Behaviour Change - Science and Technology Committee Contents


Different kinds of evidence

3.1.  Current understanding of how to change human behaviour is derived from the various sciences of human behaviour and from two overlapping types of research.[29] First, basic research, consisting of the development of theory describing the processes which shape behaviour and empirical, including experimental, tests of this theory. Secondly, applied research, which is the application of basic research to understanding how behaviour can be changed in everyday setting. When applied research is conducted using samples that are representative of the population to demonstrate effective behaviour change it is particularly relevant to policy makers. In this report, we refer to the latter form of research as "research at a population level". Of course, for evidence to be of most use it must have been evaluated rigorously and over the long-term. We discuss evaluation further in Chapter 7.

Understanding behaviour: basic research


3.2.  Basic research confirms that human behaviour is the product of a multitude of interrelated factors. This is true both of particular actions and also of patterns of behaviour over a lifetime. Given the complexity of factors underpinning behaviour, it is impossible to summarise concisely what is known about those factors and how they interact. Influences on behaviour can, however, be characterised broadly as comprising: genetics, individual thoughts and feelings, the physical environment, social interaction (with other individuals), social identity (interaction within and between groups), and the macro-social environment.

3.3.  We can also say that some actions are consciously planned, or deliberative, while others are governed by automatic, or non-deliberative, processes (the focus of "nudges"). For example, a decision to buy a new car will usually be made only after much conscious deliberation (coupled with unconscious motivations), but when a car is being driven down a familiar route the driver will be able to navigate without thinking about where they are going, so acting automatically. The distinction between deliberative and non-deliberative choices and actions are described in terms of dual process theories. Professor Theresa Marteau, Professor of Health Psychology, King's College London, provided an overview in her evidence to us:

    "We can understand people's behaviour as comprising the interaction between two systems. The first is a reflective system, whereby what we do is a result of goals that reflect our values and where we're aware of what we're doing. The other system, which actually accounts for much more of our behaviour, is an automatic system, whereby we're often not aware of the impulses that have generated our behaviour. There is an increasing recognition that both these systems are very important in explaining our behaviour. Often they work synergistically, so they work together well. Sometimes they work antagonistically. This is one of the reasons why, while many of us have very good intentions, we often find ourselves behaving in ways that go against our intentions."[30]

Some witnesses argued that public policy has placed too much emphasis on the reflective system or deliberative decision-making, leading to an assumption that behaviour change can only be achieved by appealing to knowledge and values and, as a result, underestimating the importance of the automatic or non-deliberative aspect of making choices.[31]

3.4.  Both deliberative and non-deliberative choices and actions can be affected by social factors (such as personal interaction and interaction within, and between, groups) and the large-scale social context (such as state of the economy). Behaviour is also influenced by the physical environment in which it takes place. The ready availability of cheap and unhealthy food, for example, makes it more likely that people will consume it. Similarly, if there are very busy roads and no cycling lanes, people are less likely to travel by bike. Professor Marteau acknowledged the contribution of behavioural economics in highlighting the contextual and automatic determinants of behaviour.[32] She observed that "... behavioural economists have been extremely successful ... in highlighting to policy makers the potential behaviour change gains from going beyond information-based campaigns, which rarely effect significant behavioural change, to alter 'choice architecture' with its potential to be far more effective".[33]


3.5.  Several witnesses identified a number of gaps in understanding about human behaviour. Examples given to us included a lack of understanding about aspects of the automatic system, particularly in relation to how emotional processes regulate everyday behaviour;[34] a lack of comparative research into the limits to the transferability of behaviour change interventions across cultural differences;[35] uncertainty about how genes interact with environmental and social factors to cause behaviour;[36] and, a lack of understanding about the effect of social dynamics on behaviour.[37] Other witnesses commented on the challenges involved in integrating the numerous theories of behaviour which were emerging from across the range of sciences of human behaviour. In this regard, Professor Michie, Professor of Health Psychology at University London, argued that, though there had been advances in multi-disciplinary working, more work needed to be done.[38]

Applied research at a population level

3.6.  Whilst theoretical understanding of behaviour change appears to be strong, several witnesses drew our attention to the comparative lack of research at a population level.[39] NICE, for example, commented in relation to public health interventions that:

    "The majority of experimental evidence about behaviour change relates to individual approaches, and comes largely from disciplines within psychology ... much of the evidence is limited and it is rare that evidence can be extrapolated or generalised from those interventions to the wider population with confidence and without caveats ... There is less experimental evidence about what works to influence behaviour when working with or at community or population levels."[40]

They further noted that there is "a marked lack of information about what works to change behaviour at policy level".[41]

3.7.  Richard Bartholomew, joint head of the Government Social Research service (GSR), said that though there were theories explaining why people behaved in certain ways there was a dearth of clear evidence about how to translate that into change.[42] The British Psychological Society (BPS) agreed to some extent, noting that further research was required "to develop cost-effective strategies that can be adopted and utilised in practice".[43] The Sustainable Development Commission said that there needed to be more "understanding of what interventions work best in practice".[44]

3.8.  Our impression that there is relatively little evidence of the effectiveness of particular behaviour change interventions at a population level has been reinforced by how few substantial responses we received following our request for examples of successful interventions. A number of witnesses also alluded to a lack of evidence about the cost-effectiveness of interventions[45] and to a disappointing lack of long-term data against which to judge the effectiveness of interventions over sustained periods.[46]

3.9.  Businesses, on the other hand, have demonstrated success at changing behaviour patterns on a large scale through measures like advertising and product promotion.[47] However, governments can face greater challenges than businesses in changing behaviour. Government may often wish to establish new behaviour patterns, such as getting people to take more exercise, or helping people to break ingrained habits, like smoking cigarettes. This is difficult to achieve. By contrast, businesses normally seek to sell people those things that they like and want.


3.10.  There is a lack of applied research at a population level to support specific interventions to change the behaviour of large groups of people (including a lack of evidence on cost-effectiveness and long-term impact). This is a barrier to the formulation of evidence-based policies to change behaviour. To address this problem, the Government will need both to evaluate their own behaviour change interventions rigorously and establish new evidence by commissioning and funding more applied behavioural research on this scale. Recommendations are made in Chapters 4 and 6 about how this can be achieved.

29   We recognise that the terminology used to distinguish between different types of research is the subject of debate and that distinctions between different categories of research are not clear cut. Identifying categories is, however, necessary for the purposes of discussion. Back

30   Q 331. Back

31   BC 71, BC 96. Back

32   BC 103, BC 110. Back

33   BC 110. Back

34   BC 54, BC 72, BC 96, BC 108. Back

35   BC 13. Back

36   BC 45, BC 108. Back

37   BC 54, BC 90. Back

38   Q 88. Back

39   BC 52, BC 58, BC 73, BC 83, BC 94, BC 105, BC 108, BC 110. Back

40   BC 52. Back

41   IbidBack

42   Q 25. Back

43   BC 105. Back

44   BC 83. Back

45   BC 42, BC 52, BC 67, BC 103, BC 108. Back

46   BC 52, BC 70, BC 105, BC 109 Back

47   BC 101. Back

previous page contents next page

© Parliamentary copyright 2011