Effects of Contingent and Non-Contingent Sound on Video Game Performance

John Baxa, Siu-Lan Tan and Matthew Spackman

Extended Abstract

As the video game industry has advanced, sound has come to play an increasingly integral role in games. The player can now take an active role with sound, using it as an aural guide to learn more about the virtual environment. The present study aims to investigate exactly what effects different levels of interaction with sound have on players' performance and experience of gaming in The Legend of Zelda: Twilight Princess. Whether sound was contingent or not with players' actions and onscreen events was also of interest. We also wanted to examine the new playing experience offered by the new Nintendo Wii Console, and how its unique interface would impact game play. We predicted that players would perform better and experience higher levels of immersion in conditions in which sound is closer to normal playing conditions.

The study employed a within-subjects design. Twenty-three undergraduate males who had experience with Role-Playing Games and the Wii Console participated. All participants attended five one-hour individual sessions. The first of the five sessions was an orientation meeting. The next four sessions were held consecutively and run under the following experimental conditions in a randomized order: Full Sound, Half-Sound (Wiimote sound only), No Sound, and Non-Contingent Music (unrelated background music played on a boombox).

During each session, researchers recorded the participants' performance as they played and administered a series of rating scales to assess quality of experience after participants had finished playing. A tournament structure was employed, serving as an incentive to encourage participants to complete all five sessions and to play as well as they would in real-world settings.

A series of one-way repeated measures analyses of variance were conducted. Surprisingly, no significant relationships were observed between the levels of sound and a majority of measures for performance or quality of experience (including immersion, telepresence, effectance, flow, among others). However, significant differences were found between sound condition and participants' use of "continues," which were used when participants ran out of life. Follow-up repeated measures ANOVAs that took participants' use of continues into account revealed significant differences across the experimental conditions for the number of completed tasks and length of play before using a "continue." In general, the means showed that higher levels of sound enhanced play performance. Surprisingly, the means also revealed that participants performed best in the Non-Contingent Music condition. A trend we observed was that Full Sound enhanced the performance of the top third of the participants more than that of the other participants, suggesting that the highest-scoring players may attend more closely to sound cues than other players. The wide variances in rating scale responses suggest that players are not a homogenous group, and may respond quite differently to the same playing conditions. However, it appears that players with gaming strategies that incorporate sound tend to play better, and that their audiovisual strategies may be a key factor to improving gaming performance.