Neural interfaces based on surface electromyography (sEMG) represent an important non-invasive approach for human-computer interaction. While most current interfaces based on sEMG focus on the forearm, the wrist may be a more suitable location for practical use. This study investigates the feasibility of wrist neural interfaces by comparing the propagation and activation characteristics of motor units (MUs) decomposed from forearm and wrist signals. First, the presence of MU propagation trends was determined by assessing wrist activation and innervation zone distribution. Second, the center of gravity of MU was calculated, and activation area was quantified to compare associations and differences between wrist and forearm signals. Experimental results indicate that MUs exhibiting propagation trends account for 61.9 ± 8.7% (wrist) and 59.0 ± 10.4% (forearm), 58.6 ± 8.3% (wrist) and 64.6 ± 9.4% (forearm) at two force levels, indicating forearm signals can propagate to the wrist via muscle fibers. Furthermore, MUs decomposed from wrist and forearm signals exhibit similar physiological behaviors and activation characteristics. These findings expand potential applications for wrist neural interfaces and wearable devices.