Models and Theories of Behaviour Change Robert West
Models and Theories of Behaviour Change Robert West University College London June 2014 1
Definitions • A scientific model is a representation of features of a system consisting of constructs, their characteristics and relationships between them – Used to • describe relevant characteristics of the system • make predictions • A theory is a model that purports to explain a set of phenomena – Used to • understand the phenomena • guide observation • develop interventions 2
Theory Quality Criteria • Scope – How far does theory explain what it sets out to? • Parsimony – Does theory have the minimum required number of elements? • Accuracy – How well does theory fit observable data? • Clarity – How easy is theory to understand apply? • Evaluability – To what extent can the propositions within theory be assessed for accuracy? 3
Examples of models of behaviour change • Cigarette consumption price elasticity 1 Δconsumption = 0. 4 x Δprice • First law of smoking cessation 2 cessation rate = proportion trying to stop x success rate 1 Chaloupka 2 West et al Tob Control 2002; 11: i 62 -i 72 et al Addiction (2015) 4
PRIME Theory (text) For a given level of opportunity and capability, changing responses (behaviour) requires changing the balance between competing momentary impulses and inhibitions • Impulses and inhibitions are influenced • – Directly • By stimuli operating on perceptual, memory and inference processes • By wants and needs – Indirectly • By plans – Directly • by stimuli operating on learned (habit) and unlearned (instinct) associations – Indirectly • by motives (wants and needs) • Motives are influenced • – Directly • By stimuli operating on learned and unlearned associations with positive and negative emotions and drives – Indirectly • By evaluations Evaluations are influenced Plans are influenced – Directly • By stimuli operating on perceptual, memory and inference processes • By evaluations West & Brown (2013) Theory of Addiction. Oxford: Wiley-Blackwell 5
PRIME Theory (graphic) External environment Internal environment Plans Evaluations Unidirectional influence Motives Impulses/ Inhibitions Responses Bidirectional influence 6
PRIME Theory (hypotheses) At every moment we act in pursuit of what we most want or need at that moment Prior intentions and evaluations can only influence behaviour if they are remembered and generate sufficiently strong motives (and thus impulses/inhibitions) to overcome others operating at the time Improving self-control requires improving the strength of plans and efficiency of connections with lower level sub-systems and/or weakening other influences 7
PRIME Theory (application) • Predicting relapse during an attempt to stop smoking • Relapse occurs when urge (impulse) exceeds resolve (desire for restraint and efficiency of inhibition) 15 Arbitrary units 13 11 9 7 Urge 5 Resolve 3 1 1 2 3 4 5 6 7 8 9 10 Days 8
PRIME Theory (application) • Urge to smoke is proportional to – Hypo-activity in the mesolimbic dopamine pathway – Strength of association between smoking triggers and impulse and motive to smoke – Degree of exposure to smoking triggers • Resolve is proportional to – Negative feelings about smoking – Positive feelings about not smoking – Capacity to inhibit responding 9
PRIME Theory (application) • Translation to dynamic systems model with time varying parameters and feedback loops Timms, Rivera, West (In preparation) 10
Some key messages • Models and theories can be useful at very different levels of specificity • Choosing an appropriate behaviour change model for intervention design depends on the levers of change one is able to pull and therefore the relevant constructs • Models and theories can be useful even when they are imprecise as long as they are evaluable • It is worth searching for precision 11
- Slides: 11