行为技能训练在促进身体活动中的挑战与机遇
Challenges and opportunities in behavioural skills training for physical activity
行为改变技术(BCT)作为干预措施的关键成分,其效果依赖于受试者主动执行与维持。Knittle等人基于现有分类体系,编纂了包含123项可自我执行的行为改变技术(SE-BCTs)汇编,如目标设定与自我监控。尽管相关研究结果令人鼓舞,但目前尚无专门针对SE-BCT干预的系统综述,且大多数干预措施未明确评估技术执行保真度。
Main Text
A key challenge in the field of exercise medicine and health is how to support long-term behaviour change, as many important health benefits, such as reduced risks of cardiovascular disease, cancer, type 2 diabetes and other chronic conditions, are contingent on physical activity (PA) sustained over time1. A broad consensus in this field is that PA interventions need to be multi-level, targeting individual, interpersonal, community and policy factors2. Nevertheless, the challenges of implementing and evaluating such interventions are considerable (e.g. need for expertise from multiple disciplines, time and coordination demands, extensive funding and unpredictability3). In this Perspective, we argue that interventions aiming to empower individuals to manage their self-motivation and behaviour have an important role to play within the PA and public health agenda. Such interventions can strengthen complex multi-level approaches by delivering better outcomes at the individual level and, potentially, by forging stronger synergies across levels (e.g. via technological solutions that can personalise key intervention components and are scalable to the population level). In our Perspective, we focus on the opportunities and challenges involved in training individuals in behavioural skills, in particular, self-enactable behaviour change techniques (SE-BCTs). We discuss issues of personalisation, clustering and sequencing, as well as how to test their potential effectiveness.
The potential of using SE-BCTs
Behaviour change techniques (BCTs) have been defined as the active ingredients of interventions designed to change behaviour4. Examples include goal setting, planning, demonstrating the target behaviour and providing feedback on behaviour. More recently, it has been argued5 that for BCTs to have their intended effects, recipients need to play an active role in enacting them, for example, by successfully adjusting their goals over time. Having an active role is important because the effectiveness of many BCTs may depend on whether recipients understand, practise, adapt and sustain their use6. With this perspective in mind, Knittle and colleagues have published a compendium of self-enactable BCTs (SE-BCTs), defined as actions that individuals must themselves undertake to change their behaviour. The SE-BCT compendium strengthens attention to enactment fidelity and supports the development of scalable digital approaches that aim to build individuals’ behavioural self-management skills.
Many SE-BCTs are already embedded within existing BCT taxonomies and ontologies (e.g. setting goals, self-monitoring) alongside non-SE-BCTs (e.g. demonstration of target behaviour by the intervention deliverer). The compendium5 features 123 SE-BCTs for changing and self-managing one’s motivation and behaviour and is based on the synthesis of six existing BCT taxonomies. Each SE-BCT is defined in simple language and accompanied by an instructive example to facilitate self-enactment. For instance, technique #98, reflect on desire to perform behaviour, asks individuals to rate how much they want to perform a healthy behaviour, such as becoming more physically active, using a 0–10 scale. If they give any rating above 0, they are encouraged to reflect on why they chose that number rather than 0. As another example, technique #43, comparative imagining of future outcomes, asks individuals to compare the likely consequences of changing a target behaviour with the likely consequences of not changing it.
Similar approaches have been advocated in other areas of psychology and health care. For example, boosting, defined as fostering people’s competencies to make their own choices, has been proposed as a complementary strategy to environmental interventions7. Likewise, the World Health Organization8 has developed guidelines for self-care interventions that view individuals as active decision-makers in their own health rather than passive recipients of health services. It is important to note here that self-care should not be understood as shifting responsibility for health entirely onto individuals. Self-care and self-empowerment approaches should not replace necessary interpersonal, environmental and structural supports.
At present, there are no reviews of SE-BCT-based interventions. Reviews of the broader BCT literature have identified effective BCTs based on coding of intervention manuals, including SE-BCTs, for PA interventions across different population groups9,10. Although these findings are encouraging, drawing conclusions based solely on coding intervention manuals for BCT presence has limitations. For instance, many interventions do not report how recipients enact the targeted BCTs; if individuals do not fully understand the skills they are taught or do not practise them consistently, BCTs may fail to produce their intended effects, even when fidelity of intervention delivery is high6. Without fidelity checks of BCT enactment, conclusions about BCT efficacy can be problematic. Notably, very few interventions have been explicitly designed to promote or assess the fidelity of BCT enactment6. This commentary is inspired by an ongoing project in which we have used the SE-BCT compendium to develop a digital platform designed to train physically inactive individuals to use SE-BCTs to become active. We outline key challenges encountered in this process and propose potential solutions.
Challenges and potential solutions in training for SE-BCT use
The first challenge is pragmatic: the SE-BCT compendium includes 123 techniques, which is a very large number to train individuals in. Moreover, the description of each technique in the compendium is quite succinct, providing only a brief 3–4 line overview of what the technique entails alongside an example. As such, it is likely too brief to function as a training tool capable of engaging participants over time and guiding them to implement each technique successfully. Additional resources could therefore be considered, integrating up-to-date evidence on behaviour change, explaining the relevance of SE-BCTs for participants’ self-management, and offering practical tips on how to adapt and implement these techniques within an individual’s own context and challenges. Within a digital training platform, such resources could include summary sheets, podcasts, quizzes, gamification elements and periodic refreshers. However, integrating these resources has implications for how many SE-BCTs can realistically be included in a digitally delivered training programme.
Several strategies can be used to reduce the number of techniques from 123 to a more manageable set. For example, some techniques (e.g. self-penalties, imaginary punishment) can be problematic because they undermine individuals’ basic psychological needs and autonomous motivation11,12. Others (e.g. prayer) may lack acceptability among certain population groups, while some techniques may be less feasible to implement in practice (e.g. seeking biofeedback). Even after excluding such SE-BCTs, the remaining number may still be too large for inclusion in a training programme. As with any PA intervention, training in SE-BCTs should be co-designed with participants to enhance acceptability and engagement. Co-design workshops can therefore be used to further prioritise and reduce the number of techniques. Finally, given that many techniques overlap conceptually (e.g. self-monitoring of behaviour and self-monitoring of outcomes of behaviour), a practical approach is to group related techniques and teach them together. Although no formal SE-BCT clusters currently exist based on the recent compendium5, researchers can draw inspiration from existing BCT taxonomies13 and the new BCT ontology14.
A second consideration is that researchers should be mindful that existing clusters of behaviour change techniques have typically been developed based on expert surveys identifying techniques that target similar ‘mechanisms of change’ (p. 85)13, or through methods such as hierarchical cluster analysis and ontological structuring. While such clusters can be useful for illustrating conceptual similarities among techniques, they have some limitations. For instance, many BCTs operate through multiple mechanisms of action (depending on how they are delivered and on which other BCTs they are combined with), and the strength of evidence for particular BCT–mechanism of action links varies14,15. Further, intervention designers must consider how (SE-)BCTs can be combined in ways that are both conceptually meaningful and practically effective. For example, a review16 of the efficacy of different BCT combinations suggested that providing feedback on performance may be counterproductive when clear instruction on how to perform the behaviour is not provided. Similarly, a trial on habit formation for PA17 found that the techniques of commitment, prompts and cues were more effective when delivered together than in isolation. Unfortunately, evidence regarding which combinations of BCTs, and SE-BCTs more specifically, are most effective for health behaviours, including PA, remains limited, largely due to a range of methodological challenges discussed in detail elsewhere18.
Taking the first two points into account, we recommend that, when developing training programmes for SE-BCTs, researchers should consider not only existing clusters in the wider BCT literature based on conceptual similarities among techniques, but also evidence regarding which combinations of techniques show evidence of effectiveness. Both considerations have informed our current research project, alongside feedback from co-design workshops, in which we have developed and tested customised clusters of SE-BCTs. We aim to present these clusters once our trial commences.
A third challenge we are addressing in our project, which builds on the previous two, is that different SE-BCTs may be more (or less) effective at different stages of the behaviour change journey. Existing BCT clusters do not typically account for this issue, as they focus primarily on content rather than the timing of BCT implementation. Several options can be considered when sequencing techniques. For example, researchers may draw on existing reviews of BCTs and theoretical models of behaviour change. At earlier stages of behaviour change, a systematic review19 of interventions designed to increase motivation for PA identified techniques associated with increased intentions and with progression from earlier to later stages of change (e.g. self-monitoring of behaviour and behavioural goal setting). Another interesting review20 of 100 theories that contain explicit propositions about behavioural maintenance identified five cross-cutting themes (i.e. habit formation, support for autonomous motivation and exercise identity formation, coping with lapses and seeking social support). These themes can be used to identify and sequence relevant SE-BCTs.
The vast majority of such theories and models are not PA-specific, and coverage of them is beyond the scope of this Perspective. An important exception is the Multi-Process Action Control (M-PAC)21 approach, which has been developed specifically for PA. M-PAC proposes that reflective processes (e.g. affective judgements, perceived capability, instrumental attitudes) are most salient during the intention formation stage. For these processes, M-PAC highlights the relevance of BCTs such as seeking or receiving social support, behavioural practice and providing information about social and environmental consequences. For individuals in the action control phase, regulatory processes (behavioural, affective and cognitive) become more salient, and techniques such as action planning, problem solving and cognitive reframing are likely to be useful. Finally, during the action maintenance phase, reflexive processes (e.g. habit and identity) predominate, and BCTs such as prompts and cues and identity-based techniques should be emphasised.
In our project, we are developing a delivery sequence of SE-BCTs based on the evidence outlined above and guidance from the SE-BCT compendium regarding which techniques serve as prerequisites for others (e.g. behavioural goal setting is a prerequisite for action planning and forming a behavioural contract), both within and, in some cases, across different stages of behaviour change. Our aim is to develop clusters of SE-BCTs, referred to as learning modules within our digital platform, based on these considerations. Figure 1 illustrates how selected techniques from the SE-BCT compendium (identified by their reference numbers; e.g. #6, outcome goal setting) can be clustered into higher-order groups (e.g. goal setting), sequenced within a given stage of change according to technique prerequisites (e.g. goal setting preceding action planning and behavioural contracting within the action control stage), and mapped onto distinct stages of change (intention formation, action control and maintenance). Completion of all modules within a given behaviour change stage, alongside attainment of a PA-related behavioural goal, unlocks access to modules for the subsequent stage, while modules from earlier stages remain available.
Despite the availability of relevant reviews and theoretical guidance, what remains lacking in the literature is the explicit and systematic testing of different temporal sequences of BCT delivery. To this end, we recommend that researchers consider N-of-1 designs27 for more idiographic approaches to SE-BCT sequencing. An N-of-1 trial is a within-person randomised experiment in which an individual is repeatedly exposed to one or several intervention conditions in varying or counterbalanced orders, using sustained treatment periods and washout phases, to estimate individual-level causal effects. N-of-1 trials are useful because a nomothetic ‘one-size-fits-all’ approach to (SE-)BCT selection and sequencing may be inadequate for capturing individuals’ idiosyncratic responses to specific techniques and their timing. This is partly because the appeal of SE-BCTs can vary across individuals, shaped by prior experience, intuitive prior use, beliefs about perceived effectiveness and perceived usefulness or burden. Thus, not all SE-BCTs need to be taken up by all participants within an intervention. Providing individuals with some choices of SE-BCTs can foster their sense of autonomy, which is an important consideration for increased engagement with digital health interventions28.
New technologies and digital solutions22 offer substantial opportunities to enhance both the scalability and personalisation of interventions designed to train SE-BCTs. Digital platforms make it possible to deliver SE-BCT training at scale while simultaneously adapting content to individual users’ needs, preferences and experiences. In particular, recent advances in artificial intelligence provide exciting avenues for improving how SE-BCT training is delivered and experienced. Large Language Models (LLMs) can be used to generate diverse forms of training content, including text-based materials, quizzes, podcasts and videos, potentially reducing development costs and enabling rapid iteration. However, such content must be carefully reviewed and quality-assured by experts in behavioural science to ensure theoretical fidelity and prevent misinformation. A further opportunity afforded by technology is the possibility of using contextual information to guide not only when SE-BCTs are delivered, but also which techniques are recommended at a given moment23. With appropriate consent and privacy safeguards, digital platforms may draw on contextual data such as time of day, location, recent activity patterns, self-reported availability, mood, fatigue, or prior engagement with specific techniques to tailor SE-BCT prompts. For example, if a participant has repeatedly struggled to be active after work, reports low energy in the evening, and is currently at home, the platform might prompt problem-solving or coping planning by asking the person to identify a smaller, more feasible activity option and a specific barrier-response plan. In contrast, if the participant is near a park, has previously responded well to prompts and cues, and indicates that they have 20 min available, the platform might recommend action control or a context-specific prompt to initiate a short walk. In this way, contextual tailoring can help move SE-BCT training beyond static personalisation based only on baseline preferences, towards adaptive support that is sensitive to varying contexts.
Natural Language Processing (NLP) techniques can be leveraged to analyse participants’ written input, such as reflections on their experiences with specific SE-BCTs, allowing researchers and training systems to better understand engagement, perceived usefulness and challenges in enactment. Importantly, NLP can also support participant-specific content generation, enabling reminders, prompts, or feedback that directly reference users’ own goals, plans, or sources of social support, thereby increasing relevance and personal meaning. Beyond content generation and tailoring, gamification24 represents another promising strategy for enhancing engagement with SE-BCT training. By incorporating game-like elements, digital platforms may strengthen cognitive, behavioural and affective engagement, which are all important for sustained engagement with training. Nevertheless, we acknowledge that gamification features do not appeal to everyone, and engagement should not be narrowly interpreted as engagement with the platform per se, but rather as engagement with practising, reflecting on and adapting, when needed, the use of relevant techniques25. From an applied perspective, engagement should be treated as a dynamic process that may influence intervention effects over time. With N-of-1 designs, researchers could examine longitudinal associations between micro-engagement with a digital platform, participants’ understanding and enactment of SE-BCTs, and macro-engagement with the target behaviour, namely PA.
Reflections and conclusions
Long-term PA maintenance remains difficult to achieve. This Perspective argues that digitally training individuals to apply SE-BCTs can complement multi-level PA interventions. Such training aligns with calls for empowering individual self-care and does not replace the need for multi-level interventions. While advocating for our Perspective, we acknowledge that although individual-level interventions can support self-regulation and agency, their effectiveness is likely to depend on the broader social and structural contexts in which PA occurs. Opportunities to enact SE-BCTs may be shaped by social determinants such as income, time constraints, caregiving responsibilities, neighbourhood safety, access to facilities and supportive social environments. Thus, individual empowerment should not be understood as a substitute for addressing structural inequalities, but as one component of broader ecological approaches to PA promotion.
Building on the recent SE-BCT compendium and available evidence from earlier BCT classifications, this Perspective highlights four linked challenges and possible solutions. First, the sheer number (123) of available SE-BCTs requires pragmatic selection based on feasibility, empirical evidence and co-design input. Second, because many techniques overlap conceptually, grouping SE-BCTs into teachable clusters can reduce burden and enhance learnability. Different clusters are available in the wider BCT literature, but when used for intervention purposes, SE-BCT clusters should also be informed by evidence of effectiveness. This leads to the third point, namely, that the field needs stronger guidance on which combinations of techniques work best, given that interactions among BCTs may be additive, synergistic, or counterproductive.
Fourth, the literature rarely tests temporal sequencing in the delivery of (clusters of) BCTs, despite theory and evidence suggesting that different techniques may be most useful at different stages of behaviour change (e.g. initiation vs. maintenance). This Perspective, therefore, recommends idiographic N-of-1 approaches to advance personalised and stage-sensitive sequencing. We also discuss how technological solutions can drive the development of adaptive and personalised SE-BCT training platforms to support ongoing skill development and long-term behaviour change. It is important to clarify that we do not advocate for a purely instructional/cognitive approach to behaviour change. For instance, the compendium includes techniques that help individuals find task enjoyment, deal with pressure and monitor the emotional consequences of the target behaviour. We also acknowledge that PA should not be understood as merely the result of combining behaviour change techniques. People’s engagement with physical activity is also shaped by how they see themselves, the meanings they attach to movement, their relationships with others and the bodily sensations and emotions that accompany activity26. However, we do not see SE-BCT training as contradicting this view, as a few techniques focus on aspects of the human experience of PA related to identity, social relationships and bodily sensations.
Overall, digital SE-BCT training is positioned as a promising way to strengthen PA promotion by improving enactment, scalability and personalisation of various techniques for behaviour change, but future research should systematically evaluate clustering, sequencing and adaptation of SE-BCTs over time. We hope to report on the testing of some of these avenues for future research once our trial is completed. Finally, although our Perspective focused on PA, the issues we discuss regarding grouping, sequencing and personalisation of SE-BCTs are also pertinent to other health behaviours, such as alcohol consumption, screen time, smoking and sleep.
Data availability
No datasets were generated or analysed during the current study.
References
Piercy, K. L. et al. The physical activity guidelines for Americans. JAMA 320, 2020–2028 (2018).
Oh, A. Y. et al. Advancing multi-level health communication research: a Delphi study on barriers and opportunities. Transl. Behav. Med. 12, 1133–1145 (2022).
Sallis, J. F. Needs and challenges related to multilevel interventions: physical activity examples. Health Educ. Behav. 45, 661–667 (2018).
Michie, S. & Johnston, M. Theories and techniques of behaviour change: developing a cumulative science of behaviour change. Health Psychol. Rev. 6, 1–6 (2012).
Knittle, K. et al. The compendium of self-enactable techniques to change and self-manage motivation and behaviour v.1.0. Nat. Hum. Behav. 4, 215–223 (2020).
Hankonen, N. Participants’ enactment of behavior change techniques: a call for increased focus on what people do to manage their motivation and behavior. Health Psychol. Rev. 15, 185–194 (2021).
Hertwig, R. & Grüne-Yanoff, T. Nudging and boosting: steering or empowering good decisions. Perspect. Psychol. Sci. 12, 973–986 (2017).
World Health Organization. WHO Guideline on Self-Care Interventions for Health and Well-Being, 2022 Revision. https://www.who.int/publications/i/item/9789240052192 (2022).
Al-Walah, M. A., Donnelly, M., Cunningham, C. & Heron, N. Which behaviour change techniques are associated with interventions that increase physical activity in pre-school children? A systematic review. BMC Public Health 23, 2013 (2023).
Carraça, E. et al. Effective behavior change techniques to promote physical activity in adults with overweight or obesity: a systematic review and meta-analysis. Obes. Rev. 22, e13258 (2021).
Knittle, K., Fidrich, C. & Hankonen, N. Self-enactable techniques to influence basic psychological needs and regulatory styles within self-determination theory: an expert opinion study. Acta Psychol. 240, 104017 (2023).
Ntoumanis, N. & Moller, A. C. Self-determination theory informed research for promoting physical activity: contributions, debates, and future directions. Psychol. Sport Exerc. 80, 102879 (2025).
Michie, S. et al. The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: building an international consensus for the reporting of behavior change interventions. Ann. Behav. Med. 46, 81–95 (2013).
Marques, M. M. et al. The behaviour change technique ontology: transforming the behaviour change technique taxonomy v1. Wellcome Open Res. 8, 308 (2024).
Johnston, M. et al. Development of an online tool for linking behavior change techniques and mechanisms of action based on triangulation of findings from literature synthesis and expert consensus. Transl. Behav. Med. 11, 1049–1065 (2020).
Dusseldorp, E., van Genugten, L., van Buuren, S., Verheijden, M. W. & van Empelen, P. Combinations of techniques that effectively change health behavior: evidence from Meta-CART analysis. Health Psychol. 33, 1530–1540 (2014).
Baretta, D., Gillmann, N., Edgren, R. & Inauen, J. HabitWalk: a micro-randomized trial to understand and promote habit formation in physical activity. Appl. Psychol. Health Well Being 17, e12605 (2025).
Michie, S., West, R., Sheals, K. & Godinho, C. A. Evaluating the effectiveness of behavior change techniques in health-related behavior: a scoping review of methods used. Transl. Behav. Med. 8, 212–224 (2018).
Knittle, K. et al. How can interventions increase motivation for physical activity? A systematic review and meta-analysis. Health Psychol. Rev. 12, 211–230 (2018).
Kwasnicka, D., Dombrowski, S. U., White, M. & Sniehotta, F. Theoretical explanations for maintenance of behaviour change: a systematic review of behaviour theories. Health Psychol. Rev. 10, 277–296 (2016).
Rhodes, R. E. & Brennan, C. Closing the gap between physical activity intentions and sustained behavior: an overview of evidence. Exerc. Sport Mov. 4, e00059 (2026).
Arigo, D., Jake-Schoffman, D. E. & Pagoto, S. L. The recent history and near future of digital health in the field of behavioral medicine: an update on progress from 2019 to 2024. J. Behav. Med. 48, 120–136 (2025).
Nahum-Shani, I., Dziak, J. J., Walton, M. A. & Dempsey, W. Hybrid experimental designs for intervention development: what, why, and how. Adv. Methods Pract. Psychol. Sci. 5, 25152459221114279 (2022).
Mazeas, A., Duclos, M., Pereira, B. & Chalabaev, A. Evaluating the effectiveness of gamification on physical activity: systematic review and meta-analysis of randomized controlled trials. J. Med. Internet Res. 24, e26779 (2022).
Mazéas, A. et al. Effect of a digital health physical activity program integrating gamification for obesity management in comparison with usual care: randomized controlled trial with an ideographic approach. J. Med. Internet Res 27, e78376 (2025).
Allender, S., Cowburn, G. & Foster, C. Understanding participation in sport and physical activity among children and adults: a review of qualitative studies. Health Educ. Res. 21, 826–835 (2006).
Shaffer, J. A., Kronish, I. M., Falzon, L., Cheung, Y. K. & Davidson, K. W. N-of-1 randomized intervention trials in health psychology: a systematic review and methodology critique. Ann. Behav. Med. 52, 731–742 (2018).
Smit, E. S., Zeidler, C., Resnicow, K. & de Vries, H. Identifying the most autonomy-supportive message frame in digital health communication: a 2x2 between-subjects experiment. J. Med. Internet Res. 21, e14074 (2019).
Acknowledgements
We acknowledge funding from TrygFonden (Grant ID 158431) for supporting our research on SE-BCTs.
Author information
Authors and Affiliations
Danish Centre for Motivation and Behaviour Science (DRIVEN), Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark
Nikos Ntoumanis & Alexandre Mazéas
School of Sport, Exercise, and Rehabilitation Sciences, University of Birmingham, Birmingham, UK
Nikos Ntoumanis
School of Education, Language and Psychology, York St. John University, York, UK
Nikos Ntoumanis
Authors
- Nikos Ntoumanis
- Alexandre Mazéas
Contributions
N.N. conceptualised the perspective and wrote the first draft. A.M. contributed to both conceptualisation and writing of the draft. Both authors reviewed and approved the final submission.
Corresponding author
Correspondence to Nikos Ntoumanis.
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Ntoumanis, N., Mazéas, A. Challenges and opportunities in behavioural skills training for physical activity. npj Exerc. Med. Health 1, 9 (2026). https://doi.org/10.1038/s44437-026-00016-z
Received: 20 January 2026
Accepted: 20 August 2026
Published: 05 October 2026
Version of record: 05 October 2026
DOI: https://doi.org/10.1038/s44437-026-00016-z
来源:Nature 健康护理主题 · nature.com