The amount of processing that message recipients give to persuasive messages has been identified as an important determinant of the nature of persuasive processes and effects. Dual-process models of persuasion, such as the elaboration likelihood model (Petty & Cacioppo, 1986; Petty & Wegener, 1999) and the heuristic-systematic model (Chaiken, 1987; Todorov, Chaiken, & Henderson, 2002), emphasize that as the amount of message processing varies so can the role that various elements play in influencing persuasive outcomes. For instance, where the communicator's credibility serves as a peripheral cue, increases in message processing commonly are associated with a decreased impact of credibility variations on persuasive effects (and an increased impact of argument-quality variations; see, e.g., Petty, Cacioppo, & Goldman, 1981).
A variety of factors have been identified as influencing the amount of processing message recipients give to persuasive messages. Many of these factors concern either some characteristic of the receiver (e.g., need for cognition; for a review, see Cacioppo, Petty, Feinstein, & Jarvis, 1996) or the relationship between the receiver and the topic of advocacy (involvement or personal relevance; e.g., Johnson & Eagly, 1989, 1990; Petty & Cacioppo, 1990). Obviously, however, it may also be valuable to explore how intrinsic features of messages might engender greater message scrutiny.
Among the various message features that might influence the degree of message processing, one natural candidate is whether the message's appeals are gain-framed or loss-framed. A "gain-framed" appeal emphasizes the desirable consequences associated with compliance with the advocated viewpoint; a "loss-framed" appeal emphasizes the undesirable consequences associated with noncompliance. Considerable research has addressed the question of the relative persuasiveness of gain-framed and loss-framed persuasive appeals (for some reviews, see O'Keefe & Jensen, 2006; Salovey, Schneider, & Apanovitch, 2002), but it is an open question whether gain- and loss-framed appeals systematically differ in the amount of message processing that they engender.
There are two reasons to suppose that loss-framed appeals will generally produce greater engagement with a message than will gain-framed appeals. The first is the observed effects of fear-arousing appeals on message processing. Abstractly, a fear appeal message has two components: One presents material designed to induce fear or anxiety (about a possible threatening event); the other presents a recommended action aimed at avoiding the fearful consequences. Although fear appeals need not be phrased using gain- and loss-framed language, implicitly (if not explicitly) the fear-arousal component emphasizes the disadvantages of noncompliance ("if you don't floss regularly, you can suffer horrible gum disease") and the recommended-action component emphasizes the advantages of compliance ("if you floss regularly, you can avoid gum disease").
Fear-inducing messages (compared to messages not inducing fear) often evoke greater message processing, as reflected in larger numbers of issue-relevant thoughts, increased differentiation of strong and weak arguments, and so forth (e.g., Baron, Logan, Lilly, Inman, & Brennan, 1994; Meijnders, Midden, & Wilke, 2001; Slater, Karan, Rouner, & Walters, 2002). There are limits to this effect, as when chronic issue-relevant fear reduces message processing (Jepson & Chaiken, 1990). But the common finding that greater fear arousal is associated with greater message processing gives reason to suspect that loss-framed appeals--conceptually akin to the fear-induction component of a fear appeal--might correspondingly evoke greater message engagement than will gain-framed appeals.
The second reason to suspect that loss-flamed appeals will be more engaging is the phenomenon of negativity bias, that is, heightened impact of and sensitivity to negative information (for a review, see Cacioppo, Gardner, & Berntson, 1997). …