REF-629 · Teixeira, Carraça, Markland & Silva, 2012 · International Journal of Behavioral Nutrition and Physical Activity, · Paper · PubMed
Self-determination theory & exercise: across 66 studies, autonomous motivation (identified + intrinsic) consistently predicts exercise
Identified regulation for starting, intrinsic motivation for keeping it up
Key Findings
66 empirical studies (experimental, cross-sectional, prospective) with actual or self-reported exercise/physical activity, including attendance, as the outcome; SDT-based interventions were also reviewed.
Consistent support for a positive relation between more autonomous forms of motivation and exercise, with a trend for identified regulation to predict initial/short-term adoption more strongly than intrinsic motivation, and intrinsic motivation to be more predictive of long-term adherence.
Competence satisfaction and more intrinsic motives consistently predicted participation across samples and settings.
Mixed evidence for other motive types (health/fitness, body-related) and for the nature and consequences of introjected regulation (guilt/approval-driven) — the base does not claim introjection "drives short-term activity"; the review calls it mixed.
Most studies are descriptive (non-experimental), but similar results appear across cross-sectional, prospective and experimental designs.
Not from this paper (related work; checked against its own abstract): Ntoumanis 2021 meta-analysis of 73 experimental SDT-based health interventions (DOI 10.1080/17437199.2020.1718529) — small-to-medium changes in SDT constructs and health behaviours at the end of the intervention and at follow-up; increases in need support and autonomous motivation (not controlled motivation) went with positive behaviour change; effects modest and heterogeneous.
Practical Application
Across 66 studies, autonomous motivation predicts exercise, with identified regulation ("this matters to me") mattering most for getting started and intrinsic motivation (enjoyment) for long-term adherence. Coach toward both: tie each block to the athlete's own valued goal, and protect enjoyment in the sessions themselves rather than leaning on guilt over misses.
Be autonomy-supportive: offer choices, give the "why", acknowledge their perspective, minimise controlling language — the interventions that raised need support and autonomous motivation were the ones that changed behaviour (Ntoumanis 2021, above).
Feed competence (completable progressions, mastery feedback), the need whose satisfaction most consistently predicted participation; relatedness (group sessions, coach responsiveness) is part of the SDT model but the review's strongest consistent signal is competence and intrinsic motives.
Limitations
Systematic review without meta-analysis; mostly descriptive studies; intervention effects small-to-medium; general-population samples (Tier 1–2), so magnitudes in committed athletes are untested.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A: systematic review without pooling, and the majority of included studies are descriptive (non-experimental)
Content correction 2026-09-27
"Introjected/external regulation can drive short-term activity but poorly predicts maintenance" -> the abstract reports mixed evidence on introjected regulation; the adoption-vs-maintenance split it does report is identified (adoption) vs intrinsic (long-term), now stated.
"Autonomy, competence, relatedness … positively associate" -> the abstract's consistent finding is competence satisfaction and intrinsic motives; relatedness is not singled out.
Ntoumanis 2021 "small-but-significant" -> small-to-medium and modest/heterogeneous, 73 studies, per its abstract. Level A -> A- (SR without pooling, mostly non-experimental studies).