FIG. 7.

Conceptual hypothesis illustrating descending and ascending influences of sacral and lumbar spinal activation, respectively, during stepping enabled by electrically-enabled motor control (eEmc). (A) eEmc at L2 excites the locomotor networks (LNs) for motor output. Optimal facilitation of the interneuronal and motoneuronal pools depends upon the excitatory input from sacral spinal cord stimulation. Short interconnecting and long arrows indicate short and long ascending propriospinal interneurons, respectively. For L2 monopolar and L2-S1 trials, the ascending facilitation by S1 is minimal because the sacral cord is not stimulated. With S1 monopolar and S1-L2 trials, there is ascending excitation of the LNs from the sacral cord stimulation. Because of the lack of an optimal level of pattern generation directly by L2 stimulation, however, the motoneuronal pool excitation is not optimized for robust stepping. For trials 5-7, the rostral cord is activated for rhythm generation, but the S1 activation frequency is not adequate enough for effective stepping. For trials 8-9, the LNs that respond to L2 stimulation also are facilitated by S1 stimulation and the interaction between these networks is maximized for optimal stepping. (B) Spinal cord cartoon demonstrates that the relative timing of the stimulation pulses at L2 and S1 produce unique interactions in the LNs shown in (C). For Windows 1 and 3, the modulation of the LNs is not adequate for robust stepping. In contrast, the relative timing of the stimulation pulses in Windows 2 and 4 allows for appropriate modulation and facilitation. (D) Interaction-evoked responses (black) are superimposed over an evoked response from a single stimulation pulse (red or blue) to depict the interactions described in (C). Color image is available online at www.liebertpub.com/neu