Two experiments investigated temporal generalization in humans using a computer based task which presented red dots with a range of lines at different angles and durations. After training with a standard S+ stimulus duration, generalization testing commenced with an asymmetrical series of presentations of lines of varying angles and durations. Experiment 1 had four conditions, with a standard S+ duration being the presentation of a red dot for a fixed duration. Two of the conditions had the addition of the line tilt. In Experiment 1, 11 participants produced a peak shift effect in all four conditions. Experiment 2 was the same as Experiment 1 except that there were two conditions. Condition 2 was the same as Condition 1 except that the participants were given a verbal instruction to think of the line tilt as if hands on a clock. All 9 participants produced a peak shift effect in both conditions. In Experiment 2, the effect of categorising the stimuli and in turn changing the stimuli from a continuous dimension to discrete stimuli (one in which could be labelled) and the verbal instruction of to think of the line tilt as if hands on a clock did not have an effect on the peak shift as predicted. The results for both experiments were in accordance with predictions of adaptation level theory.