Chewing requires a low level of muscle activity for jaw movement. Additional muscle activity is required to overcome the resistance of the food. The additional muscle activity consists of two contributions, an anticipating contribution before food contact and a peripherally induced contribution, about 23 ms after food contact. The amounts of both contributions depend on the information about food resistance obtained in preceding chewing cycles. It is not known whether this information is preserved if the resistance is absent during only a limited number of chewing cycles. Our aim was to investigate the extent to which information about food resistance obtained during chewing is used during subsequent cycles to generate anticipating and peripherally induced muscle activity. Subjects made rhythmic open-close movements at their natural chewing frequency, controlled by a metronome. Food resistance was simulated by an external force acting on the jaw in a downward direction during part of the closing movement. Jaw movement and surface EMG of the masseter and suprahyoid muscles were recorded during experiments in which sequences of at least 20 cycles with the force were alternated with a small, random number (from 1 to 10) of cycles without the force. The amount of anticipating muscle activity as well as the peripherally induced muscle activity in the first cycle with the force gradually decreased as a function of the number of preceding forceless cycles. About 30% of the additional muscle activity had an anticipatory origin, whereas the rest of the activity was evoked by the force regardless of the number of preceding forceless cycles.