This robust structural modeling study, with over 23,000 responses to 240 advertising messages, found that affect when measured by a visual measure of emotional response dominates over cognition for predicting conative attitude and action. AFFECTIVE AND COGNITIVE-BASED ATTITUDEFor many years, there was a tendency of focusing on cognitive-based attitude, suggesting that, with advertising involvement, cognition predominates over affective processing and that affective reactions are always mediated by cognition (Greenwald and Leavitt, 1984; Tsal, 1985). In fact, the derivation and strength of the attitude toward the ad (Aad) process is based on the relationship between attitude toward the ad and attitude toward the brand (Abd), and the determination that Abd predicts purchase intention (Mitchell and Olson, 1981;Lutz, MacKenzie, and Belch, 1983;MacKenzie and Lutz, 1986). Fishbein (Fishbein and Middlestadt, 1995) also heralded the notion of cognitive-based attitude by suggesting that a consumer's attitude is a function of (cognitive) beliefs and those beliefs predict intentions of behavior.Studies examining the role and relationship of emotion as the mediator of responses to advertising (Edell and Burke, 1987; Holbrook and Batra, 1987), however, have found that cognition can drive affect. In fact, some researchers (Brown and Stayman, 1992; Cohen and Areni, 1991; Petty et al., 1991) have argued that affect can directly influence attitude and that cognitive-based models fail to properly measure feelings associated with the sources of information (Edell and Burke, 1987; Schwarz, 1997). Failing to understand the role of

others. Over the past few years, he has been developing a model (AdSAM) for analysing emotional response to marketing communications. ChongMoo Woois a doctoral student majoring in advertising in the Department of Advertising at the University of Florida. He received a BA and MA, both in advertising and public relations from the Chung-Ang University, Seoul, South Korea, and an MA in advertising from the University of Florida. During the programme of study, he has completed course work toward an MA in statistics and has published in the Abstract The Elaboration Likelihood Model (ELM), developed by Petty and Cacioppo, proposed two routes to attitude change: central and peripheral. The central route emphasises a high relevance of the message to the individual. In the peripheral route, the individual concentrates on heuristic cues like attractive expert sources and number rather than the content of arguments employed by the message to process the message. If these cues produce an attitude change, this change is likely to be shorter lasting and unpredictable of that individual's behaviour. Hence, the cognitive (central) aspect of the ELM overshadows its affective (peripheral) aspect, and the underlying suggestion of this model is that an attitude change is mostly reached through cognition as opposed to emotion. This study attempts to show that the emotional aspect is as important as the cognitive aspect. The basis for this conclusion is that even as an individual processes a message cognitively, that cognition has an emotional core. In addition, there is a possibility that content processing (elaboration) gives rise to emotions and that this leads to a longer-lasting change in attitudes.

No abstract