Identification as a Mechanism of Narrative Persuasion Revised

[cayala15 (Chris Ayala)]

Speaker: So-eun Park

Affiliation: University of Haifa; University of Verona
Title: Identification as a mechanism of narrative persuasion revised.

Abstract (long version below): Experiencing narratives often involves seeing the story world according to the narrator’s lead, which makes the audience’s experience vary depending on which narration is followed. A study by De Graaf et al. (2012) used a change of narrative perspective in order to manipulate identification, and to test whether perspective leads to different attitudes regarding the given narrative topic. The current paper replicated the previous study with detailed perspective manipulation, using the segment from the story “The Fifth Child” by Doris Lessing (1988).

Long abstract

Stories are pervasive in virtually all aspects of human culture, from those told around campfires in caves millennia ago to the virtual reality games today. The importance of stories in shaping people’s views resulted in a great number of studies showing that narratives engage people's attention and emotions, and impact people's beliefs (​​Appel & Richter, 2007) and attitudes (Oschatz & Marker, 2020).

Previous studies found associations between identification and story-consistent attitudes. De Graaf et al. (2012) used a change of perspective in order to manipulate identification (Cohen, 2001), and test whether perspective leads to different attitudes regarding a given narrative topic. They found that perspective indeed led to identification, and to attitudes consistent with the story version they read. It is noteworthy that these results were found using two texts which differed in contents, topics, and style in which they were written, indicating their robustness.
The main purpose of the current study is to similarly explore the effects of narrative perspective. The main research question is, ‘To what extent may story perspectives lead to narrative persuasion?’ This idea has been tested before as described above, but the current research was intended to investigate and replicate it with improvements. Firstly, with shortcomings of De Graaf et al. (2012) in mind, the present story was designed to balance the two opposing views within the narrative. Secondly, the present study improved on external validity by employing a story based on the existing literature, “The Fifth Child” by Doris Lessing (1988) instead of experiment-written reading material. Lastly, a new operational definition of perspective was formulated that would lead to more sensitive manipulation.
Two different manipulations of narrative perspectives were used in this study; linguistic and psychological. Linguistic perspective manipulation (LPM) merely changes the narrative from a first-person to a third-person point of view, and vice-versa. If the story was written from a third-person author’s point of view, the linguistic manipulation is accomplished by simply changing ‘he/she’ (or characters’ names) to ‘I’ - ‘I’ being the story narrator. Psychological perspective manipulation (PPM) further includes the perspectivizing characters’ inner thoughts and feelings. PPM is a higher level of manipulation than LPM, as PPM can fill the gaps of the story that are typically left out in LPM.
Based on extant theory and research it was expected that 1) readers will identify more with the perspectivizing character and 2) agree with them more, and 3) that identification will mediate the relationship between perspective and attitudes. Furthermore, it was expected that 4) stories in the first person will engender stronger transportation and 5) that transportation will also mediate the relationship between perspective and attitudes.
To test these hypotheses, a segment from the story “The Fifth Child’’ by Doris Lessing (1988) was used. It involves a family’s difficult life choice; David and Harriet’s plan of having a big family went according to plan until their fifth child (Ben) was born. Ben’s wild behaviors and threatening character made everyone in the family miserable. Harriet thinks Ben should stay with the family, while David thinks Ben staying in a special institution is better for everyone, including their four other children. The research subjects were 287 adults (+18) (50.52% female, 49.48% male) with a minimum of high-school-level reading skills in Hebrew. Their mean age was 44.52 years (SD = 15.97), ranging from 18 to 85. The mean education level was 2.61 (SD = 1.00), meaning that the average participants obtained a Bachelor’s degree.
The study was conducted online. The study employed a between-subject design with five conditions; each participant was randomly assigned to one of five groups and each group read a different version of the same story (i.e. PPM Harriet, LPM Harriet, Neutral, LPM David, PPM David). The questionnaire consisted of two parts; pre-exposure and post-exposure. The pre-exposure questionnaire contained general questions about the participants (age, gender, etc.) and a measure of trait empathy. The post-exposure questionnaire included measures of identification, transportation, attitudes, social-distance, and a reactance scale.
Preliminary results revealed that the perspective manipulation did not work as expected. Nonetheless, several noteworthy findings emerged. Firstly, identification levels with both protagonists were significantly correlated with each other, as well as with transportation, suggesting that readers found both characters relatable and saw their virtue beyond the characters’ argument. However, transportation was more strongly correlated to identification with Harriet than with David. Secondly, trait empathy was more significantly associated with Harriet’s scales (i.e. identification - Harriet, social distance - Harriet, attitude slider). Thirdly, participants who were more transported into the narrative also indicated higher psychological resistance after reading it. Lastly, there was an unexpected negative correlation between Perspective Level (0 = Neutral, 1 = Linguistic, 2 = Psychological) and identification with Harriet; this indicates that identification with Harriet is associated with lower levels of perspective manipulation.