| Few people engage in recommended amounts of physical activity, and as such, it is important methods are sought out that could help people improve their motivation to sustain exercise. One potential way to do this is to leverage the power of group dynamics, specifically, the Kohler effect (Stroebe, Diehl, & Abakoumkin, 1996). When two people of moderately unequal abilities are partnered together on a conjunctive task, the weaker group member tends to perform better, and this performance gain is known as the Kohler effect (Kerr & Hertel, 2011).;A growing body of research supports the idea that the Kohler effect can be combined with exergames to produce improvements in the length of time people hold plank exercises (e.g., Feltz, Kerr, & Irwin, 2011). However, the research thus far has only made use of pre-recorded confederates purported to be live human partners. While this approach has been successful, it may not be practical on a large scale. Therefore, the purpose of this dissertation is to explore whether or not the Kohler effect can be achieved with different types of partners.;One approach is to use a computer-generated partner. Some research shows that people will team up with computers, suggesting that there is the potential for these types of interactions to improve motivation and performance (Nass, Fogg, & Moon, 1996). Another potential partner is an imagined partner. Research has shown that using associative imagery (imagery focused on the task) and dissociative imagery (imagery focused on something unrelated to the task) helps extend performance on muscular endurance tasks, suggesting that combining imagery and the Kohler effect could be an effective match (Razon et al., 2010).;To test this, participants (220 undergraduate students) were randomly assigned to one of six experimental conditions (virtually-real partner, altered virtually-real partner, software-generated partner, imagined partner, associative imagery, dissociative imagery) or to a no-partner, no-imagery control condition. Participants performed two blocks of plank exercises, holding each exercise for as long as possible. The first block was a baseline measure, performed individually and without imagery. For the experimental conditions, the second block was performed either with a partner or with the use of imagery. Measures of exercise duration, perceived exertion, self-efficacy beliefs, enjoyment, intention to exercise the following day, and impressions of one's partner and group were collected.;Results from the main analysis indicated that participants in the experimental groups generally held plank exercises for longer during the second block compared to the control participants, supporting the idea that computer-generated partners, imagined partners, and imagery can be used to increase the time people hold plank exercises. Ancillary analyses indicated that participants in the partner conditions perceived themselves as working harder during the second block, while participants in the imagery and control conditions did not. There were no differences in self-efficacy beliefs, enjoyment, and intention to exercise the following day.;These findings are consistent with past research (Feltz et al., 2011) and suggest that incorporating computer-generated partners into exergames may boost how long people play these games. The use of imagined partners and imagery may also be effective methods to improve performance during traditional exercise. As a muscular endurance task was used, future researchers could use a similar paradigm to test these results with aerobic exercise, which would have a greater potential to positively influence health. |
