The research adds weight to the idea humans are smarter eaters than previously thought and shows people ‘right size’ portions of high-calorie foods.

The study, to be published in The American Journal of Clinical Nutrition (to publish 00:01 BST Monday 13 June), is especially significant as it challenges a common view among researchers that people are apt to over consume high-energy foods.

Remarkably, the results indicated a degree of ‘nutritional intelligence’ whereby humans managed to adjust the amount they consumed of high-energy density options.

“For years we’ve believed that humans mindlessly overeat energy-rich meals. Remarkably, this study indicates a degree of nutritional intelligence whereby humans manage to adjust the amount they consume of high-energy density options,” said lead author Annika Flynn, Doctoral Researcher in Nutrition and Behaviour at the University of Bristol.


Thursday, June 9, 2022
11:00AM – 12:00PM EST

Researchers will discuss their findings in an expert panel. 

Media, register to attend here



Media, register to attend here



Thom: Hello and welcome to today’s Newswise Live event, we have with us a couple of researchers doing a study that’s coming out in a journal next week, about nutrition and food choices, so I would like to introduce briefly our panellists so that they can begin to tell us about this study.

So first we have Annika Flynn, she’s a doctoral researcher in Nutrition and behaviour at the University of Bristol and she's the lead author of the paper, and we also have Jeff Brunstrom, he’s PhD and he's professor at Bristol and – he’s professor of experimental psychology there, in the school of psychological science. Thank you both so much for joining, Annika as the lead author I want to give you a chance to tell us, the conclusions of your study suggest that the average person is able to adjust the amount of food that they eat, and this correlates to the caloric density of the food. How were you able to reach this conclusion from the data that you studied?

Annika: Yeah perfect, thanks for your question and thanks for the opportunity to speak to everyone here today, its really exciting. 

So what we did in our paper is we completed a secondary data analysis of meals which were eaten in a controlled environment, and these meals were actually part of a study conducted by Dr. Kevin and his colleagues to asses the effects of food ultra-processing on energy intake.

So, we actually then repeated our analysis and data from free living participants who were part of the UK national diet and nutrition survey and they completed a 7 day diet diary where they weighed and recorded all the food and beverages they consumed in the process of 7 days. 

What we ended up seeing is that in both data sets, that is Kevin’s Halls dataset and the NDNS data set, we saw that there was this tipping point. So as meals became more energy dense the caloric content of those meals actually started to decrease. So in other words people actually adjusted the amount of food they put on their plate in response to the energy density of the meal that they were going to consume. 

Thom: So this is in contrast to a lot of conventional wisdom especially what we hear from the dieting industry suggesting that processed foods for example, but all foods in general, people are not very good at controlling their portions or recognising whether the food has high caloric value or lower and adjusting their intake. So how does this study of yours change that perception and what are your thoughts about somewhat debunking that conventional wisdom?

Annika: Yeah so our study does challenge the idea that people do passively overconsume these really energy rich meals. There actually has been a tussle in literature about this key question about whether people are really sensitive to the composition of the foods that they're eating, including the energy content of the foods and there's been quite a lot of different methodologies that have been used to sort of approach this question. 

So kind of in part born out of necessity, cause of Covid lockdowns, we ended up deciding to explore this question by looking at real people eating real food and Kevin had this amazing really gold standard data from this controlled trials and then we decided to repeat the analysis in the UK and the United States where people were in free living conditions and we think that – and more importantly these findings really encourage us to recognize the complexity and how people are interacting with food in this modern food environment and the data do seem to suggest that we aren’t sort of passively over consuming these energy rich foods. And the really nice thing too is the pattern of the results that we do find, do fit with a variety of other research findings, specifically that we might eat to capacity in lower energy dense foods, so eat larger portions of these lower energy dense meals, but sort of as the meals start to increase and their caloric density – we kind of start to actually decrease the amount of  food that we’re putting on our plate. 

So we are showing a sensitivity per se in these more energy rich meals.

Thom: So, some of the data that you're looking at was unique compared to some of the conventional approaches to food studies. Id love to turn to Jeff and ask you, how was that different and in what ways would you say that this makes your study and the conclusions maybe better than some of these kinds of previous methods of monitoring food intake in a clinical setting.

Jeff: Thanks for the question. It’s worth saying I suppose that there are lots of different ways to study human dietary behavior and each come with their advantages and disadvantages.

When we think about this key question about whether humans are sensitive to the composition, the fairly traditional way to do that would be to just take the food and then to manipulate the food and see where they will not change the behavior. So for instance in protein, we’ll just take a few more protein and you’ll see behavioral changes and traditionally this question about calories and sensitivity of calories has been addressed by doing just that. By adding calories to food and that’s what they told us that yeah, we tend to [inaudible 05:24] passive over consumption.

What was unique about this data and the terrific opportunity here, was as Annika said, the access to personally use data from Kevin Hall, so he has people who come into this metabolic ward and stay there, live there, day after day, eating food. So it’s a semi naturistic environment, so there's no ecological virility in our data, but also he’s got this tremendous capacity to monitor very, very precisely the amounts of food consumed and food choices. 

So we’ve got this lovely coupling of ecological validity with accurate measurement. That’s unusual and I think that’s a real strength to us – and then - I mean to triangulate, to couple that with diet from NDNS, our national diet and nutrition survey here in the UK, we’ve got really comprehensive data, a very large sample of natural living behaviors.

Its self-report but it’s in a very large cohort. So we compliment a relatively small sample in Kevin Halls Metabolic ward, with data obtained at a population of that, logical level with many thousands of people, they pair this together and then that’s the answer, and that’s the strength here I think and we were able to do that and we reached the same conclusion in both data sets.

Thom: We have a question in the chat from Anna Guildford at Medical News today, Annika if you could answer this, do you think this behavior would change in people who are overweight?

Annika: I mean yeah that’s a really good question and that is kind of the avenue for future research. To be honest we didn’t really have scope in this paper but it is really exciting to think of all the research questions that could come out of this initial publication, one of them being whether or not there is sort of individual differences and how people might express this behavior and sort of the levels and degrees that they could be expressing in and what factors could be influencing how they might express that behavior. So long story short, we don’t know but we’re excited to try and figure it out.

Thom: So something that we spoke about a little bit in preparation for this is how you centered the data for each individual. Explain a little bit about that because I think that helps to answer the question about how it’s adjusted at least to correlate to the people’s weight in the caloric intake, maybe having something consistent there, can you explain that a little bit about how the data was analyzed that way?

Annika: Yeah so what we did before we even ran the regression analysis is we mean centered each meal within each participant. So we tried to account for individual variation across the sample. So it could try and address the fact that a larger person might eat a larger meal than a smaller person. I don’t know if you want to really get into the details but basically it means just calculating the average energy intake of all the meals and then sort of subtracting that from the actual caloric content. So you kind of look at try and wash out as much individual variation across the different participants, if that makes sense. 

Thom: So this is an area where there might be further exploration and Annika you're working on your PhD in this area of studying peoples behavior about food. Is that focus – so do you have plans for what next you might want to study based off of this?

Annika: Yeah so we touched on that a little bit just now, looking at individual variation, regarding the extent to which people, different groups of individuals might show different degrees of sensitivity. Another avenue of research as well is repeating the analysis in different datasets, because importantly these were data from US and the UK and it would be important to see if this pattern is consistent across different data sets as well. so that’s kind of the most immediate next step I think.

Jeff: I would just maybe add to say this also throws up, I think the question about bearing mind about individual differences is really, really important, I think cultural differences is important. I think the question here about – well where does this nutritional intelligence come from? What are the limits of our nutritional intelligence, so we’re just sort of scratching the surface here, but I guess refocusing the narrative perhaps around a more complex interaction we have is perhaps helpful and then thinking about how where this ability to discriminate calories comes from – is it something that is innate, is it something that is learnt at a personal level or is it something that forms as part of a collective form of learning that occurs within and across generations, it forms part of our collective cuisine or collective food practice, all these questions are fascinating and we’ll probably be wanting to explore them in different ways. 

Thom: We have another question in the chat, hunger and perception and acknowledgement of hunger can be at times complex, so she's wondering if you’ve also looked at emotional eating, overeating, by stress and anxiety – any factors such as those?

Annika: No, not in this data set. There is obviously like a breath of research that does look at emotional eating and individual susceptibility to emotional eat, there is different measures such as the three factor eating questionnaire, I don’t know that’s perhaps a bit specific, that do sort of try and capture different psychometrics, psychological properties or traits that might fall into that path and I believe the NDNS does have access to that, so that would be a potential avenue sort of looking at whether or not you have different patterns in individuals who score higher or lower on those different traits. I don’t know if that answers your question.

Thom: Yeah that’s great. Another question – and if you have any follow ups, please feel free to follow up. Another question from Anne again – how would you describe nutritional intelligence to a lay person?

Annika: Yeah that is kind of the golden question and to be honest that is something that Jeff and I have really struggled with to a certain extent. I would – this is my interpretation, how I might say and explain it to my mom, is that there is a ability whether learned or unlearned we don’t know, a ability for people to respond to the food composition of the foods that they are either choosing or getting ready to actually consume. The extent, the breath how reliable this might be, we really are just scratching the surface at understanding but there is to some degree an ability to respond to the composition of the food that the person is going to consume.

Jeff: Yeah I think that’s really good – I can say what it isn't and what we’re not referring – so we’re not referring here to our ability to be able to read a nutrition text book and to be able to articulate some of the science of nutrition, that’s not what we’re talking about. We’re talking about – as Annika says, about an ability expressing humans and other species to detect differences in the composition of foods and food considerably in their macro and micro nutrient content, to be able to detect that and to respond in some way, and there are different ways to go about exposing about nutritional intelligence and we show evidence around calorie detection.

Thom: So the process obviously involving some measurement and recording of the food intake. So, how in your opinion does that factor at all to – are the subjects under better behaviour because they know that they're recording it and they’re then there is something subconscious or otherwise going on that helps them to then execute better choices?

Jeff: I think it’s a great question, I guess in science we have to be concerned that we might be changing the outcomes simply by observing something, I think that’s called the Hawthorne effect and here one concern is that just by asking people what they're eating or by monitoring and peering over their shoulder, they somehow start to behave themselves or they change, they moderate their behaviour in some way. 

Now I guess there are two different influences here – when you’ve got people on a metabolic road, we can look at the data overtime – a suspicion here is – yes, people will be a little bit concerned about being monitored to initially, but our participants were there 28 days after where they do acclimatise and we have got a greater faith on the data on that basis.

The other point here is that the larger data that the NDNS has – again there is a real concern there about under reporting, but what really got me excited is when you look at both datasets, they provide the same answer. So if there is a bias that’s affecting the data in one context, its still enabling the fundamental underlying packing to match up with data that we got in a very different context. It would seem unlikely that the big picture, the completions that we were drawing from both studies, were affected by any particular bias.

Thom: Would it be fair to say then in your opinion that the recording and the measuring is a factor potentially in helping the humans to access that nutritional intelligence as you’ve sort of –

Jeff: I don’t – it’s possible that – yes. Its an interesting idea – asking people to focus and be mindful or to encouraging them to think about things they're consuming might possibly promote behaviours that enhanced our evidence in nutritional intelligence. Yes I think that’s plausible, completely untested.

Thom: That’s certainly a part of what we’d like to hear about these kinds of findings and especially when it’s something that – like you’ve done, contradicts conventional wisdom or other assumptions, it opens doors to look at what can be next – so I'm very curious about that. If anybody else on the audience has further questions – please do chat them. 

Any other thoughts about what plans there are next for this study or other take homes, other kind of – what people can use this for. Lay people as well as other scientists that may want to build on this in reference to your work. Annika any thoughts?

Annika: Yeah its exciting because it really is just the beginning and I think that – we’ve been saying this the whole time. But I think that is sort of the honest take home message is – across the last few decades of research, we’ve had a bunch of different studies that have tried to look at this from a bunch of difference perspectives and there have been different findings and different conclusions drawn and we just sort of decided to take a different perspective and we ended up with this really exciting result and as much as I would like to be able to say concretely – yeah this is what you can take away to be honest – I think in part we don’t really know a 100% and I think every time we sit down and have a conversation, especially with folks outside of our lab, we just end up with a bunch of new questions and I see that in the chat about age and level of activity – yeah – I need to – in this paper yeah we didn’t actually cover that but that is something that we’ve been discussing and I know level of physical activity, there's a whole bunch of research out there that looks at physical activity and sensitivity to calories and how people make decisions about food in relation to physical activity – so that is something to take into consideration. 

I think in part, the centring process that we did take in this study sort of did try to minimise all the individual differences such as age and level of physical activity, so in a sense we kind of washed that out in this study, but that doesn’t mean to say that isn't something that we can look at in the future, as Jeff said we keep getting new research ideas – so thank you.

Thom: Building on that question about age and physical activity, is there anything then that suggests to you similarly kind of approaching the topic with this kind of study and then isolating out more of those sorts of variables or is there a way that this can be used as the basis for identifying disordered eating for example?

Jeff: Yeah I'm going to say much of the same thing that Annika said – yeah these are really interesting questions and they are questions that we have been pondering ourselves. At this stage, we think about behaviours that lead to over consumption, or indeed we have to think about behaviours that lead to under consumption and thinking about eating disorders. I think maybe its worth – just coming back to a question that arise earlier about hunger inception and acknowledgment or hunger and emotional eating, over eating and so on – and for us I think a really exciting and interesting idea or opportunity here is to take the data that we’ve got at the moment and look at it like a snapshot, but we also recognise and have to recognise that peoples dietary patterns and behaviours are very much intimately bound up with other things that are going on in their lives. Stress, anxiety, depression and so on. These things both chronically and acutely, are unlikely to change the way we interact with food. 

Now, that’s important because it is in perhaps those interactions that we can start to understand how nutritional intelligence actually changes. Maybe from week to week or even from moment to moment, depending on our mood. The extent to which we discriminate calories, the extent to which our breakpoints between – the extent to which we respond to energy density, the extent to which they are expressed can be very much influenced in the moment as well – so I guess its my way of saying that- that we’re excited about this because it may result to refocus on things that humans do rather than our failings. But having said that I think it’s important for us to recognise that we’re not in this research saying that – in a sense earlier research and passive over consumption was wrong and we’re now showing nutritional intelligence – no, actually with very low energy densities we do see that people pretty much eat a fixed amount of food they put on the plate and we explained why in terms of gastric capacity.

So one of the things – great things that I think about this paper is that we’re not dismissing previous research, actually we’re explaining lots of different types of research, we’re showing how one can lead to eat food by volume for low energy density but also this sense that keep calories for more energy rich foods as well.

Thom: Another question in the chat – have you found differences in terms of dieting choices like veganism or intermittent fasting and does that nutritional intake, does that have an effect on food consumption? Again maybe not something in this study but any insights that this provides into looking into that further?

Annika: I mean yeah so dietary restraint is a factor that we’ve started to look at, especially in the NDNS data, we have – as I mentioned previously there is a questionnaire to try and capture those psychological traits, specifically dietary restraint and yes it could be a factor, but we don’t think it really changes the pattern that much, based off of very preliminary evidence that we’ve seen. So with the idea being that, if you are sort of higher in dietary restraint, you are restricting the portion of especially calorie rich foods that you would put on your plate in response to perhaps concerns about body weight, and again the very preliminary evidence that we’ve started to look at is that the pattern in individuals who score low in dietary restraint and high in dietary restrain are largely similar. But that is just a tiny drop in the bucket of all – as Jeff said, all the different factors that could be influencing our interactions with food. So again, anxiety stress, whether traits are relatively fixed, sort of acute stressors again.

In terms of veganism – that could be really interesting actually because you might expect that the caloric density of the meals that vegans would consume – I will throw myself out there, I am vegan, obviously we’re eating a lot of fruit and veg, short of grains – those foods tend to be lower in energy density, so that could be interesting because you basically sort of shift the pattern of results that you – sort of shift the energy density more into the lower density meals, so the patterns might be completely different in vegans and that could be really cool because you could look at the transition to veganism and you could see – for people who just start, is that pattern completely wonky, they have no idea what they're doing and then as they become more established vegans, we might start to see a more robust pattern where people kind of start figuring it out. 

So another really good research idea.

Thom: Okay, if there are no other questions, I think that might cover it for today. thank you both for the time to talk about your study – for anyone in the audience we can email the study to you, its also embedded as a downloadable pdf on the news release that I've linked in the chat on and also you can email the media contact Victoria Tag at Bristol, to follow up with further questions – maybe if you want to interview Jeff and Annika further or ask for a copy of that pdf if you're not able to access it on Newswise.

We will have a video recording and transcript of this that we will send out tomorrow morning, so you can use any of the commentary if you’d like. 

Thank you so much to Professor Brunstrom and Annika, please keep us posted on the next phase of interesting research and I'm sure that if you dig into any of the questions that have come up today, you will find even more interesting answers to this. Thank you so much for your time today and best of luck, congratulations on your paper.