This programme of research uses choice test methodologies to quantify hungry broiler breeder chickens’ preferences for qualitative or quantitative dietary restriction. It begins with an outline of quantitative dietary restriction, its severity and welfare implications before discussing methods of qualitative feed restriction and the difficulties ascertaining whether it represents a welfare improvement. Chapter two reviews the factors affecting diet preferences and discusses implications for feed restricted broiler breeder diet preferences. Chapter three outlines the use of a closed economy T-maze task to quantity the diet preferences of feed restricted broiler breeders. It concludes that broiler breeders can learn a food versus no food task but find it very difficult to learn a task in which both of the options are rewarded with food and this impeded diet preference quantification. Chapter four demonstrates that severity of feed restriction underlies these difficulties in learning. In Chapter five, a conditioned place preference task to identify the effects of diets on affective state (hunger versus satiety) is reported. A method validation group demonstrated that broilers show a state dependent preference for an environment associated with ad libitum access to food. However, birds failed to show a preference between an environment associated with quantitative dietary restriction and one associated with qualitative dietary restriction. Chapter six applies state- dependent learning (SDL) to quantifying the satiating effects of quantitative and qualitative dietary restriction. However, a validation group suggested that SDL preferences were probably an artefact of the test rather than a genuine state-led preference. Finally, the overall conclusion that no evidence was found that broiler breeders want, or that their welfare is improved by, qualitative feed restriction was drawn. However, the conditions under which a preference was reliably observed and the presence of hunger – state dependent effects on learning and expression of learnt preferences complicates the interpretation of any findings. Recommendations for further research are highlighted.