OBJECTIVE To investigate the influence of fatigue on phasic muscle-activation patterns during dynamic trunk extension exercise. METHODS Fifteen healthy volunteers performed dynamic trunk-extension exercise through a 30-degree range-of-motion (ROM) exercise to task failure at an intensity of 50% of maximum. Electromyography (EMG) signals were recorded unilaterally from the lumbar extensor, gluteus maximus, and biceps femoris muscles, and signal amplitude was analyzed in 10-degree increments during the unfatigued and fatigued states (0-10 degrees from torso horizontal to the ground was considered extension, and 11-20 and 21-30 degrees of flexion relative to this were considered midphase and flexion, respectively). RESULTS Lumbar extensor EMG was approximately 75% of maximum EMG, with no differences being observed with respect to ROM or fatigue state. The gluteus maximus demonstrated an altered phasic activation pattern with fatigue, with an increased recruitment during the extension phase (fatigued-state extension-phase EMG: 89.1 +/- 8.3% > flexion phase EMG: 37.8% +/- 9.1%). The biceps femoris demonstrated a similar response during both the fatigued and unfatigued states (fatigued-state extension EMG: 77.8 +/- 5.4% > midphase EMG: 65.8 +/- 5.7% > flexion EMG: 46.8 +/- 4.0%; unfatigued-state extension EMG: 46.1 +/- 3.7% > flexion EMG: 27.1 +/- 2.6%). CONCLUSIONS During this exercise, as one moves from flexion to extension, hip extensor muscle activity increases, whereas lumbar extensor activity does not. Additionally, fatigue results in an altered recruitment pattern, with the hip extensors being activated to a greater extent in the extension phase. These findings suggest that when this exercise is performed in the prone position, it can be used to stimulate the lumbar and hip extensor muscles, but the specific exercise protocol in terms of set/repetition number and ROM will influence which muscles are primarily targeted.