Can Simulated Partners Boost Workout Effort in Long-Term Exercise?

Abstract

We tested whether exercising with a stronger simulated (i.e., software-generated) partner leads to greater work effort compared to exercising alone, to help those seeking to maintain or improve fitness levels with long-term high-intensity training, but who find it necessary or practical to exercise in social isolation. Forty-one middle-aged adults, who participated in at least 30 minutes of vigorous exercise 3x·wk-1, trained on a cycle ergometer 6 days per week for 24 weeks in an alternating regimen of moderate-intensity 30-minute continuous and 3 types of high-intensity interval sessions (8 × 30-second sprints, 6 × 2-minute ladders, and 4 × 4-minute intervals). They were assigned either no partner (control), an always superior partner, or a not always superior partner. Participants varied cycle power output to increase or decrease session intensity during the repeated moderate-intensity sessions (30-minute continuous) and 1 of the 3 high-intensity sessions (4 × 4-minute intervals). Changes in intensity were used as a measure of effort motivation over time. Nested multilevel models of effort trajectory were developed and alpha was set to 0.05. For continuous and interval sessions, effort trajectory was positive and significant for those with an always superior partner, but not significantly different from control. Within interval sessions, those with an always superior partner significantly increased effort in the fourth interval compared to control (p = 0.02). Exercising with an always superior partner leads to greater work efforts during the hardest interval training compared to exercising alone.