Are You What Your Avatar Eats? Health-Behavior Effects of Avatar-Manifested Self-Concept

Abstract

Drawing on the self-concept activation and goal-priming account of the priming effect, this study examined how self-concept—i.e., ideal self, ought self, and actual self—can be harnessed as a model for avatar customization in digital games to promote healthy-eating behavior. Female participants (N = 133) customized an avatar in a digital game to reflect either the ideal, ought, or actual self. Participants then selected food items for their avatar within the digital game as well as food items for themselves to eat afterward. Results suggest that for participants using an ought-self avatar, the extent to which they were conscious of their health was positively related to healthier food choice both within and after playing the game. No such effect emerged for participants who used an ideal- or actual-self avatar, indicating that participants formed the goal of being healthy only with regard to the ought self. This study demonstrates that avatar customization in a digital game can serve a regulatory function by representing individuals’ duties and responsibilities, thus, causing them to adopt such attributes manifested in their avatar during and after the game.

ACKNOWLEDGMENTS

We would like to thank the following individuals for supporting these research activities: Daniel Allen, Xinning Guo, Jackson Hopcroft, Annette Yunwoo Kim, William Wisz, Joey Sprow, Justin Tokarski, and Kimberly Vroman.

FUNDING

This research was partially supported by the AT&T endowment to MSU’s Department of Media & Information

Notes

¹. The vegetables included the following: cauliflower, broccoli, celery, tomato, and carrot. The fruits included the following: pineapple, grape, cantaloupe, watermelon, honeydew, and banana. Items in the vegetable and fruit category varied slightly depending on their availability at the grocery store. The salty and sweet snacks included the following options: Doritos, Cheetos, Ritz Bits, Cheez-it, SunChips, mini Chips Ahoy, Oreo Mini, Nutter Butter Bites.

². On average, participants consumed over 80% of the food items they had chosen.

³. These value assignments are somewhat arbitrary, but because we use ordinal regression in our analysis, the specific values of the categories are less relevant than their ordering. We should note that we examined different value assignments (e.g., red = 0, yellow = 1, green = 3) to confirm this assumption and this did not change the results.