Audience recall of issues and image in Congressional debates.

• Author: Hullett, Craig R.

His rhetoric is as empty as a walnut that a squirrel has condemned. Response of Audience Member to 5th District Congressional Debate, Winston-Salem, NC

A focal point of the research and corresponding argument surrounding political debates concerns what the voters learn from the viewing experience. In academic circles and among practitioners, the controversy regarding the worth of political debates remains intense. Although large numbers of voters watch this increasingly common campaign event, the value and nature of citizen learning remain uncertain(1). A general consensus seems to have emerged among academics that voters "learn" from viewing political debates by attending to issue messages from the candidates and elaborating on their images of the candidates. Many existent models, however, seek to demonstrate that voters remember either issue or image information. A few theorists acknowledge that voters may integrate issue and image judgments, but little has been done to discern the nature of combined information acquisition.

Unfortunately, most of the literature examining political debates regards image-based information with disdain. Authors generally hold that issues are the "right-stuff" of decision making, the pertinent reasons associated with proper democratic participation, whereas image is devalued as extraneous character judgments (e.g., Jamieson & Birdsell, 1988; Zhu, Milavsky, & Biswas, 1994). Some inconclusive research suggests that there may be differences in receptivity to candidates' statements according to voter allegiance, but to date little research has tested whether image perceptions indicate a lack of knowledge about the candidates' statements. Also, although voters' perceptions of the candidates are often reliably sorted into the image/issue dichotomy, there is a shortage of inquiry regarding whether these categories are as clearly opposed in voters' minds as they are represented in academic discussion. In this study, we sought to discover: 1) whether voters' party affiliations were related to their overall learning from the candidates; 2) whether voters' tendencies toward image or issue perceptions were related to their overall learning from the candidates; and 3) how image and issue are interrelated in the voters' minds.

Learning

Some argue that learning from debates reflects the value that the voters place on the issues discussed. Judd and Kulik (1980) claim that people, in general, are more likely to recall arguments they have heard when their views about those arguments are highly polarized toward either agreement or disagreement with the issues. This heightened learning stems from the fact that people are typically more receptive to information they perceive as important (Judd & Kulik, 1980).

Current research conflicts concerning whether viewers of debates are entirely open to the information presented by both candidates (Hellweg, Pfau, & Brydon, 1992). Carter (1962) found that those who watched the Kennedy-Nixon debates similarly recalled the arguments presented by both their preferred and the opposing candidate. The only significant differences in recall of the two candidates' arguments occurred when the viewers either discounted the effectiveness of both candidates' arguments or were inattentive to the debates (Carter, 1962). Hagner and Rieselbach (1978) reported that some viewers of the Carter-Ford debates were receptive to both candidates' arguments; twenty percent of the viewers converted their allegiance from one candidate to the other. Those who converted, however, typically had weaker voting convictions than those whose votes followed their previous preferences. Others have found evidence of selective and greater recall for preferred candidate's arguments (Bothwell & Brigham, 1993). Still, other contextual variables (e.g., how informed viewers are) have been advanced to explain audience reception to debate information (Hellweg et al., 1992). Overall, these findings are uncertain as to whether or not the viewers of debates are more receptive to their preferred candidate's arguments. We sought to determine whether or not the viewers' preference for a candidate affects their reception of the statements made by both candidates.

RQ 1. Are the viewers of a Congressional election debate more or less receptive of opposition versus preferred candidate messages?

Image/Issue Distinctions

Political researchers are not only concerned about if and when voters acquire information from candidate debates, they are also interested in the substance of what voters learn. The most prevalent concerns expressed in political communication literature focus on voters' perceptions of issue-stances and images of candidates. Much of the argument centers on the conception of issue knowledge as authentic and image perceptions as diversions (Converse, 1964). Lubell (1962), examining the Kennedy and Nixon debates, illustrates a preference for what voters should learn from political debates; he argued that the debates lacked sufficient issue distinctions for the voters to make proper voting decisions. His fear was that the voters were being tricked by the theatrics of image rather than informed by the substance of issues. Lubell (1962) argued that the more rational approach entails focusing upon issue differences; image distinctions only serve to cloud the proper decision-making processes of the voters.

Part of the responsibility for voters' reliance on image impressions is placed on the media. Media insistence on pinning down winners and losers, amplifying gaffs, and revealing character traits (faults) induce voters to frame their understanding via image distinctions (Diamond & Fiery, 1986). Meadow and Beek (1978), for example, conducted a comparative study involving the '60 and '76 Presidential debates that revealed that approximately 95 percent of debate content for each election was devoted to issue statements. Yet, in comparing the frequency of issues in the debate to the media discussion of the debates, Meadow and Beek (1978) found that the media were not accurate in their issue reporting and seemed to choose, instead, to focus on matters related to image.

Voters' reliance on image-related information does not reside solely in adopting media interpretations, but may reflect individuals' natural preference for image-oriented cognitions. Even while Lubell (1962) was criticizing the debate's lack of issue distinctions, much of his data implied that voters' perceptions of image were a key outcome of the debates. There is evidence that the debates do increase voters' cognitions of issues (Bishop et al., 1978; Drew & Weaver, 1991), but that this effect is not necessarily enduring (Miller & MacKuen, 1979). More recent work on political cognition clearly suggests that voters-sophisticated and uninitiated, involved and uninvolved - rely on issue and image-based information for making voting decisions (Miller, Wattenberg, & Malanchuk, 1985, 1986; also see Lau & Sears, 1986; Lodge & McGraw, 1995). These image-oriented cognitions, many argue, are inherent in candidate evaluation and serve as the natural interpretive schema for voters (see Popkin, 1991; Page & Shapiro, 1992). This research suggests that while issues are often perceived by the voters, the primary and enduring perceptions give emphasis to the candidate's image (Lanoue & Schrott, 1991).

Taken as a whole, then, the literature seems to indicate that either image perceptions act as a confound to accurate reception of the candidates' messages, or that image perceptions are a naturally co-occurring cognitive process to the reception of the candidates' messages. Those taking the former stance have been able to point to the potential cultivation effects of the media's focus on image information. Hence, an immediate audience of a debate may serve as a way of controlling for this possible effect.

RQ 2. Does message reception correspond with viewers' reliance on issue/image information?

Considering the possibility of differences in viewers' message reception between their preferred and opposed candidate's statements, a third question regarding tendencies toward image and issue perceptions was posed:

RQ 2.1 Do the viewers of a Congressional election recall more image or issue items for the opposition rather than for their preferred candidate?

Methodological Concerns

The final issue discussed in terms of debate effects involves whether or not research accurately reflects voter perceptions. Notwithstanding endeavors to move beyond a strict issue/image dichotomy when investigating voters' debate perceptions, research generally continues to impose external categorizations of issue and image on debate content and voters' cognitions. An innovative study examining voters' evaluations of candidates found voters generated a majority of image-based evaluations (vs. issue-based) even though the stimuli were political spots that focused exclusively on issues (Johnston, 1986; 1989). As with most research in the debate arena, this project imposed an a priori classification to order the subjects' thoughts.

The question is not only whether coding criteria like Johnston's (1989) are reliable classifications of what image and issue perceptions are; the question is whether these distinctions are consistent with how the voters make sense of political events. Louden (1994) argued that the separation of image and issue cognitions is largely an artificial one. He offered instead that these dichotomies are often convenient for research and criticism, but are strained in understanding what count as legitimate reasons when framing a voting decision. This difficulty in establishing the primacy of issue information may be a result of the fact that the process of reification of political understanding requires that images, not issues, are the social reality (Louden, 1990). Some argue that one needs primarily to discover the idiosyncratic image criteria that determine...