Effects of immersive stories on prosocial attitudes and willingness to help: testing psychological mechanisms

ABSTRACT

In recent years, the journalism and marketing industry has seen an increasing use of immersive stories, in the format of 360° videos and virtual reality. However, the impact of immersive stories on individuals’ beliefs, attitudes, and behaviors is understudied. Drawing upon literature in narrative persuasion and immersive media, this study examined the persuasive effects and underlying psychological mechanisms of immersive stories on prosocial attitudes and behavioral willingness to help. Findings from a laboratory experiment (N = 216) indicated that stories presented in immersive virtual (vs. traditional mediated) environments led to a higher level of spatial and social presence, which intensified users’ transportation and identification. The enhanced transportation led to less counterarguing and then promoted prosocial attitudes. These findings have important theoretical contributions to the study of persuasion in immersive virtual environments.

Acknowledgment

This work is part of the author’s dissertation project. The author would like to thank Drs. Xiaoli Nan, Robert Feldman, Dale J. Hample, Anita Atwell Seate, and Leah Waks for their suggestions and support. The author would also like to thank Dr. Amy Nathanson and two anonymous reviewers for their insightful comments and guidance throughout the review process.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. Participants also watched another 360° video about driving under the influence (DUI) and completed another set of questionnaire. Two videos were randomly played for each participant. The DUI case was dropped from the analysis because the pre-existing and post-test attitudes were virtually identical, making this key variable a constant instead of a variable.

2. See Appendix for a full report of the measures.

3. The anchors of the response metric were expanded to detect greater variability among the extreme values (Pelham & Blanton, 2012).

4. $x^2$ was computed based on the formula provided by Satorra and Bentler (2010).

5. The coefficient of Mardia’s kurtosis was 15.66, suggesting that the assumption of multivariate normality was violated (Bentler, 2004).

6. Gender had a significant effect on spatial presence and willingness to help, with females (vs. males) experienced greater spatial presence and behavioral willingness. Social desirability predicted greater identification and less favorable attitudes. Baseline attitudes had a positive effect on transportation, favorable attitudes, and behavioral willingness. Cybersickness had a negative effect on transportation. We also ran the same model without covariates. The majority of the results were similar to the model with covariates reported in the text. Yet, when no covariates were included, condition had a significant negative effect on identification, b = − .16, p = .03. In addition, the specific indirect effect of immersion on behavioral willingness through spatial presence and transportation was significant, b = .04, p = .04, 95% CI [.002, .075].

7. All parameters reported are standardized parameters under the STDYX output.

8. RMSEA = .144, 90% CI [.094, .199], SRMR = .040, CFI = 0.952, and χ² (5) = 27.405, p < .001.