C. Yoo, D. Macinnis /Journal of Business Research 58(2005)1397-1406 1399 by beliefs engendered from the communication(e.g, Batra information and because information is less directly stated, and Stephens, 1994; MacInnis and Jaworski, 1989; Drodge beliefs may be weak and play a less powerful role than Aad 1989: Mitchell and Olson, 1981) in affecting brand attitude. Although beliefs may be weakly established in this emotional format condition. when an 2. 2. Brand attitude formation process using an emotional emotional ad creates strong feelings and creates favorable thoughts about the ads credibility and meaningfulness, it creates a sense that the brand is"for them"independent of When the ad execution format is emotional, the primary specific attribute-oriented beliefs. Thus, we expect that responses to advertising exposure are likely to be feelings- H2. When the ad execution is emotional, thoughts about the as the ad execution is designed to appeal to the viewer's ads credibility influence brand attitudes(a)directly and (b) emotions. Feelings evoked in response to emotional ads are indirectly through the mediating influence of Aad hypothesized to influence Aad in two ways. First, they may do so directly, through an affective transfer mechanism. Past If an ad is regarded as more meaningful and relevant to research is consistent with this proposed effect (e. g, the self, consumers may form stronger beliefs about the niszewski, 1988). For example, the literature on mood product's ability to deliver its claims. Edell and Burke states indicates that mood can bias evaluations and judg-(1987) and Burke and Edell(1989) found that feelings and ments in mood congruent conditions(Gardner, 1985; Isen, ad evaluations influenced beliefs and that feelings influ 1989). Second, the impact of feelings on Aad may also be enced beliefs directly and indirectly through ad evaluations mediated by thoughts about the ads credibility. Emotional We expect that for ads with an emotional format this same ads may draw the consumer into the ad and allow them to route occurs. Beliefs may be inferred based on the emotional experience vicariously the experiences of the actor(Batra ad content and the associated feelings and evaluations it and Stayman, 1990; MacInnis and Stayman, 1993). To the produces. For example, based on an emotional ad, which extent that feelings associated with the actor(s) can be shows a daughter talking to her father on the telephone, a experienced by the viewer, the ads may be seen as more consumer may believe that a long-distance phone service convincing(Bagozzi and Moore, 1994). Ads regarded as elps bring people closer together. Feelings of warmth from convincing should in turn be better liked. Hence part of thethe commercial may engender such beliefs. Thus we effect of feelings on Aad may be mediated through thoughts propose about the ads credibility. Relatedly, prior research suggests that positive feelings induced by mood can create more H3. When the ad execution is emotional, feelings influence cognitive flexibility, allowing consumers to draw associa- beliefs(a) directly and(b) indirectly through the mediational tions and see things as meaningful that might not have been influence of credibility seen in the absence of such a mood(e.g, Lee and Sternthal, 1999: Barone et al., 2000). Since meaning of the ad is H4. When the ad execution is emotional, beliefs influence brand attitudes enhanced by mood-induced cognitive flexibility, we would expect that positive feelings induced by an ad would lead a view of the ad as meaningful and that responses about 2.3. Brand attitude formation process using an meaningfulness would induce more favorable attitudes informational ad forman toward the brand. Thus we propose that HI. When the ad execution is emotional, feelings influence An informational ad execution is designed to appe Aad (a) directly and(b) indirectly through the mediating the rationality of the audience(MacInnis et al., 2002; influence of credibility thoughts Vaughn, 1980). Hence, as Fig. I suggests, we expect that the primary response of audience concerns evaluative thoughts When the ad execution is emotional, we also expect that regarding the ads credibility. Credibility thoughts may in positive evaluative thoughts generated about the credibility turn affect consumers'feeling responses For example, an ad of the ad will contribute to consumers'attitudes toward perceived as credible is less likely to make consumers feel ad. Given that past research has found a strong relationship skeptical, irritated or annoyed. Likewise, when the ad between ad and brand attitudes under a variety of execution is informational, consumers should like ads that conditions, we expect that positive evaluative thoughts will are convincing and meaningful. Moreover, to the extent that affect brand attitudes in part through the mediational consumers feel less skeptical, irritated, and annoyed, they influence of Aad( Brown and Stayman, 1992). Considerable should like the ad better. Thus, positively valenced research has shown that attitudes toward ads are more credibility thoughts may also affect ad attitudes by positive for ads regarded as convincing and believable(e.g the negative feelings consumers have from the ad. Schlinger, 1979; Plummer, 1971; Aaker and Norris, 1982). although feelings have typically been modeled However, credibility may also directly affect brand attitudes. cedents to ad evaluations, here we propose that credibility Moreover, this direct effect may not operate through beliefs. evaluations drive feeling responses. Given that informa Specifically, because an emotional ad contains less factual tional ads are unlikely to elicit much emotion from theirby beliefs engendered from the communication (e.g., Batra and Stephens, 1994; MacInnis and Jaworski, 1989; Drodge, 1989; Mitchell and Olson, 1981). 2.2. Brand attitude formation process using an emotional Ad format When the ad execution format is emotional, the primary responses to advertising exposure are likely to be feelings— as the ad execution is designed to appeal to the viewer’s emotions. Feelings evoked in response to emotional ads are hypothesized to influence Aad in two ways. First, they may do so directly, through an affective transfer mechanism. Past research is consistent with this proposed effect (e.g., Janiszewski, 1988). For example, the literature on mood states indicates that mood can bias evaluations and judg￾ments in mood congruent conditions (Gardner, 1985; Isen, 1989). Second, the impact of feelings on Aad may also be mediated by thoughts about the ad’s credibility. Emotional ads may draw the consumer into the ad and allow them to experience vicariously the experiences of the actor (Batra and Stayman, 1990; MacInnis and Stayman, 1993). To the extent that feelings associated with the actor(s) can be experienced by the viewer, the ads may be seen as more convincing (Bagozzi and Moore, 1994). Ads regarded as convincing should in turn be better liked. Hence part of the effect of feelings on Aad may be mediated through thoughts about the ad’s credibility. Relatedly, prior research suggests that positive feelings induced by mood can create more cognitive flexibility, allowing consumers to draw associa￾tions and see things as meaningful that might not have been seen in the absence of such a mood (e.g., Lee and Sternthal, 1999; Barone et al., 2000). Since meaning of the ad is enhanced by mood-induced cognitive flexibility, we would expect that positive feelings induced by an ad would lead to a view of the ad as meaningful and that responses about meaningfulness would induce more favorable attitudes toward the brand. Thus we propose that: H1. When the ad execution is emotional, feelings influence Aad (a) directly and (b) indirectly through the mediating influence of credibility thoughts. When the ad execution is emotional, we also expect that positive evaluative thoughts generated about the credibility of the ad will contribute to consumers’ attitudes toward the ad. Given that past research has found a strong relationship between ad and brand attitudes under a variety of conditions, we expect that positive evaluative thoughts will affect brand attitudes in part through the mediational influence of Aad (Brown and Stayman, 1992). Considerable research has shown that attitudes toward ads are more positive for ads regarded as convincing and believable (e.g., Schlinger, 1979; Plummer, 1971; Aaker and Norris, 1982). However, credibility may also directly affect brand attitudes. Moreover, this direct effect may not operate through beliefs. Specifically, because an emotional ad contains less factual information and because information is less directly stated, beliefs may be weak and play a less powerful role than Aad in affecting brand attitude. Although beliefs may be weakly established in this emotional format condition, when an emotional ad creates strong feelings and creates favorable thoughts about the ad’s credibility and meaningfulness, it creates a sense that the brand is ‘‘for them’’ independent of specific attribute-oriented beliefs. Thus, we expect that: H2. When the ad execution is emotional, thoughts about the ad’s credibility influence brand attitudes (a) directly and (b) indirectly through the mediating influence of Aad. If an ad is regarded as more meaningful and relevant to the self, consumers may form stronger beliefs about the product’s ability to deliver its claims. Edell and Burke (1987) and Burke and Edell (1989) found that feelings and ad evaluations influenced beliefs and that feelings influ￾enced beliefs directly and indirectly through ad evaluations. We expect that for ads with an emotional format this same route occurs. Beliefs may be inferred based on the emotional ad content and the associated feelings and evaluations it produces. For example, based on an emotional ad, which shows a daughter talking to her father on the telephone, a consumer may believe that a long-distance phone service helps bring people closer together. Feelings of warmth from the commercial may engender such beliefs. Thus we propose: H3. When the ad execution is emotional, feelings influence beliefs (a) directly and (b) indirectly through the mediational influence of credibility. H4. When the ad execution is emotional, beliefs influence brand attitudes. 2.3. Brand attitude formation process using an informational ad format An informational ad execution is designed to appeal to the rationality of the audience (MacInnis et al., 2002; Vaughn, 1980). Hence, as Fig. 1 suggests, we expect that the primary response of audience concerns evaluative thoughts regarding the ad’s credibility. Credibility thoughts may in turn affect consumers’ feeling responses. For example, an ad perceived as credible is less likely to make consumers feel skeptical, irritated or annoyed. Likewise, when the ad execution is informational, consumers should like ads that are convincing and meaningful. Moreover, to the extent that consumers feel less skeptical, irritated, and annoyed, they should like the ad better. Thus, positively valenced credibility thoughts may also affect ad attitudes by reducing the negative feelings consumers have from the ad. Notably, although feelings have typically been modeled as ante￾cedents to ad evaluations, here we propose that credibility evaluations drive feeling responses. Given that informa￾tional ads are unlikely to elicit much emotion from their C. Yoo, D. MacInnis / Journal of Business Research 58 (2005) 1397 – 1406 1399