Previously, we have used a 'method of constant stimuli' to quantify the satiety that different foods are expected to deliver. Our data indicate that foods differ considerably (some are expected to deliver 5-6 times more satiety than others [per kcal]). In the present study we explored the relative importance of 'expected satiety' in decisions about portion size. For eight different snack foods, we measured 'ideal' portion size and compared these values with corresponding measures of liking, expected satiety, and intention to restrict intake. Across participants (N=60), ideal portion size was predicted by both liking and expected satiety. Individuals differed in the relative importance of expected satiety and liking. In particular, expected satiety was a more important predictor in restrained eaters and in individuals with a higher BMI. In this study we also included a measure of food reward. For each food, reward was inferred from a measure based on cash spend per kcal. Again, food liking and expected satiety were both significant predictors. Together, our findings confirm the importance of expected satiety and they demonstrate the quantification of separate affective and non-affective determinants of food reward and portion size.