The Influence of Audio-Only Character Narration on Character and Narrative Engagement

Abstract

Using the Showtime crime drama Dexter (2006–2013), this posttest-only experimental design revealed that narration by the title character (i.e., Dexter Morgan) was significantly associated with more parasocial relationship (PSR) interaction than no narration. The influence of narration was examined while controlling for biological sex, evaluation of the narrating character's motive, evaluation of the narrating character, and previous exposure to the stimulus program. Favorable evaluations of the narrating character predicted more identification and PSR interaction with him. Women were more transported into the stimulus program than men. Findings are discussed in terms of the growing use of character narration in narrative television, Auter's (1992) experimental evaluation of audiovisual narration in a situation comedy, and the limitations of using a morally ambiguous narrator.

Keywords:

• Identification
• Media Effects
• Narrative Television
• Narrative Transportation
• Parasocial Interaction
• Soliloquy

Notes

Copies of the measures are available from the first author.

Original alpha reliability of Green and Brock's (2000) 11-item transportation scale was .54. To improve the overall reliability, four items were deleted from the final scale: “While watching this episode of Dexter, activity going on around me was on my mind,” “I could easily picture myself inside the events of this episode of Dexter,” “I was mentally involved in this episode of Dexter while watching it,” and “I wanted to learn how this episode of Dexter would end.”

Tests of narration's influence on EPSI, in H1, and narrative transportation, in H2, were not significant but might have been frustrated by a lack of statistical power. A post hoc power analysis, conducted with G*Power (Faul, Erdfelder, Lang, & Buchner, 2007), revealed that this study's experiment was underpowered for narrative's influence on both EPSI (1 − β = .46) and narrative transportation (1 − β = .38). With a larger sample size, future researchers could cite these empirical findings to support a case for narration's influence on EPSI and narrative transportation. Another method of increasing the power of a similar experiment would be to use a more likeable narrating protagonist. The lack of findings for identification may have been frustrated by Dexter Morgan's willing embrace of his psychopathology. This interpretation is consistent with this study's finding that less-favorable Morgan evaluations predicted significantly less identification with Morgan. It is also consistent with Tal-Or and Cohen's (2010) finding that subjects were more willing to identify with a conventionally moral protagonist than a conventionally immoral protagonist. Even the findings for transportation might have been frustrated by Morgan's psychopathology. Tal-Or and Cohen (2010) found that audiences were less willing to transport themselves into a movie segment in which the protagonist was anticipated to act immorally. Thus, future research should seek to replicate this study with a more conventional narrator.

Additional information

Notes on contributors

Shane M. Semmler

Shane M. Semmler (PhD, University of Oklahoma, 2010) is an associate professor in the Department of Communication Studies at the University of South Dakota.

Travis Loof

Travis Loof (MA, University of South Dakota, 2014) is a PhD student in the College of Media and Communication at Texas Tech University.

Collin Berke

Collin Berke (MA, University of South Dakota, 2014) is a PhD student in the College of Media and Communication at Texas Tech University.