PurposeLumbar multifidus (LM) and transversus abdominis (TrA) show altered motor control, and LM is atrophied, in people with low-back pain (LBP). The Functional Re-adaptive Exercise Device (FRED) involves cyclical lower-limb movement against minimal resistance in an upright posture. It has been shown to recruit LM and TrA automatically, and may have potential as an intervention for non-specific LBP. However, no studies have yet investigated the effects of changes in FRED movement amplitude on the activity of these muscles. This study aimed to assess the effects of different FRED movement amplitudes on LM and TrA muscle thickness and movement variability, to inform an evidence-based exercise prescription.MethodsLumbar multifidus and TrA thickness of eight healthy male volunteers were examined using ultrasound imaging during FRED exercise, normalised to rest at four different movement amplitudes. Movement variability was also measured. Magnitude-based inferences were used to compare each amplitude.ResultsExercise at all amplitudes recruited LM and TrA more than rest, with thickness increases of approximately 5 and 1 mm, respectively. Larger amplitudes also caused increased TrA thickness, LM and TrA muscle thickness variability and movement variability. The data suggests that all amplitudes are useful for recruiting LM and TrA.ConclusionsA progressive training protocol should start in the smallest amplitude, increasing the setting once participants can maintain a consistent movement speed, to continue to challenge the motor control system.Electronic supplementary materialThe online version of this article (doi:10.1007/s00421-017-3648-3) contains supplementary material, which is available to authorized users.