The role of autonomy support and integration in predicting and changing behaviour: theoretical and practical perspectives on self-determination theory

This thesis reports six studies adopting a self-determination theory (SDT; Deci & Ryan, 1985) approach to understanding motivation and behaviour in health and social contexts. The research focuses on the internalisation and integration of goals and motives extrinsic to the individual and the effects of internalisation on psychological and behavioural outcomes. Two studies also explore the role of social agents in facilitating internalisation through provision of autonomy support. \n\nThe research addresses gaps in the SDT literature and contributes to the advancement of theory and practice. A meta-analysis of effects of autonomy support on health-related psychological and behavioural outcomes (Chapter 2), based on the methods of Hunter and Schmidt (1994), indicated the significance and consistency of adaptive effects of autonomy support across the literature. A path analysis was also used to test a modified representation of Williams et al.’s (2006) SDT process model of health-related behaviour. Results supported the motivational sequence postulated within the model, as the effect of autonomy support on behaviour was mediated by need satisfaction and autonomous motivation. \n\nThe studies reported in Chapters 3 and 4 make a novel contribution to the SDT health literature by employing measures of chronically-accessible physical activity outcomes and motives to represent spontaneous motivational influences on behaviour. The results presented in Chapter 3 indicated that chronically-accessible appearance-related outcomes are associated with controlling forms of motivation, while the findings reported in Chapter 4 showed that planning-based strategies to maintain physical activity under situations of success and failure in goal striving are differentially effective for chronically autonomous and controlled individuals. These studies also offer guidance for health practitioners in promoting physical activity, by highlighting the potentially maladaptive effects of appearance-related goals and the importance of tailoring planning-based interventions to individuals’ chronically-accessible motives.\n\nThe study reported in Chapter 5 substantiated a core theoretical assumption of SDT by providing empirical support for people’s inclination to distinguish between intrinsic and extrinsic goals. Adopting methods from the literature on memory and attitudes, cluster analysis was performed on participants’ self-generated and recalled physical activity goal data to determine the presence of clustering by goal type. Although findings supported individuals’ ability to distinguish intrinsic and extrinsic goals at some level of representation, participants were not able to reliably code their goals at an explicit level. It was therefore concluded that differentiation between goal types may not occur consciously. \n\nThe scale-development study in Chapter 6 also supported a key tenet of SDT in establishing construct, nomological, and predictive validity of a scale measuring integrated regulation for physical activity. The factorial validity of the scale, developed through an extensive literature search, expert ratings, and confirmatory factor analyses, was supported in both a high and a lower-active sample. Consistent with predictions, latent means analysis indicated the high active sample reported significantly greater integrated regulation. The scale provides a valid and reliable tool that may be used to evaluate the process of integration following autonomy-supportive interventions in health-related contexts.\n\nFinally, Chapter 7 details the development of a brief autonomy-supportive intervention and observational checklist system for ensuring fidelity to protocol that can be modified for use in a number of contexts requiring behaviour change. The intervention was implemented in a higher education setting over the duration of a single course module and significantly increased two autonomy-supportive teaching behaviours in postgraduate tutors. However, the intervention did not significantly increase the perceived autonomy support, self-determination, or coursework grades of the experimental tutors’ students relative to the control condition, although there was a trend towards a trend towards an interaction between time and experimental condition for level of self-determination towards studying. While students in the experimental group reported an increase in self-determination over time, students within the control condition reported a decrease in self-determination between the first and second, and first and third waves of data collection. \n\nThe thesis concludes with a general discussion of findings and directions for future research and practice.