMULI

M. Tuan Pham, University of Florida

ABSTRACT -

This paper examines the effects of involvement with a sports event, and arousal and pleasure in reaction to this event, on the recognition of embedded sponsorship stimuli. A study shows that involvement with a soccer game has a curvilinear effect on the recognition of embedded billboards. Arousal in reaction to the game has a negative effect on this recognition. These effects are distinct from each other. Pleasure did not have the expected positive effect. It was also observed that the recognition of the sponsorship stimuli was significantly lower among females than among males.

INTRODUCTION

Sponsorship can be defined as "the provision of resources (e.g., money, people, equipment) by an organization directly to an event or activity in exchange for a direct association to the event or activity" (Sandler and Shani 1989, p.10). Sports Sponsorship as a communication tool has recently become very popular among marketers. According to Meenaghan, (1989), in 1987 in the United States, 35,000 companies spent $1.75 billion to sponsor sporting events, and in 1989 worldwide sponsorship expenditures were valued at $4.1 billion. Marketing practitioners' marked interest in sponsorship communication (e.g., Otker 1988; Ryssel and Stamminger 1988) is in sharp contrast with academic researchers' lack of attention (Gardner and Shuman 1986; Meenaghan 1983; Sandler and Shani 1989).

One of the main reasons for the practice of sponsorship is the achievement of media coverage (Abratt, Clayton, and Pitt 1987). Hence, sponsorship effectiveness is most often expressed in terms of the amount of media coverage generated through it (Waite, 1979; Mintel 1986). However, while sponsorship stimuli (e.g., billboards) embedded in sports events can receive dramatic exposure (see e.g., Welling 1986), there is evidence that attention to and memory for these stimuli are, in fact, very limited (Nebenzahl and Hornik 1985; d'Ydewalle et al. 1987). Therefore, there is a need to better understand the factors that affect the processing of sponsorship stimuli and, consequently, moderate the effectiveness of their exposure.

The purpose of this paper is to explore how involvement with a sponsored event, and affective reactions to this event, moderate the effectiveness of exposure to embedded sponsorship stimuli. The measure of effectiveness considered here is the subsequent recognition of these stimuli. The effects of event-induced affective reactions are studied along two main dimensions, pleasure and arousal (cf. Mehrabian and Russell 1974). The literature about the effects of involvement, arousal, and pleasure is briefly reviewed, leading to three hypotheses. The results of a study testing these hypotheses are then reported and discussed.

SPONSORSHIP STIMULI IN CONTEXT

The selection of events to sponsor is partly determined by the assumption that sponsorship stimuli will be processed somewhat differently as a function of the events in which they appear. This issue parallels the stream of research that relates the context of an ad to its performance (e.g., Axelrod 1963, Thorson et al. 1985). For instance, Pavelchak, Antil, and Munch (1988) have recently investigated the effects of emotional experience induced by a Super Bowl game on the recall of embedded commercials. They found that arousal induced by the game decreases the number of commercials recalled. There is, however, a critical difference between commercials and the type of stimuli considered in this study: sponsorship stimuli (e.g., billboards around a soccer field, brand names on racing cars, logos on team players' shirts) do not interrupt the programs in which they are embedded. Therefore, the issue is to determine how factors related to the processing of a sports event affect the concurrent processing of sponsorship stimuli. Three variables are considered here: 1) involvement toward the sponsored event, 2) arousal in response to this event, and 3) pleasure when watching the event. Although involvement and arousal are strongly correlated, especially in a sports context, it is suggested here that they have distinct effects on the processing and subsequent recognition of sponsorship stimuli.

Effect of Involvement

Consumers' processing of a sponsored event is influenced by their motivation, ability, and opportunity to process this event (cf. Batra and Ray 1986; MacInnis and Jaworski 1989). These factors should also affect the concurrent processing of any embedded stimuli. Motivation to process information is often referred to as involvement with the informational object (e.g., Celsi an Olson 1988; Petty, Cacioppo, and Schumann 1983). This motivational state is governed by the perceived relevance of the target object (Greenwald and Leavitt 1984; Zaichkowsky 1985). Consistent with Celsi and Olson (1988) and Houston and Rothschild (1978), it is proposed that the phenomenological experience of involvement with a particular sponsored event results from the combination of enduring (or intrinsic) and situational sources of personal relevance. For instance, while an individual may be involved with soccer in general or with a specific team in particular (i.e., enduring involvement), situational factors such as the opponent (e.g., two teams from the same city), the stakes (e.g., a World Cup final), or the score (e.g, 3-3) can increase or decrease the perceived personal relevance of a given game. This perceived personal relevance, or "felt involvement" (Celsi and Olson 1988), will influence the processing of the event, and thereby, the processing of the sponsorship stimuli.

Consistent with Mitchell (1979;1981), (felt) involvement with a sponsored event is conceptualized as having both intensity and directional properties, each exerting a different effect on the processing of embedded stimuli. The more a person is involved with say, a soccer game, the more intense his/her processing of the game (cf., Burnkrant and Sawyer 1983). At low levels of involvement, little attentional capacity (effort) will be allocated to the event as a whole. Because they are inlaid in the event, sponsorship stimuli should also receive little attention. As involvement increases more overall attention is devoted to the event and also, as a result of their embeddedness, to the sponsorship stimuli.

However, involvement also has a directional property. Because of limited cognitive capacity (Kahneman 1973), as involvement or motivation to process reaches high levels, attention becomes more focused on the relevant source of information (Celsi and Olson 1988; MacInnis and Jaworski 1989), i.e. the game itself, and away from irrelevant information such as the sponsorship stimuli. Thus, at high levels of involvement, attentional selectivity should overcome the positive effect of processing intensity on the processing of sponsorship stimuli. Highly involved people are no longer "willing" to process irrelevant billboards when watching a soccer game. Therefore, a curvilinear relationship is hypothesized between audience involvement with a sponsored event and the processing of sponsorship stimuli. This should be reflected on people's ability to recognize the stimuli in a subsequent task.

H1 Felt involvement with a sponsored event has an curvilinear (inverted-U) effect on the recognition of embedded sponsorship stimuli.

This curvilinear relationship between felt involvement and the recognition of sponsorship stimuli is essentially driven by people's motivation to process the event. It is argued that this effect is distinct from the effect of arousal, discussed here below.

Effect of Arousal

Sports events can induce intense affective reactions among audience members. The intensity dimension of affective reactions is often operationalized as arousal (e.g., Clark 1982; Pavelchak et al. 1988). Arousal refers to the activity of the autonomic nervous system, particularly in its sympathetic division (Mandler 1975). Its experience is defined as a feeling state varying along a single dimension from drowsiness to frantic excitement (Mehrabian and Russell 1974, Humphreys and Revelle 1984).

In dual task studies, heightened arousal has been shown to impair performance on the secondary task (Eysenck 1982). Easterbrook (1959) suggested that heightened arousal produces an attention narrowing process, restricting the range of cue utilization. Primary task cues are attended to at the expense of secondary task cues. This attention narrowing process results from capacity limitations associated with high arousal. Two explanations can account for this increased attentional selectivity. First, it has been proposed that arousal cues arousal-related material in memory (Clark 1982; Clark, Milberg, and Ross 1983). Because more, and more diverse, material enters consciousness, there is a decrement in the amount of processing capacity available (Worth and Mackie 1987). It is also plausible that arousal in the autonomic nervous system generates internal cues that compete with external cues for the limited attentional capacity in the conscious field (Mandler 1975).

A parallel can be drawn between the dual task paradigm and the concurrent processing of a sponsored event and of embedded sponsorship stimuli. When watching sports event, limited processing capacity must be shared by the event itself (e.g., a tennis match) and other sources of information including sponsorship stimuli. As arousal increases, available processing capacity decreases. Because of this reduced ability to process information, attention is narrowed to the game itself. Therefore, arousal induced by a sponsored event will have a negative effect on the processing of embedded sponsorship stimuli. This reduced processing of the stimuli should result in a lower ability to subsequently recognize them.

H2 Arousal induced by a sponsored event has a negative effect on the recognition of embedded sponsorship stimuli.

High involvement with a sporting event should result in higher arousal in reaction to this event. People to whom the event is personally relevant should experience more intense affective reactions to it. Nevertheless, it is speculated that the curvilinear effect of involvement is distinct from the negative effect of arousal. Conceptually, whereas the former reflects people's motivation to process information, the latter results from their ability to process such information (Sanbonmatsu and Kardes 1988).

Effect of Pleasure

Besides their intensity, affective reactions can be characterized by their polarity (i.e., pleasure versus displeasure). Sporting events have the potential to create highly polarized affective states among audience members. It has been suggested that people learn more material when in a positive mood than when in a negative mood. For instance, Goldberg and Gorn (1987) observed that commercials were better remembered when inserted in a happy TV program than when inserted in a sad program. Mayer (1986) proposes that negative affective states impede learning by decreasing motivation and inducing the interjection of negative thoughts. Isen (1984) suggests that positive affective states produce changes in cognitive organization that enable a more efficient processing. Specifically, people in a positive affective state categorize material in fewer but broader categories. This use of broader categories allow them to organize the material more efficiently and consequently to memorize it better. In line with these arguments, it is hypothesized that event-induced pleasure will have a positive effect on the recognition of sponsorship stimuli.

FIGURE 1

HYPOTHERSIZED RELATIONSHIPS AMONG INVOLVEMENT, AROUSAL, PLEASURE, AND RECOGNITION OF SPONSORSHIP STIMULI

H3 Pleasure experienced in reaction to a sponsored event has a positive effect on the recognition of embedded sponsorship stimuli.

The path diagram in Figure 1 summarizes the hypothesized effects of involvement, arousal, and pleasure. Although they are not central in this study, the expected relationships between these independent variables are also presented. It is anticipated that both arousal and pleasure will be enhanced by an increase in felt involvement. As already stated, people to whom the sporting event is personally relevant should react more intensely than people to whom the event is less relevant. Also, although highly involved people might feel displeasure if their favorite team or player is losing, on average, watching a sporting event should be a more pleasurable experience to involved viewers than to noninvolved ones. Pleasure and arousal are depicted as uncorrelated because they are conceptually orthogonal constructs (Russell, Weiss, and Mendelson 1989). However, a positive correlation is sometimes observed between them (e.g., Pavelchak et al. 1988). An artificial variable, the square of felt involvement, is included to express the curvilinear effect of involvement on the recognition of sponsorship stimuli (H1). Felt involvement and its square are described here as uncorrelated, which is expected if the former is normally distributed and expressed in deviation form. Arousal will decrease the recognition of sponsorship stimuli (H2); pleasure will increase it (H3).

METHOD

Overview. Eighty-five undergraduate students at a Belgian university took part in the study to fulfill a course requirement. They watched, in a single session, a 25-minute edited version of an unaired recent soccer game with embedded billboards. After viewing the game, subjects were administered a surprise recognition test concerning the embedded billboard.

Material. Given the objectives of the study, it was necessary to generate sufficient involvement, arousal, and pleasure among the subjects. The perceived relevance of the material to the subjects was enhanced in three ways. First, the selected game concerned a confrontation between two of the most prestigious Belgian teams. Second, it was known in advance that the game would not be broadcast prior to the study, except for short excerpts during TV news. (This was also essential for the internal validity of the study.) Finally, the study took place only six days after the actual game, and before either team had played another game. A full recording with original commentaries was made available by Belgian French speaking Television (RTBF). The recording was professionally edited in order to select the most interesting sequences (as in real televised summaries) and to control exposure to the billboards. In order to obtain sufficient power (Sawyer and Ball 1981), the edited program was 25-minute long (i.e., much longer than usual summaries).

Procedure. The study was run in a single session in order to minimize communication effects. The projection of the stimulus material took place in a conference hall with excellent viewing facilities. The subjects were first assigned well-spaced seats, and then given the following cover story: "The Belgian Soccer League has recently observed a steady increase in the attendance at the matches of the Top League Championship (a slide was projected to support graphically this true assertion). The Soccer League attributes it notably to increased quality of the games resulting in a higher potential to interest the viewer. Our study aims at analyzing the ability of the matches to induce interest among TV viewers". They then completed a questionnaire concerning their possible "centers of involvement" (i.e., soccer in general, or a team in particular), their level of enduring involvement toward each "center", and their knowledge of soccer. They were also questioned about their acquaintance with the particular game (e.g., unseen, seen portions during TV news, attended the game).

After the first booklet had been collected, subjects were shown the game on a giant screen. They were instructed to imagine that they were in their living room watching sports news on TV.

After the projection, a second booklet containing the independent variables measures was distributed. Subjects were asked to report their felt involvement while watching the game and their affective reactions (pleasure and arousal) to it. To measure any demand characteristics, control questions were included about the objectives of the study. After the second booklet was collected, the recognition task was administered. After completing this task, subjects were asked to estimate how well they performed on it. They also explained why they thought that their recognition performance was good or poor.

Independent Variables. Zaichkowsky's (1985) 20-item PII scale was adapted to better capture the phenomenological nature of felt involvement. Zaichkowsky's antonyms (e.g., "important/ unimportant") were inserted into small sentences such as "Watching this game has been important/ unimportant to me" and "I have found this game boring/interesting." In order to measure experienced pleasure and arousal, the scales developed by Mehrabian and Russell (1974) were adapted in a similar way. For example, the 6-item pleasure scale included the item "While watching the game, I was happy/unhappy." The 6-item arousal scale included items such as." During the game I was jittery/dull". All three scales were presented in a single random order with end-points counterbalanced. Because the effects of involvement are easily confounded with the effects of knowledge, the latter was also measured in this study. Thirteen multiple-choice questions about soccer and the teams, with various degrees of difficulty, were developed. The measure of knowledge was simply the number of correct answers.

Dependent variable. Recognition of the sponsorship stimuli was measured on the basis of 8 billboards that were clearly visible during the program (mean duration of exposure= 118.5 sec). Each of these 8 billboards was projected to the subjects for 10 seconds among 3 other billboards that did not appear in the program. No other cue but the billboard itself (e.g., a player, a portion of the stadium) was projected. The subjects were asked to indicate which of the billboards presented in groups of 4 they recognized as appearing during the program. They were instructed not to rely on their beliefs nor to attempt to guess. The recognition score was simply the number of correct answers.

RESULTS

Control of Artifacts

Control questions about the objectives of the study were asked just after the projection of the game and before the recognition test. They revealed that no subject guessed the true purpose of the study. After the recognition test, as a second control, subjects were asked to explain their recognition performance. Seven subjects explicitly attributed their good performance to their acquaintance with the stadium or with the home team; they were thus disregarded for subsequent analyses. Two other subjects were dropped because of evident carelessness. The final sample was composed of 76 subjects, 35 males and 41 females (mean age = 19.2). While 16 of these subjects had been previously exposed to small portions of the game broadcast during TV news, there was no significant effect of this previous exposure on the recognition score (F(1,73)=.09, p > .77). Hence, these 16 subjects were retained in the analyses.

Reliability and Discriminant Validity of the Independent Variables

Scales for felt involvement (Cronbach's alpha=.98), arousal (alpha =.86), and pleasure (alpha=.88) exhibited high internal consistency. To test the discriminant validity of the felt involvement, arousal, and pleasure constructs, the 32 items measuring them were factor-analyzed. Because involvement and arousal were anticipated to be highly correlated, an oblique rotation (OBLIMIN) was applied to the extracted factors. A three-factor solution accounted for 72% of the total variance.

The rotated factor loadings (not reported here because of space limitation, but available from the author) clearly indicated that three different constructs were indeed measured. All felt involvement items loaded on factor 1, pleasure items on factor 2, and arousal items on factor 3. As anticipated, factor 1 (involvement) and factor 3 (arousal) were strongly correlated (r=.462). The correlation between factor 1 and factor 2 (pleasure) was r=.368. Although pleasure and arousal are theoretically orthogonal (Russell, Weiss, and Mendelson 1989), factor 2 and factor 3 were positively correlated in this study (r=.264).

Effects of Involvement, Pleasure and Arousal

To review, felt involvement was hypothesized to have a curvilinear effect on the recognition of sponsorship stimuli (H1). Arousal was hypothesized to have a negative effect, distinct from that of felt involvement (H2). Pleasure was hypothesized to have a positive effect on recognition (H3). The hypothesized relationships depicted in Figure 1 were tested using LISREL-VI maximum likelihood procedure (J÷reskog and S÷rbom, 1984). It is recognized that the inclusion of a squared variable (felt involvement squared) violates the assumption of multivariate normal distribution made by maximum likelihood estimation. However, a previous study (Hayduk 1987) also using a squared variable in LISREL, found maximum likelihood estimates to be little affected by this violation. Hence, maximum likelihood estimates are reported here. (generalized least squares estimates were very similar.)

FIGURE 2

OBSERVED RELATIONSHIPS AMONG INVOLVEMENT, AROUSAL, PLEASURE, AND RECOGNITION OF SPONSORSHIP STIMULI

Rather than directly path-analysing the observed summated scores (see e.g., Pavelchak et al. 1988), it was judged preferable to take measurement error into account. The recognition score had an estimated reliability of KR-20 = .62. Following common practice, this measurement error was introduced in the model by setting the corresponding factor loading equal to the square root of the reliability estimate and the measurement error equal to 1 minus this estimate. Although both pleasure and arousal were measured with good internal consistency (alpha = .88 and .86, respectively), it was decided to split each of the two 6-item scales into two 3-item scales to have two separate measures of each construct. Finally, since felt involvement had a very high reliability (alpha = .98), it was assumed that the observed summated score was an error-free single indicator of the latent construct. This allows to consider the squared involvement score to be itself an error-free measure of the latent squared factor. (This is clearly a simplifying assumption, but it would also have been made had path analysis or multiple regression been used instead.)

The final estimated model is depicted in Figure 2. It shows a very good fit to the data (Chi-square (df=10)= 13.94, p > .18), goodness-of-fit index = .955, root mean square residual= .042). Felt involvement had a strong positive effect on experienced arousal (.617, t= 6.88, p < .001). It also increased subjects' pleasure experiences when viewing the game(.528, t= 5.39, p < .001). There was a residual correlation between pleasure and arousal that was no accounted by the structural model (.140, t = 2.01, p < .05). This is probably because, on average, subjects experienced more pleasure than displeasure in reaction to the game (mean = 27.07 on a scale from 6 to 42). This resulted in an overall polarized affective reaction that was accompanied by heightened arousal. In support for the first hypothesis, the effect of felt involvement on the recognition of the stimuli was clearly curvilinear. The path from (unsquared) felt involvement to recognition was positive (.722, t= 3.19, p < .01), whereas simultaneously the path from felt involvement squared to recognition was negative (-.396, t = -2.26, p < .01). It can be observed that felt involvement, although expressed in deviation from its mean, was strongly correlated with its square. This unexpected correlation probably stemmed from the skewness of this sample's felt involvement distribution. Even when the effects of felt involvement and its square were accounted for, arousal significantly decreased the recognition of the stimuli (-.440, t= -1.898, p < .05). This effect of arousal corroborates the second hypothesis. However, it is possible that involvement has a curvilinear effect on recognition only because it has both a positive direct influence and a negative indirect influence mediated by arousal. To test that mediation explanation, the direct path from squared involvement to recognition was removed, resulting in a nested "mediation" model. This mediation model fitted the data significantly worse than the model depicted in Figure 2 (Chi-square difference (df=1) = 4.98, p < .05). Adding a path from felt involvement squared to arousal did not improve the fit of the mediation model (Chi-square difference (df=1) = 1.01, p > .30). Therefore, it appears that the curvilinear effect of felt involvement and the negative effect of arousal are distinct from each other. Contrary to the prediction of third hypothesis, the effect of pleasure on recognition was not significant (t = .615, p > .25). The corresponding path was consequently dropped. Felt involvement (squared and unsquared) and arousal accounted for 22% of the variance in the recognition of sponsorship stimuli.

Effects of Knowledge and Gender

An alternative interpretation of the data is that the effects of felt involvement are confounded with the effects of knowledge. Therefore, a measure of subjects' knowledge of soccer and of the teams was taken before they saw the program. This knowledge index had a strong internal consistency (KR-20 =.84). As could be expected (Celsi and Olson 1988; Sujan 1985), it was strongly correlated with felt involvement (r = .595, p <.001). To test for this potential confound, a knowledge factor (adjusted for measurement error) was included in the causal model as a predictor of recognition. Its effect on recognition was not significant (t = .885, p > .15).

Marked gender effects were also observed. As could be expected, females had less knowledge of soccer and of the teams than males (F(1,74)= 27.465, p < .001, Omega-squared =.26), and reported a lower level of felt involvement (F(1,74)= 18.826, p < .001, Omega-squared =.19). Females were also less aroused (F(1,74)= 4.417, p < .039, Omega-squared =.043), but their pleasure when watching the game did not differ significantly from males (F(1,74)= 2.457, p > .12). Females scored consistently poorer (M= 2.65) than males (M = 3.80) on the recognition test (F(1,74)= 8.195, p < .01). Unexpectedly, their performance was still significantly lower when, previous exposure to the game, knowledge, arousal, felt involvement, and its square were entered as covariates (F(1,67)= 4.053, p < .05, Omega-squared =.037).

CONCLUSION AND DISCUSSION

In a recent editorial, Lutz (1991) called for more research on substantive consumer behavior issues that are important to someone other than the researchers themselves. Sponsorship communication is important to marketers. If they are to continue to rely on generated exposure as a measure of sponsorship effectiveness, understanding what happens beyond this exposure is clearly needed. In this study, extensive exposure to the billboards yielded only limited effects on recognition. On average, the billboards were visible for 118.5 seconds. Yet, on average, they were recognized by only 39 percent of the subjects (although the a priori probability to choose the correct billboard in a set of four is 25%). This replicates the findings of previous controlled studies (d'Ydewalle et al. 1987; Nebenzahl and Hornik 1985). That billboards achieved only limited recognition stresses the need to identify potential influences on the processing of sponsorship stimuli. Three of these were studied here, (felt) involvement, arousal, and pleasure. Felt involvement was found to have an inverted-U effect on the recognition of sponsorship stimuli, arousal a negative effect, but pleasure no effect. Involvement and arousal accounted for a fairly large proportion of recognition variance. This has some implications both for the selection of events for sponsorship and for the evaluation of sponsorship effectiveness. For a given audience size, highly involving and arousing events might not offer the sponsor his/her best advantage. On the other hand, it is also true that potentially involving and arousing events are likely to attract a larger audience.

Pleasure did not affect the recognition of the billboards. The effects of pleasure on memory for billboards may be too subtle to be captured in a recognition task. It may also be that the incidental learning of billboards does not lend itself to the processing efficiency principles suggested by Isen (1984). These principles may apply only to more goal-directed processing. In a related context, it was also found that pleasure experience when watching a football game did not increase recall for the embedded commercials (Pavelchak et al. 1988).

The distinction between involvement and arousal needs to be clarified, especially when involvement is defined as a state of arousal or activation (e.g., Cohen 1983; Mitchell 1979; Rothschild 1984). For instance, Thorson et al. (1985) purportedly measured and manipulated involvement with TV programs. However, both their manipulation of program involvement (sexual program segments) and their measurement of involvement (cortical arousal recorded by EEG) could be interpreted in terms of arousal. In this study, a distinction was drawn between (felt) involvement and (physiological) arousal. One difference between these two constructs is that involvement is object-specific (e.g., Celsi and Olson 1988; Mitchell 1979; Zaichkowsky 1985), whereas arousal is nonspecific (Derryberry 1988; Eysenck 1976) and can be misattributed (cf. Schachter and Singer 1962). This distinction between involvement and arousal seems to exhibit construct validity, even in a context where these two constructs are likely to be confounded. A factor analysis indicated that distinct factors underlie the felt involvement and arousal scales. Moreover, this study shows, in a nomological sense, that involvement and arousal have distinct effects on a critical dependent variable. Conceptually, felt involvement reflects people's motivation to process the event and the embedded stimuli, whereas arousal affects their ability to do so (cf. Sanbonmatsu and Kardes 1988).

Males unexpectedly outperformed females on the recognition task. This gender effect remained significant when arousal, felt involvement, knowledge, and previous exposure to the game were controlled. A post hoc explanation can be suggested for this gender effect. As reflected by their knowledge scores, females were presumably less expert than males in processing soccer. A well known consequence of expertise in a task is a reduction of the amount of cognitive effort required to perform the task (cf. Alba and Hutchinson 1987). Therefore, processing the game (as demanded by the experimental task) should have been comparatively more effortful for females than for males. As females invested comparatively more effort in processing the game, less capacity was available for the processing the sponsorship stimuli. For instance, Navon (1984; Navon and Gopher 1979) has argued that an increased processing efficiency in the performance of a primary task should result in enhanced performance on a concurrent task that draws on the same pool of resources. This processing-expertise explanation of the gender effect is worth testing in future research.

Several limitations of this study deserve mention. First, the independent variables were not experimentally manipulated. Hence, third-variable explanations cannot be fully excluded. It should be noted, however, that a knowledge interpretation of the results has been ruled out. Second, the type of sponsorship stimuli investigated here is restricted to billboards. Other stimuli (e.g., brand names on players' shirts or on racing cars) that are closer to the audience members' "center of involvement" may be affected differently by involvement and arousal. For instance, in a racing context, it is possible that stimuli on the cars, provided that they are visible, may actually benefit from the increased focus of attention created by high involvement and arousal. In contrast, this study suggests that signs around the circuit should be negatively affected. The moderating effect of the proximity of the stimuli to the audience's center of involvement awaits future research. Third, no measure of focused attention was collected. Therefore, it is difficult to ascertain whether the observed curvilinear effect of involvement and the negative effect of arousal resulted, as hypothesized, from increased attentional selectivity. Fourth, background factors such as subjects' familiarity with the brands advertised on the billboards were not controlled. This is a common drawback of using real marketing stimuli instead of artificial ones. However, to have biased these result, these background factors should have had very specific interactions with the independent factors. Finally, this study only considered conscious processing of and explicit memory for the stimuli. However, there is growing evidence that even nonattented material that is subconsciously processed can be influential (e.g., Janiszewki 1990). The subconscious processing effects of sponsorship stimuli deserve future research.

[The data reported in this paper were collected while the author was at the Catholic University of Mons (FUCAM, BELGIUM). The author is indebted to Michel Bomboir for his assistance at various stages of the study and to Richard Celsi, Christian Derbaix, Susan Fournier, Rich Lutz, Piet Vanden Abeele and Luk Warlop for helpful comments on earlier drafts of this paper.]

REFERENCES

Abratt, Russell, Brian C. Clayton, and Leyland F. Pitt (1987),"Corporate Objectives in Sports sponsorship," International Journal of Advertising, 6, 299-311.

Alba, Joseph W. and J. Wesley Hutchinson (1987), "Dimensions of Consumer Expertise," Journal of Consumer Research, 13 (March), 411-454.

Andrews, Craig J. (1988), "Motivation, Ability and Opportunity to Process Information : Conceptual and Experimental Manipulation Issues," in Advances in Consumer Research, Vol. 15, M.J. Houston, ed., Provo:UT, Association for Consumer Research, 219-224.

Axelrod, Joel n. (1963), "Induced Moods and Attitudes toward Products," Journal of Advertising Research, 3, 19-24.

Batra, Rajeev and Michael L. Ray (1986), "Situational Effects of Advertising: The Moderating Influence of Motivation, Ability, and Opportunity to Respond," Journal of Consumer Research, 12 (March), 432-445.

Burnkrant, Robert E. and Alan G. Sawyer (1983), "Effects of Involvement and Message Content on Information Processing Intensity," in Information Processing Research in Advertising, ed. Richard J. Harris, Hillsdale, NJ: Lawrence Erlbaum Associates, 43-64.

Celsi, Richard and Jerry C. Olson (1988), "The Role of Involvement in Attention and Comprehension Processes," Journal of Consumer Research, 15 (September), 210-224.

Clark, Margaret S. (1982), A Role for Arousal in the Link between Feeling States, Judgments and Behavior, in Affect and Cognition, M. Clark and S. Fiske, eds., N.J., Lawrence Erlbaum Associates, 263-289.

Cohen, Joel B. (1983), "Involvement and You : 1000 Great Ideas," Advances in Consumer Research, Vol.10, R. Bagozzi and A. Tybout (eds.), Ann Arbor, MI: Association for Consumer Research, 161-170.

Derryberry, Douglas (1988), "Emotional Influences on Evaluative Judgments: Roles of Arousal, Attention, and Spreading Activation," Motivation and Emotion, 12 (1), 23-55.

d'Ydewalle, GTry, Piet Vanden Abeele , Johan Van Rensbergen, and Patrick Coucke (1987), "Incidental Processing of Advertisements while Watching Soccer Games Broadcasts," in Practical Aspects of Memory : Current Research and Issues, M. Gruneberg, P. Morris and R. Sykes (eds.), John Wiley & Sons, 478-483.

Easterbrook, J.A. (1959), "The Effects of Emotion on Cue Utilization and the Organization of Behavior," Psychological Review, 66 (May), 183-201.

Eysenck, Michael W. (1979), "Arousal, Learning, and Memory," Psychological Bulletin, 83, 3, 389-404.

Eysenck, Michael W. (1982), Attention and Arousal, Cognition and Performance, New York:Springer-Verlag.

Gardner, Meryl P. and Philip J. Shuman (1987), "Sponsorship: An Important Component of the Promotion Mix," Journal of Advertising, 16, 1, 11-17.

Golberg, Marvin E. and Gerald Gorn (1987), "Happy and Sad TV Programs: How They Affect Reactions to Commercials," Journal of Consumer Research, 14 (December), 387-403.

Greenwald, Anthony G. and Clark Leavitt (1984), "Audience Involvement in Advertising: Four Levels," Journal of Consumer Research, 11 (June), 581-592.

Hayduk, Leslie A. (1987), Structural Equation Modeling with LISREL, Baltimore: Johns Hopkins University Press.

Houston, Michael J. and Michael L. Rothschild (1978), "Conceptual and Methodological Perspectives in Involvement," in Research Frontiers in Marketing: Dialogues and Directions, S. Jain (ed.), Chicago, IL: American Marketing Association, 184-187.

Isen, Alice M. (1984), "The Influence of Positive Affect on Decision-Making and Cognitive Organization," in Advances in Consumer Research, Vol. 11, ed. Thomas Kinnear, Provo, UT: Association for Consumer Research, 534-537.

Janiszewski, Chris (1990), "The influence of Nonattended Material on the Processing of Advertising Claims," Journal of Marketing Research, 27 (August), 263-278.

J÷reskog Karl and Dag S÷rbom (1984), LISREL VI : Analysis of Linear Structural Relationships by the Method of Maximum Likelihood, Upsala: University of Upsala.

Kahneman, Daniel (1973), Attention and Effort, Englewood Cliffs, NJ: Prentice-Hall.

Lutz, Richard (1991), "Editorial," Journal of Consumer Research, 17 (March).

MacInnis, Deborah J. and Bernard J. Jaworski (1989), "Information Processing from Advertisements: Toward an Integrative Framework," Journal of Marketing, 53 (October), 1-23.

Mayer, John D. (1986), "How Mood Influences Cognition," in Advances in Cognitive Science, ed. N.E. Sharkey, Chichester, England: Ellis Hortwood, 290-314.

Mehrabian, Albert and James Russell (1974), An Approach to Environmental Psychology, Cambridge, MA: MIT Press.

Meenaghan, John A. (1984), "Commercial Sponsorship," European Journal of Marketing, 7, 2-67.

Meenaghan, John A. (1989), "Sponsorship : An up-and-coming Medium Captures Advertisers' Fancy," International Advertiser, (March-April), 15-17.

Mintel (1986), Special Report on Sponsorship, London.

Mitchell, Andrew A. (1979), "Involvement: A Potentially Important Mediator of Consumer Behavior", in William L. Wilkie (ed.), Advances in Consumer Research, 6, Ann Arbor: Association for Consumer Research, 191-195.

Mitchell, Andrew A. (1981), "The Dimensions of Advertising Involvement", in Kent B. Monroe (ed.), Advances in Consumer Research, 8, Ann Arbor: Association for Consumer Research, 25-30.

Navon, David (1984), "Resources-A Theoretical Soup Stone," Psychological Review, 91 (2), 216-234.

Navon, David Daniel Gopher (1979), "On the Economy of the Human-Processing System," Psychological Review, 86 (3), 214-255.

Nebenzahl, I. and Hornik J. (1985), "An Experimental Study of the Effectiveness of Commercial Billboards in Televised Sports Arenas," International Journal of Advertising, 4, 27-36.

Otker, Ton (1988), "Exploitation: The Key to Sponsorship Success," European Research, 16, 2 (May), 77-86.

Pavelchak, Mark A, John H. Antil, and James M. Munch (1988),"The Super Bowl: An Investigation into the Relationship Among Program Content, Emotional Experience, and Ad Recall," Journal of Consumer Research, 15 (December), 360-366.

Petty, Richard E., John T. Cacioppo, and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness : The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-146.

Rothschild, Michael L. (1984), "Perspectives on Involvement: Current Problems and Future Directions," in Advances in Consumer Research, Vol.11, T. Kinnear (ed.), P.216-217.

Russell, James A., Anna Weiss, and Gerald A. Mendelson (1989),"Affect Grid: A Single-Item Scale of Pleasure and Arousal," Journal of Personality and Social Psychology, 57, 3 (March), 493-502.

Ryssel, Christian and Erich Stamminger (1988), "Sponsoring World-Class Tennis Players," European Research, 16, 2 (May), 110-116.

Sanbonmatsu, David M. and Frank R. Kardes (1988),"The Effects of Physiological Arousal on Information Processing and Persuasion," Journal of Consumer research, 15 (December), 379-385.

Sandler, Dennis M. and David Shani (1989), "Olympic Sponsorship vs. 'Ambush' Marketing: Who gets the gold ?," Journal of Advertising Research, 29 (Aug/Sept), 9-14.

Schachter, Stanley and Jerome E. Singer (1962), "Cognitive, Social, and Physiological Determinants of Emotional States," Psychological Review, 69, 5 (September), 379-403.

Sawyer, Alan G. and Dwayne A. Ball (1981), "Statistical Power and Effect Size in Marketing Research," Journal of Marketing Research, 18 (August), 275-290.

Sujan, Mita (1985), "Consumer Knowledge: Effects On Evaluation Strategies Mediating Consumer Judgments," Journal of Consumer Research, 12 (June), 31-46.

Thorson, Esther, Byron Reeves, Joan Schluedes, Annie Lang , and Michael Rothschild (1985), "Effects of Program Context on the Processing of Television Commercials," Proceedings of the 1985 Conference of the American Academy of Advertising, ed. Nancy Stephens, Tempe, Az : College of Business, Arizona State University, R58-R63.

Waite, Nigel (1979), "Sponsorship in Context," Unpublished PhD thesis, Cranfield Institute of Management.

Welling B. (1986), "The Biggest Game on Earth," Business Week, June 2, 20-22.

Worth, Leila T. and Diane M. Mackie (1987), "Cognitive Mediation of Positive: Affect in Persuasion," Social Cognition, 5, 1, 76-94.

Zaichkowsky, Judith L. (1985), "Measuring the Involvement Construct," Journal of Consumer Research, 12 (December), 341-352.

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