During human bipedal locomotion neuronal features of quadrupedal coordination between arms and legs are likely conserved [1–5]. One method for assessing this coordination is by measuring the modulation of segmental reflexes. For example, after stimulation of nerves in the hand and foot, interlimb cutaneous reflexes in legs and arms are phase-modulated during the walking cycle  and during arm and leg cycling . Segmental central pattern generators (CPGs) have been implicated in this coupling regulating arm and leg motion . Coordination of arm and leg motion during walking, creeping, and swimming have also been ascribed to coupled CPG activity [1, 8].
Interaction between CPGs presumed to contribute to movement of the arms and legs can be estimated through examination of coupling effects during human rhythmic movement. A coupling effect is operationally defined as a measurable effect of limb movement or motor output on background or reflex muscle activity in another limb. For example, rhythmic arm cycling alters lumbar spinal cord excitability  in a manner that interacts with cutaneous inputs from the hand [10, 11]. Reciprocal and robust effects arising from leg cycling are also found in arm muscles [12, 13].
Recently, effects of rhythmic arm activity on background and reflex motor output have been examined during tasks in which both the arms and legs were rhythmically active such as arm and leg cycling [11, 14] and arm and leg stepping . These paradigms can be used as surrogates for “reduced” walking in order to separate arm and leg movement and thus estimate the contributions of the arms and legs to locomotor output.
Functional propriospinal interlimb connections between the fore and hindlimbs have been clearly demonstrated in the cat and in the neonatal rat [16, 17], confirming interplay between the cervical and lumbar pattern generators in coordinating rhythmic movement of all four limbs. These observations suggest neuronal circuits regulating the movement of all four limbs are akin to coupled locomotor oscillators. Complementary experiments have also been conducted in the human [7, 18].
An important factor related to effective and relevant arm and leg coordination is relaying movement related somatosensory feedback from the hands and feet to the legs and arms and vice versa . Over a century ago, Sir Charles Sherrington made passing mention of these effects in the form of the ‘hand foot’ reflex in the cat [19, 20]. Later, Lloyd described a kind of potentiation of interlimb reflexes from the cervical cord to the lumbar cord in the cat . In his experiments an interaction was observed wherein cervical or lumbar nerve stimulation produced potentiated responses in the opposite pathways. For example, cervical stimulation preceding lumbar stimulation produced larger lumbar responses than those arising from cervical stimulation obtained in isolation.
It remains to be determined if a potentiating effect can be induced reliably in humans during locomotor behaviours. Widespread interlimb reflexes evoked in leg muscles by cutaneous stimulation of the hand are phase-modulated and produce functionally relevant changes in ankle trajectory . These reflexes are complementary to the segmental responses evoked by stimulation at the ankle. Despite some differences in the expression of reflex amplitude based upon site of nerve stimulation, there are some common features as well, suggesting the possibility of shared pathways . Currently, little is known about possible convergence or integration of shared reflex systems from interlimb cutaneous networks during locomotion in humans.
The purpose of this study was to investigate convergent reflex effects following cutaneous stimulation of the hand and foot during arm and leg cycling (A&L) by using spatial facilitation. We tested the hypothesis that integration in reflex pathways arising from cutaneous inputs in the hand and foot converge on common interneurons presynaptic to the alphamotoneuronal pool for leg muscles during locomotor activation. Convergence in common interneuronal pathways would be revealed by non-linear summation of reflex amplitudes during simultaneous stimulation of the cutaneous fields in the hand and foot and is predicted to occur only during locomotor activity when CPG elements could make contributions. This outcome would support the assumption that the human locomotor system is organized in a similar fashion to other animals such as the quadrupedal cat.