SUMMARYThe purpose of this study was to test the hypothesis that the potentiation of concentric twitch force during work cycles is dependent upon both the speed and direction of length change. Concentric and eccentric forces were elicited by stimulating muscles during the shortening and lengthening phases, respectively, of work cycles. Work cycle frequency was varied in order to vary the speed of muscle shortening and/or lengthening; all forces were measured as the muscle passed though optimal length (L o ). Both concentric and eccentric force were assessed before (unpotentiated control) and after (potentiated) the application of a tetanic conditioning protocol known to potentiate twitch force output. The influence of the conditioning protocol on relative concentric force was speed dependent, with forces increased to 1.19±0.01, 1.25±0.01 and 1.30±0.01 of controls at 1.5, 3.3 and 6.9Hz, respectively (all data N9-10 with P<0.05). In contrast, the conditioning protocol had only a limited effect on eccentric force at these frequencies (range: 1.06±0.01 to 0.96±0.03). The effect of the conditioning protocol on concentric work (force ϫ distance) was also speed dependent, being decreased at 1.5Hz (0.84±0.01) and increased at 3.3 and 6.9Hz (1.05±0.01 and 1.39±0.01, respectively). In contrast, eccentric work was not increased at any frequency (range: 0.88±0.02 to 0.99±0.01). Thus, our results reveal a hysteresis-like influence of activity-dependent potentiation such that concentric force and/or work were increased but eccentric force and/or work were not. These outcomes may have implications for skeletal muscle locomotor function in vivo.