The role of working memory sub-components in food choice and dieting success

Evidence suggests a role for self-reported working memory (WM) in self-reported food intake, but it is not known which WM sub-components are involved. It is also important to consider how individual differences in dietary restraint and disinhibition influence WM and the impact of this on food choice. The current study assessed the relationship between WM sub-components and food choice, using computerised measures of WM sub-components and a direct assessment of food intake. The role of dieting success (measured by restraint and disinhibition) as a distal predictor of food choice that influences food choices via WM, and the role of WM more generally in dieting success were investigated. Female undergraduate students (N = 117, mean age: 18.9 years, mean BMI: 21.6 kg/m(2)) completed computer tasks assessing three components of WM (updating, phonological loop and visuospatial sketchpad) and a snack food taste-test. Greater visuospatial WM span was associated with a higher (lower) percentage of food intake that was low (high) energy dense. It was also found that unsuccessful dieters (high restraint, high disinhibition) had poorer visuospatial WM span and consumed a lower (higher) percentage of low (high) energy dense food. Visuospatial WM span significantly mediated the relationship between dieting success and percentage of low energy dense food intake. Further, dietary restraint was associated with poorer updating ability, irrespective of disinhibition. These findings suggest that better visuospatial WM is associated with a greater (reduced) preference for low (high) energy dense foods, and that deficits in visuospatial WM may undermine dieting attempts. Future work should assess whether the ability to deal with food cravings mediates the relationship between visuospatial WM and dieting success and investigate how WM may influence the mechanisms underlying behavioural control.