Information Processing To understand and explain consumer responses to coordinated television-web (TV-W) campaigns, we should understand the processes behind the ad encoding and retrieval. According to Edell and Keller (1989) each sensory mode used (e.g. audio, video) has ability either to evoke direct cognitive or affective responses or serve as a distraction to other modes. A typical TV ad uses multiple sensory modes by combining audio and video at the same time. The multimode nature of TV could lead to more processing effort since it has more elements to process (Edell and Keller 1989) but also can interfere with complete comprehension of the ad because of high level of information load (Bither and Wright 1973). A typical web ad today contains only a basic video sensory mode expressed in static/animated image and text. In its current state, the web ad is closer to “print media” than to any other type of mass media. The other distinction between TV and web is “preprogrammed” nature of the former and interactive nature of the latter. TV advertising is mostly out of person's control (assuming its not recorded on a video tape), whereas on the web a person generally has a choice whether to view the ad by clicking on it or discarding it all together. Wright (1974) argues that multimode, preprogrammed and embedded in the continuos flow of transmission nature of TV advertising may create a strainful reception environment that could limit opportunity for active thinking. On the other hand, a web ad, much like a print ad has a high degree of referability since a person can control the flow of information and can always go back and spend more time and effort to process the ad information (Wright 1981).
After the first exposure, the ad leaves a strong or weak trace in a person’s memory (depending on the focal attention and processing intensity at the time of the exposure) which then could be retrieved by exposure to the ad for the second time (Edell and Keller 1989). Edell and Keller (1989) indicate that when the second ad is presented, a person might engage in three different processing activities: retrieval of the old ad, comprehension of the new ad, and elaboration. The authors further suggest that “these three processing activities…must compete for the available processing resources,” (p. 150). Previous research within the same media (Batra and Ray 1986) suggests that exposure to the same ad demotes comprehension and retrieval efforts and magnifies the elaboration process. As for the mixed media exposures, Edell and Keller (1989) reported that subjects in mixed TV-Radio ad sequence have the same level of brand name recall as the TV-TV ad sequence. Apparently, the subjects in their study were involved in retrieval process of the initial ad video elements and generated more thoughts about the video. In this case, the audio track served as a strong retrieval cue (Edell and Keller 1989) Additionally, Keller (1987) found that point of purchase displays using image and verbal components of a print ad can improve the retrieval of the ad memory trace. In the TV-W ad sequence, an image and animated verbal components (ticker tape) used in the web ad can serve as retrieval cues for the initial TV ad exposure, having similar effect as audio track (Edell and Keller 1989) and point of purchase displays (Keller 1987). When exposed to the web ad, a person may try to retrieve the initial TV ad exposure which can result in more processing since people will try to visualize the video representation looking at the image and retrieve audio claims by reading the ticker tape.

Research Proposition The preceding conceptual discussion allows us to present the research proposition to be tested in a laboratory experiment setting. For better understanding of the proposition the conditions of the experiment are presented. We will compare 4 groups:
1. High Involvement Product-TV-TV: two exposures to the TV ad
2. Low Involvement Product-TV-TV: two