It is estimated that around 10% of the world’s entire health burden is attributable to sub-optimal dietary intake, with 5.7% of disability-adjusted life-years (DALY’s) lost due to diets low in fruit and vegetables¹. Our food and drink preferences influence strongly what we choose to eat or drink, evidenced by the fact that preferences predict actual food intake². Understanding the factors that shape our liking of fruit and vegetables is therefore important for public health initiatives that aim to increase intake.

Twins can establish genetic and environmental influence on fruit and vegetable liking

Twin studies are a powerful method for understanding the extent to which liking for fruit and vegetables (or any trait) is driven by genetic and environmental influence. The basis of the method is to compare resemblance in liking between identical twin pairs who share 100% of their genes, with liking between non-identical twins who share about 50% of their genes. Because both types of twins share their environments to a very similar extent, the only real difference between the two types of twins is the fact that identical twins are twice as similar genetically. Greater resemblance between liking for identical versus non-identical twins therefore indicates a genetic contribution to preferences; similar liking for both identical and non-identical pairs indicates that environmental factors shared completely by twin pairs are important (e.g. maternal gestational diet); and low similarity in liking between twin pairs suggests that factors unique to each individual twin are key (e.g. illness).

Environmental influence on liking for fruit and vegetables changes from early childhood to late adolescence

Previous studies have shown that aspects of the environment shared entirely by twin pairs play an important role in shaping liking for fruit (51% to 53%) and vegetables (35% to 51%) among young children^3,4, alongside modest genetic influence (fruit: 53% to 54%; vegetables: 37% to 54%). This is not unexpected given the importance of the family environment for the eating behaviour of young children. However, the relative influence of genes and the environment can change dramatically with age, and the drivers of older teenagers’ preferences were unknown until very recently. In 2016, we explored this for the first time in a large sample of 18-19 year old twins (n=2865) from the Twins Early Development Study (TEDS), a population based British cohort of twins born in 1994-96⁵. Preferences for 62 individual foods were selfreported, and categorised into six food groups: fruit, vegetables, meat/fish, dairy, starch food and snacks. In contrast to the study of younger children we found no influence of the shared environment on preferences for any type of food. Instead, aspects of the environment that are not shared by twin pairs (experiences unique to each individual twin, such as having different friends) exerted an important influence on liking for all of the foods (46% to 68%), alongside modest genetic influence (32% to 54%; in keeping with estimates observed for young children) (see Figure 1).

Figure 1. Proportions of variation in food preferences explained by genetic (black portion of bars) and unique environmental (grey portion of bars) influences at 18-19 years of age (n=2685).

How could genes influence preferences for fruit and vegetables?

Food preferences can vary considerably, even among people from the same cultural background, evidenced by the fact that the food preferences of twin pairs can differ. Genes play a part in explaining some of these differences, and specific genes and their pathways have been proposed. Variants in the TASR gene family (a family of bitter taste receptors) affect sensitivity towards bitter compounds⁶, and carriers of a variant in the TAS2R8 gene are especially sensitive to bitter tastes, and have lower liking of cruciferous vegetables^7,8. Other proposed mechanisms relate more to cognitive aspects of food preferences⁹, such as food neophobia¹⁰ or reward circuitry¹¹.

What environmental influences might be important?

Much more is known about the environmental influences that shape food preferences, especially for children. When it comes to vegetable liking, exposure is key. In short, we like what we know, and we eat what we like. Repeatedly offering (15 times or more) vegetables to children can increase both liking and intake over the short term¹². However, research is needed to establish which strategies are effective for modifying preferences for fruit and vegetables on a wider scale (e.g. for public health initiatives).

Conclusions
Our findings indicate that the effects of family upbringing on liking for fruit and vegetables (and other foods) have entirely disappeared by late adolescence; replaced instead by environmental influences that are unique to each individual twin. This suggests that efforts to improve adolescent nutrition may be best targeted at the wider environment rather than the home. However, the substantial influence of the non-shared environment suggests considerable scope for modification of food preferences by environmental means. Research is needed to establish the most effective strategies for increasing liking of fruit and vegetables in this age group on a large scale.