Not all researchers, however, agree with this conclusion. Dowey (1996) reviewed the
literature examining food and rewards and argued that the conflicting evidence may
relate to methodological differences between studies and that studies designed to change
food preference should be conducted in real life situations, should measure outcomes
over time and not just at one time point, should involve clear instructions to the children
and should measure actual food intake not just the child’s stated preference. The recent
intervention study described above incorporated these methodological considerations
into its design (Lowe et al. 1998) and concluded that food preferences could be improved
by offering rewards for food consumption as long as the ‘symbolic context’ of reward
delivery was positive and did not indicate that ‘eating the target foods was a low value
activity’ (Lowe et al. 1998: 78). As long as the child cannot think that ‘I am being offered
a reward to eat my vegetables, therefore vegetables must be an intrinsically negative
thing’ then rewards may work.
The associations between food and rewards highlights a role for parental control over
eating behaviour. Some research has addressed the impact of control as studies indicate
that parents often believe that restricting access to food and forbidding them to eat food
are good strategies to improve food preferences (Casey and Rozin 1989). Birch (1999)
reviewed the evidence for the impact of imposing any form of parental control over food
intake and argued that it is not only the use of foods as rewards which can have a
negative effect of children’s food preferences but also attempts to limit a child’s access to
foods. She concluded from her review that ‘child feeding strategies that restrict children’s
access to snack foods actually make the restricted foods more attractive’ (Birch 1999:
11). For example, when food is made freely available children will chose more of the
restricted than the unrestricted foods particularly when the mother is not present (Fisher
and Birch 1999; Fisher et al. 2000).
Food and physiological consequences: Studies have also explored the association
between food cues and physiological responses to food intake. There is a wealth of
literature illustrating the acquisition of food aversions following negative gastrointestinal
consequences (e.g. Garcia et al. 1974). For example, an aversion to shellfish can be
triggered after one case of stomach upset following the consumption of mussels.
Research has also explored pairing food cues with the sense of satiety which follows their
consumption. One early study of infants showed that by about 40 days of age infants
adjusted their consumption of milk depending upon the calorific density of the drink
they were given (Formon 1974). Similarly children can adjust their food intake accord-
ing to the flavour of foods if certain flavours have been consistently paired with a given
calorific density (Birch and Deysher 1986).Problems with a developmental model
A developmental approach to eating behaviour provides detailed evidence on how food
preferences are learned in childhood. This perspective emphasizes the role of learning
and places the individual within an environment which is rich in cues and reinforcers.
Such an analysis also allows for a moderate interaction between learning and physiology.
However, there are some problems with this perspective as follows:142 HEALTH PSYCHOLOGY