You are here

Article is available in full to IFST members and subscribers.

Register on the FST Journal website for free

Click the button to register to FST Journal online for free and gain access to the latest news


If you are an IFST member, please login through the Members Area of the IFST website. 













Nudge, nudge, eat, drink...

Nudge IT

Gareth Leng and John Menzies of the University of Edinburgh explain how a large multinational, multidisciplinary research project (Nudge-IT) tackled the question of how we decide what to eat.

In the last thirty years, the prevalence of overweight and obesity across Europe has increased dramatically, particularly among children. The full consequences of this epidemic have yet to unfold, but include an expected increase in cardiovascular disease, hypertension, type 2 diabetes and diverse mental health conditions. These have huge social and economic costs: they affect people in the midst of their working lives. Stress, anxiety and depression, together with social stigma and bias in the workplace add to the pressures on families and employment and can enhance the vicious cycle of weight-gain through ‘comfort eating’.

Nudge-IT was a multidisciplinary, multinational project funded by the European Union to address the need for a better understanding of the determinants of food choice[1]. The project engaged experts in the neurobiology of motivational behaviour, the neuroscience of reward pathways, the neuroendocrinology of homeostatic regulation of appetite, experimental psychology, functional brain imaging, behavioural economics (the scientific study of decision making) and computational neurobiology.

There is a ‘disconnect’ between our mechanistic understanding and our ‘softer’ knowledge of consumer behaviour, which can make inferences about food choices incomplete and unsafe.

Policy recommendations

To produce policy recommendations on healthy eating that are likely to be effective, we must be able to make valid, non-trivial predictions about the likely consequences of any proposed interventions. In particular we need: confidence in the scientific evidence base, an adequate understanding of the health benefits that are likely to accrue from behavioural change, an understanding of the barriers to change – the reasons why interventions may fail to be effective, and an appreciation of the possible unintended consequences of change.

These are not trivial considerations – if an intervention is successful in getting people to eat less of some foods that we have reason to believe are ‘unhealthy’ when eaten in excess, what will they eat instead and what other consequences for their behaviour might we expect to follow?

For this, we need a better understanding of the determinants of food choice[2]. These obviously include dietary components (taste, texture and palatability), but also cultural and social pressures, cognitive factors (perceived stress, anxiety and depression) and familial, genetic and epigenetic influences. Our choices of what particular foods to eat, how much to eat and when are influenced by how foods are marketed and labelled and by economic factors. They reflect both habits and impulses, moderated, albeit imperfectly, by our often incomplete understanding of what constitutes ‘healthy eating’.

But they are also influenced by our individual physiological needs – needs which we are not necessarily consciously aware of but which nevertheless affect our behaviour.

Clearly there is a need for a strong evidence base to inform public policy on the issue of healthy eating. However, there is a ‘disconnect’ between our mechanistic understanding and our ‘softer’ knowledge of consumer behaviour, which can make inferences about food choices incomplete and unsafe. For a robust evidence-based policy, it is essential that these inferences are properly validated.

In general, to inform public health policy, scientific evidence needs to include three very different types of information. There is a need for strong evidence of association between, for example, a particular behaviour and a health outcome. But evidence of association does not in itself demonstrate any causal link between the behaviour and the health outcome. So there also needs to be evidence from interventions to show that changing a behaviour will produce the intended outcome.

But if the hoped for outcomes (on health risks) are very long term and only to be expected on a population level, it may be difficult to demonstrate a link.

Evidence from intervention studies is often weak; they are typically studied for a limited subset of the population under particular conditions and over a limited time period. Accordingly there is also a need for evidence of mechanism: a detailed understanding of the pathways and mechanisms by which a particular behaviour can result in a particular health outcome.

Such mechanistic understanding can give insights, for instance, into the particular groups in society that are likely to benefit from behavioural change, the kinds of intervention that are most likely to be effective, other ‘unintended’ consequences of behavioural change and the likely barriers to change.

These engage a very wide range of questions, including, for instance, how our brains recognise and respond to different food constituents, how our brains decide how much energy and other nutrients we will need for the day ahead and how early life experiences affect our behaviour as adults.


The focus of the Nudge-IT project was on tools that will lead to knowledge that translates into policy, and understanding how to make that translation effective. In particular, it set out to develop new tools to help understand:

a) the importance of early life experience in determining food choice,

b) habitual eating behaviour,

c) impulsive choice behaviour,

d) the importance of environmental context in decision-making processes.

The Nudge-IT consortium used a very wide range of approaches to study food choice behaviour in both people and animal models[2]. These included investigating the effects of hormonal signals on particular brain pathways (brain imaging), how dietary interventions in children affect their food preferences and how different food compositions affect signalling from the gut to the brain and influence hunger and satiety.

Here we focus on one example of how mechanistic studies in animals can give a deep insight into food choice and raise important questions.

Physiology of eating

Although it might seem obvious that changes in body weight reflect choices about what food we eat, how much we eat and how much we exercise, the long-term balance between energy intake and energy output is mainly determined by unconscious physiological systems.

Two hormones are extremely important in this: leptin, which is produced by adipocytes (the cells that store fat), and ghrelin, which is secreted from the empty stomach. Both of these hormones signal to the brain, as do many other hormones secreted from the gut, fat tissue and liver[3, 4].

We now have a good mechanistic understanding of how leptin, ghrelin and some of the other physiological signals from the gut and fat stores affect eating behaviour. We also have a growing understanding of exactly where and how they act in the brain and we are beginning to understand the mechanisms by which different food constituents affect hunger and satiety. These physiological pathways have been well conserved throughout mammalian evolution, so knowledge gained from animal models has proved generally to translate well into understanding of human physiology and behaviour.

For example, human brain imaging studies using positron emission tomography and functional magnetic resonance imaging show that these mechanisms function in a similar way in humans and rodents. The brain’s response to palatable foods differs from that to bland foods and responses of subjects that crave palatable foods differ from those who do not. Importantly, cravings for palatable foods activate similar brain regions and involve the same chemical messengers in humans as in rats.

Although it might seem obvious that changes in body weight reflect choices about what food we eat, how much we eat and how much we exercise, the long-term balance between energy intake and energy output is mainly determined by unconscious physiological systems.

Regulation of weight

We know that signals from the gut to the brain contribute to what is generally a very efficient homeostatic regulation of body weight. For most people, their weight generally remains remarkably stable over very long periods of time[3]. A very simple study in rats shows one component of this homeostatic regulation: if rats are given occasional sweet ‘treats’ they do not gain weight[5] – they compensate almost exactly for the additional calorie intake in the sweet food by reducing their intake of bland food (Figure 1). How do they do this?

Figure 1 Rats given regular access to a sweet ‘treat’ (a small amount of sweetened condensed milk; SCM) reduce their voluntary food intake compared to control rats that receive no SCM. Rats compensate exactly for the energy contained in the treat. Rats in this study did not increase their total energy intake and did not put on more weight than the controls. These data highlight the complex interaction between eating for pleasure and homeostatic control systems in the brain.

Clearly there is a signal to the brain when rats eat a sweet treat and exactly which brain areas respond is something that can be revealed in extraordinary detail in laboratory animals. This led to the identification of one particular site – a small group of neurons in the hypothalamus – that were particularly strongly activated. These cells turned out to be neurones that make the neuropeptide oxytocin[6].

Oxytocin has long been known to be a potent suppressor of appetite – but very recently oxytocin neurons have also been shown to send messages to areas of the brain that are involved in signalling ‘reward’. Thus these oxytocin neurons became a particular focus of interest in the context of food choice. But what exactly was activating oxytocin neurons when the rats ate the sweet treat – was it the taste or smell or something else?

To study this, we gave rats the sweet treat directly into the stomach, by slow gavage, and while doing this we recorded the electrical activity of the oxytocin neurons[7].

Remarkably, we saw that their firing rate rose progressively as we infused the sweetened condensed milk into the stomach – the oxytocin neurones were responding to some signal from the gut itself and were responding in proportion to the amount of sweet food ingested (Figure 2). But would they respond to any food given directly into the gut? When isocaloric cream was infused instead of sweetened condensed milk, the oxytocin neurones were inhibited.

Figure 2 Gavage of a high-sugar food (SCM) into the stomach of rats increases the electrical activity of brain oxytocin neurones (expressed as the change in number of spikes per second). These neurones seem to receive signals from the gut on the nutritional composition of foods in the stomach. Understanding these neurones’ networks and signalling pathways will help us understand better how the brain
controls food choice.

From this it is clear that signals from the gut to the brain report not just the amount of food ingested – how full the stomach is – they also carry specific information about the type of food ingested.

This information is coded in the brain by different populations of neurones and the different constituents of food evoke different behavioural and physiological responses. The oxytocin neurones, for example, do not simply control appetite – they also control glucose homeostasis.

Energy homeostasis involves much more than regulation of appetite – many other mechanisms also contribute to varying extents in different people and in different animal species. In some, excess energy intake is compensated for by increased energy expenditure – either in activity levels or in thermogenesis (heat production). But if this homeostatic mechanism is so effective, why do people ever become obese?

Sweet food is rewarding even when we are fully satiated.


One clue might be in the link between the oxytocin system and the reward systems of the brain. Sweet food is rewarding even when we are fully satiated. This in itself is not an explanation, nor does it necessarily lead us anywhere obviously useful – probably the last thing we really want to do with any public health intervention is to eliminate the pleasure that people get from eating.

However rewards are interesting, because the value of a reward is in part determined by our experience – it is not wholly intrinsic to the nature of the food. Commonly adults give children sweet food as ‘treats’ for birthdays and celebrations or rewards for good behaviour. This raises an interesting question – does this further increase the reward value of sweet foods? Perhaps even more interestingly, if ‘healthy’ foods (that children like) are given as a reward – can this increase their liking for those foods[8]?

Citizen empowerment

One important element in Nudge-IT was to launch a Massive Open Online Course (MOOC) called Understanding Obesity[9]. The MOOC had as a key objective the dissemination of ideas widely and effectively. The target audience was the general public and health-related professional groups that have direct contact with the public – groups that have an ability to inform and alter behaviour.

The course comprised films and podcasts featuring consortium members discussing scientific evidence behind the major themes of the Nudge-IT project. This was supported by a discussion forum and a number of ‘citizen science’ projects, where course participants undertook scientific work in collaboration with professional scientists in the consortium. In these projects, the course instructors formulated questions, developed hypotheses and designed studies to test these hypotheses.

Course participants were asked to predict the outcome of the study then provide data to test the hypotheses. The MOOC team analysed these data and shared and discussed it with participants. In this way we encouraged participants to step beyond received wisdom about food choice and obesity, and begin to ask and answer their own questions.

Professor Gareth Leng, Professor of Experimental Physiology and Dr John Menzies, Lecturer,

Centre for Discovery Brain Sciences, University of Edinburgh, Hugh Robson Building, 15 George Square, Edinburgh EH8 9XD



The Nudge-IT project ended in December 2018 and the final report is in preparation. The many papers produced from this project are available on the project website, along with some of the key tools developed. The final stages of the work are still in the process of completion and will be added to the website as they are published. The project also involved a number of workshops and meetings to disseminate the findings to other scientists, related professionals and policy makers, and included many activities to promote public awareness and understanding of the issues and of the contributions made by the project.



2.  Leng, G. et al. 2017. The determinants of food choice. Proc Nutr Soc. 76:316-327.

3.  Pandit, R., Beerens, S., Adan, R.A.H. 2017. Role of leptin in energy expenditure: the hypothalamic perspective. Am J Physiol 312:R938-R947.

4.  Leng, G. 2018. The Heart of the Brain. MIT Press.

5. Hume, C, Jachs, B. and Menzies, J. 2016. Homeostatic responses to palatable food consumption in satiated rats. Obesity (Silver Spring). 24:2126-32.

6. Leng, G. and Sabatier, N. 2017. Oxytocin - The sweet hormone? Trends Endocrinol Metab. 28:365-376.

7. Hume, C., Sabatier, N. and Menzies, J. 2017. High-sugar, but not high-fat, food activates supraoptic nucleus neurons in the male rat. Endocrinology. 158:2200-2211

8.  Belot, M, James, J. and Nolen, P. 2016. Incentives and children's dietary choices: A field experiment in primary schools. J Health Econ. 50:213-229.


View the latest digital issue of FS&T or browse the archive


Click here

Become a member of the Institute of Food Science and Technology