We explored the hypothesis that the satiety that we expect a food to confer can influence the actual satiety that is experienced after it has been consumed. In Experiment 1 we manipulated ‘expected satiety’ by telling participants (N = 32) that a ‘fruit smoothie’ contained either a small or a large amount of fruit. All participants consumed the same smoothie. Nevertheless, those in the ‘large amount’ condition reported significantly lower hunger, 0, 1, 2, and 3 h after meal termination. In Experiment 2 we manipulated information about the volume of soup consumed in a meal. Before lunch, participants were shown either 300 ml or 500 ml of soup. Orthogonal to this, half consumed 300 ml and half consumed 500 ml. This process yielded four separate groups (25 participants in each). Covert and independent manipulation of the ‘actual’ and ‘perceived’ soup portion was achieved using a computer-controlled peristaltic pump. Immediately after lunch, self-reported hunger was predicted by the actual and not the perceived amount of soup consumed. However, 2 and 3 h after meal termination this pattern was reversed. Hunger was predicted by the perceived and not the actual amount. Together, these findings confirm a role for ‘expected satiety’ and show how memory for a recent eating episode can affect satiety in the inter-meal interval.