Data from a laboratory experiment which examines the relationship among awareness, attitude formation, and intent to behave over several levels of election races and several levels of advertising show that behavior is easily influencable by a high repetition level of innocuous advertising in the lower-level election; this phenomenon does not hold in the higher-level races. Furthermore, the authors find that this change in behavior is accomplished without an accompanying shift in attitude. Finally, it is found that the shift occurs equally among those subjects who were evaluated as having high involvement with the political process and those evaluated as having low involvement.he effects of the mass media on voting behavior have long been a source of communication research interest (Berelson, Lazarsfeld, and McPhee, 1954;Campbell, Gurin, and Miller, 1954;Campbell and Cooper, 1956; . Campbell, Converse, Miller and Stokes, 1960, 1966;Lazarsfeld, Berelson and Gaudet, 1948). But only in the last decade have studies focused on specific types of mass communication and their effects, not only on voting behavior, but also on the communication responses which might precede voting.The study reported here can be seen as a further step in this trend toward research on more specific components of