Fig. 5. Leg stiffness adaptation in the two consecutive contacts. We normalized each leg stiffness value for track types i=1–3 (_leg,i) with a subject-specific reference run on the flat, undisturbed track (_leg,i=0) separately for each contact. Open boxes represent trials for the first contact, grey boxes those for the second contact. Leg stiffness was altered between contacts on bumpy ground (track type 1–3). Significant differences were found for track types 1, 2 and 3 (see Table 3).