Felice Jacka is a leading researcher of nutritional psychiatry, a relatively new field of study that considers the role of diet in mental and brain health. She is also the co-director of the Food & Mood Centre at Deakin University, Australia, and president of the International Society for Nutritional Psychiatry Research. She spoke to the Observer about her discoveries into the links between ultra-processed foods and the health of our brains.

We usually think of food as having an effect on us from the mouth down. Why study its impact on our brain?
Our gut microbiome affects virtually every aspect of health. It affects our metabolism, our blood glucose, our body weight. It affects the way genes turn on and off, and the amount of serotonin in our brain by altering the way the gut breaks down tryptophan [from protein] in our diet. It influences the stress response system. It affects the way the [energy-producing] mitochondria in our cells work, and it very profoundly influences our immune system.

So if you start to think about all of those things, all of which are very much involved in mental and brain health, and that diet is a key factor that affects the gut microbes, then you can see why we need to think about what we’re eating and what we’re not eating in relation to our mental and brain health.

And ultra-processed foods play a particularly unusual part in that?
It’s a question that science doesn’t yet have all the answers for. There is already some research that tells us that when a western junk food diet is given to young people who normally eat a fairly healthy diet for a week, we can see there are impairments in the cognitive functions of the hippocampus [an area of the brain]. We and others have shown that people who have a less healthy diet have a smaller hippocampus, and people who have a healthier diet have a larger hippocampus.

A shrunken brain is a frightening image. How scared should we be?
The hippocampus is the one area of the brain that can grow and shrink. People with, for example, severe depressive illness on average have a smaller hippocampus. But when they’re no longer depressed, their hippocampus grows again.

So the hippocampus is really important in mental health. We know that in animal studies, if they block the proteins that help to grow the hippocampus, then antidepressants don’t work. And then, of course, the hippocampus is very important for various aspects of learning and memory. It’s really important for kids at school, for example, while any of us who want to maintain our brains and not have cognitive decline as we get older, you want to do everything you can to keep your hippocampus nice and fat and healthy.

So why are ultra-processed foods impairing our brains?
We’ve just run a very interesting study where we tested the ultra-processed food v the wholefood version of a low-calorie diet, and looked at the impact on the gut microbiome. What we think is that with ultra-processed foods, even when the packet says it’s got this vitamin, or it’s got added minerals, or it’s got enough protein, your brain or your gut microbes don’t process it as food in the same way [as wholefoods]. We just need more science to show this conclusively.

Food manufacturers often claim ultra-processed foods can be healthy for their fibre and vitamins. Is that wishful thinking?
I think the big picture is really critical here. The industrialised food system is the leading cause of illness and early death around the world. And it is the leading cause of biodiversity loss. The UN food report of 2021 worked out that our global industrialised food system costs the globe roughly $20tn a year. About $11tn of that is in the impact to human health, like heart disease, high blood pressure and cancer. And about $7tn is the impact to the environment.

Therefore we should reduce consumption of ultra-processed foods where we can?
The real issue is that many people don’t necessarily have the option because it’s very often the case that ultra-processed foods are the cheapest. This is a failure of government policy, nothing short of that.

In the UK, as it is in Australia, when you go to fill up your car with petrol, you see row after row of ultra-processed foods, sugar-sweetened beverages, and every sort of chip and lolly and soft drink you could possibly imagine. Given that those foods interact really strongly with the reward systems of the brain, if you’re getting those in your face every day when you’re on the high street, at the supermarket, filling up your car with petrol, it is very difficult for individuals to make healthy choices.

You’ve also explored a connection between ultra-processed foods and neurodevelopmental disorders. What have you found?
In 2021, we looked directly at diet quality in mothers and children, and ADHD symptoms and diagnosis. And particularly, the mothers’ intake of healthy and unhealthy foods were both independently linked to children’s ADHD diagnosis.

That’s not an experiment, it’s just an observation. But it’s very concordant with animal studies and the emerging understanding that the microbes that are in mothers’ guts while the baby is growing, and the microbes that are in the guts of infants when they’re born, not only influence immune development but also brain development. Because people’s diets are so very impaired in the west, we think that that, along with antibiotic exposure and the loss of biodiversity from the food we’re eating, limits the healthy microbiome in mothers and infants. And that might be part of what’s leading to apparent increases in neurodevelopmental disorders as well as mental health problems.

Some researchers have criticised the term ultra-processed food as being ill-defined, overly broad and unfit for purpose. Should we reconsider the term?
Food must be the most complex exposure in the world. It comes in so many forms, has so many links to culture and emotional health, and it fulfils so many different roles. So it’s inevitable there’s going to be some fuzzy borders and some misclassifications.

But what I also know is what industry does, absolutely knowingly, is to jump in and make people confused. With cigarettes, it has never been categorically shown through a randomised controlled trial that smoking causes lung cancer. How could you randomly assign people to either smoke or not and follow them for 40 years? But we know from convergent evidence – from animal studies, from epidemiology – that if people stop smoking, their risk goes down.

What the tobacco industry had done for a long time was to confuse people by saying correlation doesn’t equal causation. Now, they do exactly the same thing with food. Think of the billions of dollars that are generated in profits through the industrialised food system every year, and the power of those industries to get people, including scientists, to muddy the waters. A lot, I think, of the conversation around the Nova classification [which groups food according to their level of processing] and if it’s fit for purpose is actually coming from industry.