Journal of Communication ISSN 0021-9916 ORIGINAL ARTICLE The Relationship Between Message Recall and Persuasion: More Complex Than It Seems * Christopher J. Carpenter 1 & Franklin J. Boster 2 1 Department of Communication, Western Illinois University, Macomb, IL 61455, USA 2 Department of Communication, Michigan State University, East Lansing, MI 48824, USA A strong message recall-attitude relationship was predicted when participants were induced to use memory-based processing to form their attitudes but not when induced to use on-line processing after exposure to a persuasive message. The recall-attitude correlation in the memory-based conditions was expected to be positive when the arguments were strong and negative when they were weak; 240 participants participated in a study to test these predictions. An ample and positive recall-attitude relationship was found when participants were exposed to strong arguments in both processing conditions. A small but negative relationship was found in the weak arguments, on-line processing condition but not the memory-based condition. Parallel findings were obtained when need to evaluate was used to measure processing type. doi:10.1111/jcom.12042 In his seminal work on information processing, McGuire (1968) argued that per- suasion proceeds through stages, the early stages requiring that the audience attend to the message and learn the message content, as he expected message learning to be a prerequisite to persuasion. He argued that strong message recall indicates greater learning of the message. If the audience pays enough attention to the message and understands it well, McGuire believed that audience members would be likely to recall a greater proportion of the message than someone who did not. Subse- quently, several studies used recall as a standard measure of message comprehension and reported substantial, positive associations between message recall and message acceptance (Chaiken & Eagly, 1976; Eagly, 1974; Haugtvedt & Petty, 1992; Regan & Cheng, 1973). Corresponding Author: Christopher J. Carpenter; e-mail: cj-carpenter2@wiu.edu * This study was conducted as part of the first author’s doctoral dissertation under the direction of the second author. Journal of Communication 63 (2013) 661–681  2013 International Communication Association 661