Attitudes Toward Software-Generated Exercise Partners During High-Intensity Training

Stephen Samendinger, Christopher Hill, Brian Winn, Alison Ede, Norbert Kerr, James Pivarnik, Lori Ploutz-Snyder and Deborah Feltz

Extended Abstract

Background: The most recent Physical Activity Guidelines for Americans (PAG; Services, 2008) recommend at least 150-min/week of moderate-intensity, or 75-min/week of vigorous-intensity aerobic physical activity, or an equivalent combination of moderate- and vigorous-intensity aerobic activity for “substantial benefits to health.” One alternative to satisfy all or part of these recommendations can be in the form of high-intensity interval training (HIT), as there is compelling evidence that HIT is effective in improving health for non-diseased adults and adults suffering from cardiometabolic disease (Garber et al., 2011; Gibala, Little, MacDonald, & Hawley, 2012; Haskell et al., 2007; Hwang, Wu, & Chou, 2011). HIT exercise protocols are defined as alternating brief bursts of vigorous activity with periods of rest or active-rest (low intensity movement). Coordinating exercise, particularly high-intensity, with a partner can pose challenges due to the intense nature of the exercise, differing exercise goals and abilities, and conflicting schedules. Yet, evidence suggests exercising with a companion may improve adherence to physical activity (Carron, Hausenblas, & Mack, 1996; Sherwood & Jeffery, 2000). Software-generated exercise partners (SGPs) offer one potential solution for those who prefer exercise partners, as they can be available at any time and can be tailored to meet the requirements of the exercise protocol and the preferences of the human partner. Group dynamic principles, personal characteristics, and other motivational manipulations may also be utilized with SGPs to enhance physical activity performance. SGPs embedded in an interactive computer-based video game have been utilized successfully with 1-session physical activity experiments (Feltz, Forlenza, Winn, & Kerr, 2014) and with a 3-week aerobic exercise protocol on a stationary cycle (Max et al., 2016). Yet, little is known regarding SGP use with HIT protocols and if characteristics of an SGP are problematic for the exerciser. For example, if the human exerciser does not like the SGP or finds the character’s appearance bothersome, this incompatibility with the exercise partner may potentially interfere with motivation to initiate, maintain, or persist with high-intensity training. Therefore, attitudes about the SGP are important to game developers as they strive to create software-generated characters that are well received by the game user and considered to be an appropriate exercise companion. This study examined the use of SGPs in an exergame during a high-intensity exercise protocol over 6 days.

Methods: The current study enrolled 86 active community adults (Mage = 44.9 yr, SD = 9.5 yr) in a 6-day aerobic exercise regimen on a cycle ergometer. Sixty-four participants were assigned an SGP (see Figure 1) while 22 served as control and did not have an exercise partner. Daily sessions alternated between continuous 30-min sessions, at 75% of predicted? heart-rate maximum, and high-intensity interval sessions. Interval sessions consisted of four 4-min intervals at 90% heart-rate maximum, with 3-min periods of active rest in between. On Days 5 and 6 of the protocol (a 30-min and 4-min session day respectively), participants worked out with a same-gender SGP who also completed the protocol. The SGP was the same for both days and all participants. Prior to the Day 5 workout, the SGP and participant introduced themselves in a brief exchange of personal information: name, hometown, and some fact about them. After the introduction, participants could visualize the SGPs riding with them during the exercise but did not interact further. Subjective ratings of attitudes toward the SGP were collected at the end of Day 6. Participants completed a 4-item scale in response to the prompt: “How well does each of these statements describe your feelings toward your partner?” (1=strongly disagree, 5=strongly agree). The four items included: I liked my partner; I would be glad to exercise with my partner again in the future; I felt comfortable with my partner; and, I would like to get to know my partner better. Alternative Godspeed Indices (AGS) were gathered on Day 5 to assess whether the SGP was perceived to be uncanny or disconcerting (Ho & MacDorman, 2010; Mori, 1970). The 19-item semantic differential questionnaire contains 3 distinct sub-scales: humanness (e.g., 1=artificial vs. 5=natural), eeriness (e.g., 1=bland vs. 5=uncanny), and attractiveness (e.g., 1=repulsive vs. 5=agreeable) to capture participant impressions of anthropomorphic characters.

Results: The results indicated that there were no significant differences by gender for either attitudes toward the partner or the three AGS sub-scales so all data from both groups were combined. The 4-item Attitudes scale displayed adequate internal consistency (Cronbach’s alpha > 0.82). Participants’ attitudes toward their partners (M = 3.4, SD = .82) were significantly above the scale mid-point (3), indicating positive feelings for the exercise SGP, t60 = 3.77, P < .001. The AGS sub-scales also each displayed adequate internal consistency (Cronbach’s alpha: humanness > 0.88; eeriness > 0.79; attractiveness > 0.87). Indices for humanness (M = 2.8, SD = 0.91) and eeriness (M = 2.5, SD = 0.48) were negatively and significantly different from the scale mid-point (3), indicating that participants perceived the partner to not be human-like, but they also felt that the AGS was not eerie, t56 = -1.77, P = .08; t51 = -7.59, P < .000. The AGS index for attractiveness was positively and significantly different from the mid-point (M = 3.2, SD = 0.66), suggesting the participants perceived the SGP to be attractive, t53 = 2.5, P = .02.

Conclusions: Active video games, utilizing SGPs, may be an effective modality to aid in motivating adults participating in high-intensity training. However, just as with human exercise companions, attitudes toward one’s companion may affect participation in the activity. Therefore, SGP characteristics should be designed to avoid interfering with the participant-SGP relationship during this intense exercise. This study demonstrated that SGPs can be successfully implemented with HIT protocols in an active adult community sample. Participants liked their exercise partners and found them attractive, not eerie, while recognizing the SGP was not human-like.

Funding: This study was supported by the National Space Biomedical Research Institute, through NASA NCC-9-58.

Carron, A. V., Hausenblas, H. A., & Mack, D. (1996). Social influence and exercise: A meta-analysis. Journal of Sport and Exercise Psychology, 18, 1–16.

Feltz, D. L., Forlenza, S. T., Winn, B., & Kerr, N. L. (2014). Cyber Buddy Is Better than No Buddy: A Test of the Köhler Motivation Effect in Exergames. Games for Health Journal, 3(2), 98–105.

Garber, C. E., Blissmer, B., Deschenes, M. R., Franklin, B. A., Lamonte, M. J., Lee, I.-M., Swain, D. P. (2011). Quantity and Quality of Exercise for Developing and Maintaining Cardiorespiratory, Musculoskeletal, and Neuromotor Fitness in Apparently Healthy Adults: Guidance for Prescribing Exercise. Medicine & Science in Sports & Exercise, 43(7), 1334–1359.

Gibala, M. J., Little, J. P., MacDonald, M. J., & Hawley, J. A. (2012). Physiological adaptations to low?volume, high?intensity interval training in health and disease. The Journal of physiology, 590(5), 1077-1084.

Haskell, W. L., Lee, I. M., Pate, R. R., Powell, K. E., Blair, S. N., Franklin, B. A., ... & Bauman, A. (2007). Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation, 116(9), 1081.

Ho, C.-C., & MacDorman, K. F. (2010). Revisiting the uncanny valley theory: Developing and validating an alternative to the Godspeed indices. Computers in Human Behavior, 26(6), 1508–1518.

Hwang CL, Wu YT & Chou CH (2011). Effect of aerobic interval training on exercise capacity and metabolic risk factors in people with cardiometabolic disorders: a meta-analysis. J Cardiopulm Rehabil Prev 31, 378–385.

Irwin, B. C., Scorniaenchi, J., Kerr, N. L., Eisenmann, J. C., & Feltz, D. L. (2012). Aerobic exercise is promoted when individual performance affects the group: a test of the Kohler motivation gain effect. Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine, 44(2), 151–159.

Max, E. J., Samendinger, S., Winn, B., Kerr, N. L., Pfeiffer, K. A., & Feltz, D. L. (2016). Enhancing aerobic exercise with a novel virtual exercise buddy based on the Köhler effect. Games for Health Journal, 5(4), 1-6. DOI: 10.1089/g4h.2016.0018

Mori, M. (1970). The uncanny valley. Energy, 7(4), 33–35.

Services, H. (2008). Two Thousand Eight Physical Activity Guidelines for Americans: Be Active, Healthy, and Happy. Government Printing Office. Retrieved from

Sherwood, N. E., & Jeffery, R. W. (2000). The behavioral determinants of exercise: implications for physical activity interventions. Annual review of nutrition, 20(1), 21-44.