Jim was doing well compared to Kim, a five-year-old autistic girl referred to the same clinic. For a while, Kim too had eaten a limited diet of hotdogs, peanut butter, bacon, chocolate, eggs and toast. She too would cry, tantrum and throw food at mealtimes. After an illness, though, she stopped eating altogether and for six months had been completely dependent on feeding through a gastrostomy tube.
Most parents would feel overwhelmed at the thought of feeding these children and somehow broadening their horizons. I know I would. Food refusal is demoralizing at the best of times; all the more so when you are dealing with the other challenges of caring for an autistic child. If a child disliked most foods to the point that they provoked tears and rage, it would be very hard to bring yourself to do anything other than sigh and make another toasted cheese sandwich.
Jim and Kim sound like two hopeless cases. But they weren’t. Within two weeks of intensive treatment at the centre, Jim’s repertoire of foods had increased from three to sixty-five. Kim, meanwhile, would now eat forty-nine different foods and no longer needed the feeding tube. This huge increase in ‘likes’ (and decrease in ‘dislikes’) was achieved not through any magic but simply through a more systematic and intensive version of Lucy Cooke’s Tiny Tastes system.
Therapists at the clinic engaged the children in many repeated taste sessions to expose them to pea-sized amounts of novel foods in the course of the day. Unlike with Tiny Tastes, the therapists added in an ‘escape prevention’ element: the child was told ‘when you take your bite, you can go and play’ and were not allowed to leave the room until the bite was taken. If they screamed or cried, this was ignored, but if they ate the pea-sized bite, they were praised. There were also ‘probe meals’ at which larger quantities of the new foods were offered – three tablespoons of three different foods – with a ten-minute time limit and no requirement to eat the food.
The results of this experiment are astonishing. To go from being fed by a tube to being able to eat forty-nine different foods is life-changing, for the whole family. A three-month follow-up showed that Jim and Kim had not lost the majority of their new likes at home. They had not slipped into the old unhappy mealtimes of before. Food was no longer a trauma to them. Both sets of parents were continuing to offer the children taste sessions outside mealtimes. Jim’s range of foods was now fifty-three. This large repertoire of foods was all the more impressive considering that Jim’s parents had decided to become vegetarian since the start of the intervention, the sort of change that autistic children often find unsettling. Kim’s range of foods was still forty-seven. In place of a tube, she was now enjoying a wide range of different flavours and textures, without tears or rage. Autism goes along with restricted social interaction. Yet Kim’s new likes placed her back in the social world of the family dinner table.
Similar work is being done by therapists at specialist feeding clinics across the world, although Keith Williams, head of the feeding clinic at the Penn State Hershey Medical Center, says that these methods are by no means standard practice. Too many feeding therapists still treat limited eaters such as Jim and Kim by offering them whole platefuls of disliked food and hoping they will suddenly decide to eat it. But when these interventions succeed, they show what huge potential there is for changing our likes and dislikes for the better through a change to our eating environment. No one is doomed to like nothing but cheese sandwiches and hotdogs. If it’s possible to train a severely autistic three-year-old to love fifty-three different nutritious foods, there’s hope for us all.
The trouble is, though, that most of our food environment influences us in an opposite direction. Every day, children are exposed to messages – whether on giant hoardings and TV ads or from looking in friends’ lunchboxes – telling them that they should like the very foods that will do them the most harm.
Karl Duncker’s 1930s experiments on children’s likes and dislikes are much less well known than Clara Davis and her feeding orphanage. But they offer just as great an insight into how our tastes are formed, almost in spite of ourselves, by forces we are only dimly aware of. While Davis was interested in what tastes look like stripped of the normal social influences, Duncker wanted to pinpoint how those influences actually work.
In 1936 Duncker (born in Leipzig in 1903) was a promising young Gestalt psychologist exiled from Nazi Germany – where his parents were prominent communists – to Britain where he continued his work. One of his great philosophical interests was pleasure and what causes it. His definition of the pleasure of anticipation was a child who ‘has been told that he is soon to have a piece of candy … glowing all over with happiness’. In one of his papers, Duncker asked why eating a fine juicy beefsteak could cause such delight; he decided that it wasn’t just that it took away the pain of hunger. It was the sensory enjoyment of biting into it, and the feeling it gave that ‘life is grand’.65
On arrival in Britain, Duncker set himself the task of investigating the role of social suggestion in forming food preferences. Given that likes and dislikes varied to an ‘astounding degree’ among different cultures, he realized that there must be a process of social influence at work. His mission was to unravel the psychological processes by which likes were formed.
Duncker’s experiments involved children from Somers Town nursery school in London NW1, which was then a poor district of London.66 The first experiment was a simple one. Boys and girls aged between two and five were asked to make a food selection from carrots, bananas, nuts, apples, bread and grapes. What Duncker found was that children were far more likely to select the same foods as one another if they made their choice in the presence of other children, than if they were alone. For children younger than twenty-seven months there was a wonderful ‘social indifference’: ‘when they had fixed their minds upon the food, nothing else seemed to exist.’ Above that age, however, there was a marked tendency to copy the likes of other children, especially if the child who selected first was just a little older. There was one pair of girls where one was an extroverted five and the other a shy four. Before choosing her food, ‘B would always send some furtive glances over to A as if for reassurance’.
We’ve all seen this kind of peer influence at work. If you offer a snack to a group of young girls, they will often tie themselves in knots second-guessing what the others will go for before making up their own mind. You don’t want to be a lone wolf eating popcorn when everyone else has opted for toast. Duncker’s findings about social suggestion when eating have since been confirmed by at least sixty-nine separate experiments.67 This is a very robust phenomenon. Depending on the influence of those who share our meals, we may eat faster or slower; we choose different foods; we manage larger or smaller portions.
Duncker’s second experiment was more dramatic. He took two substances. One was a white chocolate powder flavoured with lemon – a very luxurious commodity in 1930s Britain and ‘decidedly pleasant’. The other was valerian sugar coloured brown, valerian being a herbal root traditionally used as a sedative: a very bitter and medicinal flavour that Duncker called ‘rather unpleasant’. He then asked the nursery teacher to read the children a story about a hero, Micky, a little field mouse, who hates one food – ‘hemlock’ – and loves another – ‘maple sugar’. When Micky discovers maple sugar in a tree, he realizes he has never ‘tasted such good stuff before’. But the hemlock bark is ‘sour and disgusting’.
After the story, the children were then asked to taste some actual ‘hemlock’ – which was really the delicious white chocolate powder; and ‘maple sugar’ – which was really the unpleasant valerian sugar. The deception did not exactly work. Many of the children recognized that the ‘hemlock’ was really chocolate. Yet when asked to choose which substance they preferred, 67 per cent of them opted for the nasty-tasting ‘maple sugar’ because of the positive associations in the story (only 13 per cent chose it in a control group with no story).
Can our likes and dislikes really be so easily influenced? Apparently so. Duncker’s experiment shows that a simple story is enough to make children forget – for a time – that they like chocolate. For Duncker himself, having witnessed Hitler’s rise to power, it was no surprise that human beings are suggestible in their ‘likes’ or that social forces can make them suppress their natural impulses. At the time Karl Duncker was doing his peaceful experiments with children and chocolate, his younger