Media Effects Research Lab - Research Archive
Unpacking the Audiovisual Modality
Student Researcher(s)
Tyler Augst (Masters Candidate);
Eric Meczkowski (Masters Candidate);
Faculty Supervisor
Introduction
The use of exemplars in news media is extremely commonplace and these exemplars have been shown to influence audience perceptions of the news story and the actual event being covered. One understudied form of these exemplars is the inclusion of stock video footage behind a spoken television news report. Previous research on exemplars has been limited to investigating exemplars in only one modality or extreme multimodality, like the Internet. Using an experimental design, this study, using exemplars composed of a video footage and a spoken word narration (like that seen on the nightly news), examines the relationships between modality and valence of exemplar messages and the perceived vividness, perceived credibility, and attitudes regarding the issue.
Research Question
RQ: For U.S. citizens, controlling for socioeconomic status, education, and past experience with rurality, what are the relationships between visual and audio message valence, and perceived vividness, perceived credibility, and attitudes towards rurality?
Method
Amazon's Mechanical Turk was used to recruit a total of 217 participants for this research. A 2 x 2 experimental message design that manipulated the valence (positive or negative) of audio and video content was adopted. The valence of the congruent messages was unambiguous (positive visual and positive audio, or negative visual and negative audio), whereas the incongruent messages lacked this correspondence (negative visual and positive audio, or positive visual and negative audio). Participants were randomly assigned to one of these four conditions which served as the independent variables in this research. Then, participants were asked to report upon the vividness and credibility of the message, as well as their attitudes toward rurality.
Figure 1. 2x2 Experimental Design with Valenced Audio and Visual Content as Factors
Results
ANOVA analysis revealed a significant interaction effect between visual and audio valence content for both perceptions of message vividness and perceptions of message credibility. More specifically, when visual valence was negative and audio valence was positive, perceptions of vividness and credibility were low. However, when visual valence was positive and audio valence was negative, vividness and credibility were no different from the congruent conditions. Additionally, attitudes toward rurality were the result of audio valence. No main effect or interaction with visual valence was found. Moderated mediation analysis revealed that attitudes were mediated by credibility, which in turn was mediated by vividness.
Figure 2. Effects of Visual and Audio Valence on Perceived Vividness
Figure 3. Effects of Visual and Audio Valence on Perceived Credibility
Figure 4. Effects of Visual and Audio Valence on Attitudes toward Rurality
Figure 5. Moderated Mediation Analysis of Visual and Audio Valence on Vividness and Credibility
Figure 6. Moderated Mediation Analysis of Visual and Audio Valence on Credibility and Attitudes
Conclusions
One of the most interesting findings from this study was that the relationship between the valence of one modality in the message to the outcome variables depended on the valence of the other modality. This moderation between modalities implies that future research on exemplars should examine the effects of each modality individually with an eye towards how they may interact with one another. This also project succeeded in beginning to unpack the processing of exemplar messages in an attempt to further the understanding of why these types of messages have consistently been found to be more influential than base-rate information. In the initial formulation of exemplification theory, Zillmann argued that exemplars were more readily stored and made available to individuals because of the exemplar’s memorability or vividness. This vividness of exemplars was supported by the high overall perceived vividness among all conditions. The moderated mediation analysis presented above shows that the perceived vividness of the exemplars helps to explain variations in perceived message credibility, which in turn can explain variations in attitudes toward the issue at hand.
For more details regarding the study contact
Dr. S. Shyam Sundar by e-mail at sss12@psu.edu or by telephone at (814) 865-2173