The growing popularity of television has changed the way electoral campaigns and candidates are fundamentally portrayed. Personalization, negativism, deauthentification, and more interpretative coverage are all consequences that emerged with the rise of television (Reinemann & Wilke, 2007). "In the era of television politics, there is a growing trend of style over substance, personality over issues, and emotion over information," (Coleman & Banning, 2006, p. 313). In the United States, the average length of verbal quotations of political candidates in the press, and the sound bites of their statements in classical television news has decreased significantly since the 1960s (Adatto, 1990; Patterson, 1993). Similar results exist for Germany (Wilke & Reinemann, 2000). This is an issue in many respects.
First of all, it is important for a democracy that "... a substantial part of political media content is based on political figures because this is the only way the voters can get to know what they really have to say" (Patterson, 1995, p. 330). Furthermore, deauthentification is also seen as an important reason for negative attitudes towards politicians (Donsbach & Jandura, 2003). In that respect, modern political television might have led to negative outcomes like voter decline because of its focus on interpretive coverage. On the other hand, modern television formats also provide new opportunities for authentic political communication. Especially Live TV settings like political talk shows, a common format all over the world since the 1970s, seem to be useful platforms for politicians to directly address voters. During the 1992 election campaign, presidential candidates Bill Clinton, George Bush, and Ross Perot appeared more than 90 times in shows like CBS-Morning, Good Morning America, The Oprah Winfrey Show, The Arsenio Hall Show and Larry King Live! (Diamond, McKay, & Silverman, 1993). That was the birth of the so-called "talk show campaign" (Cavanaugh, 1995, p. 158) because the candidates not only saved money when addressing the voters via free media, but also managed to bypass the evaluative tenor of the national press corps (Bucy & Newhagen, 1999). Some even say that Bill Clinton turned his stalled campaign around with his legendary appearance on the Arsenio Hall Show when he played his saxophone to a cheering audience (Newman, 1994). In the 2000 election, major party presidential candidates again addressed their voters on the Oprah Winfrey, Rosie O'Donnell, and Regis Philbin shows. Nielsen ratings indicate that political TV talk show audiences tuned in to candidate appearances in large numbers. For instance, 8.7 million households watched AI Gore's September 11,2000 appearance on the season premier of The Oprah Winfrey Show, well above the program's average of 7.5 million households during the prior (1998-1999) season, and up 27% from Oprah's 1999-2000 premier episode. George W. Bush's appearance on the program 8 days later earned even higher ratings (Baum, 2005). Similar figures can be observed for political talk shows in Germany, where up to 6 million households watch one of the two major shows every week. (Zubayr & Gerhard, 2007). In election campaigns, the performance of a candidate in one of these shows might be critical regarding voting decisions (Kepplinger, Brosius, & Dahlem, 1994). Hence, it is of particular interest for political communication practitioners to understand the mechanism of person perception via television.
Person Perception via Television
Television focuses on the visual aspects of political communication (Maurer & Kepplinger, 2003). Political talk shows often use close-ups of the speaker, or nonverbal reaction shots of the TV host, or focus on the studio audience, or others, to illustrate emotions or interpretations. Editors or producers decide what is being shown (Kepplinger, 1980). Sometimes, it seems to be more important how politicians appear, what they wear, or how they behave nonverbally more than what they talk about. It is believed that these visual aspects play a major role in a person's perception via television. In general, it is supposed that audiences associate politicians' nonverbal behavior with their attitudes (Mehrabian & Ferris, 1967), or lean on nonverbal cues in person perception (DePaulo & Friedman, 1997). More recent research indicates that public perception of politicians' character traits and personalities correlate with exposure to the media portrayals of their nonverbal communication (Coleman & Banning, 2006). This is because human ability to process information in general is limited (e.g., Brosius, 1995; Graber, 1984; Lippmann, 1922; Shoemaker & Reese, 1991). Therefore, people tend to rely on superficial cues or stereotypes to reduce the cognitive effort of forming impressions of other people (Conover, 1981; Granberg, Kasmer & Nanneman, 1988). The idea that political cues guide the cognitive process of perception and judgment received empirical support by the research of Kahneman, Slovic and Tversky (1982). The authors conducted a series of experiments to analyze "heuristic information processing." Heuristics are strategies to reduce the cognitive effort in building judgments about social issues. One heuristic is "availability heuristic" (Tversky & Kahnemann, 1973, p. 207) where information that is easily available at the time of judgment building becomes the dominant basis of judgment. Information about a politician's nonverbal behavior is readily available to recipients of TV talk shows as they watch the screen. Thus, heuristic information processing, and accordingly, "availability heuristic" explains the eminent role of nonverbal behavior in conjunction with person perception very well.
The model of "heuristic information processing" could also be applied to support this study's assumption that nonverbal reaction shots also affect impressions of a politician's image. However, more specific approaches are available. According to Festinger (1954), individuals learn about and assess themselves by comparison with other people (Social Comparison Theory). While this original work had a rather restricted focus on the "choice" of comparison targets and on the concept of "contrast" to others, social comparison theory has undergone various transitions and reformulations over the past decades (see, e.g., Suls & Wheeler, 2000). The shift of scientific attention towards the "effects" of social comparison that originated from the work of Wills (1981) is especially remarkable. Subsequent studies show that social comparison with other people might not only induce contrast with the comparison target, but also assimilation (e.g., Brown, Novick, & Kelley; 1992; Stapel & Koomen, 2000). This means that when individuals do not know how to behave or what to believe, they often copy other people or assimilate group opinions. Since individuals assume that the people know what they are doing, they serve as a prototype for how to behave. As they care what others think of them, it provides a safe course of action because they cannot be criticized for their actions or opinions if they adopted the majority position. Early social psychological research reveals that individuals tend to lean stronger towards assimilation in situations that are ambiguous (Asch, 1955; Kelley, 1952). The authors argue that in the case of person perception via television, many attributes are ambiguous because there is no objective criterion by which, for example, the competence of a speaker (politician) can be reliably judged. Thus, viewers of political talk shows might compare their own impressions of a speaker (politician) with impressions of the studio audience or the TV host that are conveyed through nonverbal reaction shots and adopt the perceived opinion.
Studies of Nonverbal Behavior
When talking about the effects of nonverbal behavior on person perception, it is important to first point out what kind of nonverbal behavior is mentioned. Research has identified three different semantic dimensions of nonverbal behavior (Mehrabian, 2007, pp. 14-15). These are (1) "positiveness dimension," (2) "responsiveness dimension," and (3) "potency or status dimension." Concrete behaviors within these dimensions emerge in several nonverbal communication channels. Burgoon, Birk, and Pfau (1990) identify "vocalic," "kinesic," and "proxemic" channels of nonverbal communication. According to this, politicians can use their voice, gestures, or body movements to be perceived as "friendly," "interested," or "superior," etc. For instance, one can demonstrate sympathy towards another person on the "positiveness dimension" by smiling or touching that person. Secondly "responsiveness dimension" is related to nonverbal communication that demonstrates the other's salience for him or her. Nodding or keeping eye contact for example is concrete behavior within this dimension. Finally, nonverbal behavior of "potency or status dimension" is used to demonstrate social control. People expand themselves and take up a lot of space by using gestures or body movements in this dimension. Ostertag (1991) finds out by doing a content analysis of German politicians' TV appearances, that nonverbal signals of "potency or status dimension" (active vs. passive nonverbal behavior styles) are most common in political reality. Therefore, this dimension is of particular interest in the present study.
In a review of empirical studies concerning effects of nonverbal behavior on person perception in general, Burgoon et al., (1990) point out that nonverbal behavior styles related to Mehrabian's "potency or status dimension" are supposed to influence only certain traits such as "competence," "dynamism," and "composure." According to that, perceptions of a persons' "character"...