Postprandial glycaemic control is important in the prevention and therapy of type 2 diabetes and related diseases. Agents that may reduce postprandial glycaemia and/or insulinaemia, such as consumption of high-amylose foods, are considered beneficial; however, little is known about the dose–response relationship and the effects of cooking. The aim of the present study was to define the dose–response curve for postprandial glycaemic and insulinaemic excursions following meals of different amylose content and to compare the dose–response curves for meals containing cooked and uncooked starches. Following an overnight fast, rats ingested a test meal and blood was sampled over 2 h. The meal was given at 1.0 g carbohydrate/kg body weight, with an amylose content of 0, 270, 600 or 850 g/kg total starch. The area under the glucose curve did not differ under any condition investigated. For the uncooked-starch diets, area under the insulin curve was higher for the 0 g amylose/kg total starch meal than all other meals (P = 0·0001). For the cooked-starch diets, area under the insulin curve was higher in the O than the 600 and 850 g amylose/kg total starch groups (P < 0·01), but did not differ from the 270 amylose/kg total starch group. These results suggest that even a relatively small proportion of uncooked amylose (270 g/kg total starch) is sufficient to achieve a maximal attenuating effect on postprandial insulin concentrations as compared with 0 g amylose/kg total starch. Following cooking, however, a much higher proportion of amylose (≥ 600 g/kg total starch) is needed to achieve a similar effect.