Reward, autonomy & voluntary persistence
Simulated results for 100 Polish participants in the timed wall-sit study. Performance rewards lift effort while present, then undermine perceived autonomy, intrinsic motivation, and post-withdrawal free-choice persistence - buffered by exercise identity.
Reward group holds longer while the incentive is present.
After reward withdrawal, voluntary persistence drops for the reward group.
Intention to keep exercising is lower in the reward group.
Group means across measures
Reward vs Control · IMI subscales (1-5) and persistence (s)
Serial mediation pathway
Reward → ↓autonomy → ↓intrinsic motivation → ↓free-choice
The indirect effect runs through both mediators in sequence: assigning a reward lowers perceived autonomy, which lowers intrinsic motivation, which in turn lowers how long participants choose to persist once the reward is gone. The direct reward → free-choice path is small and not significant.
Moderation by exercise identity
Free-choice persistence vs exercise identity, by group
Both groups persist longer as exercise identity rises, but the reward group's slope is steeper: a strong exercise identity buffers the undermining effect of reward (reward r = 0.62, control r = 0.39). At low identity the gap between groups is widest.
Participant data
100 participants · click a header to sort
| P001 | Reward | 79 | 47 | 2.7 | 3.0 | 3.2 |
| P002 | Control | 50 | 53 | 3.8 | 3.3 | 2.6 |
| P003 | Reward | 62 | 34 | 2.0 | 2.3 | 1.9 |
| P004 | Control | 55 | 66 | 3.6 | 3.4 | 2.7 |
| P005 | Reward | 103 | 18 | 2.9 | 2.3 | 3.3 |
| P006 | Control | 69 | 54 | 4.2 | 4.4 | 3.1 |
| P007 | Reward | 99 | 24 | 2.6 | 3.3 | 3.4 |
| P008 | Control | 66 | 28 | 3.5 | 3.1 | 1.7 |
| P009 | Reward | 75 | 37 | 2.7 | 1.6 | 4.0 |
| P010 | Control | 48 | 51 | 3.6 | 4.0 | 1.9 |