Understanding the effects of respiratory load on neural respiratory drive and respiratory pattern are key to understanding the regulation of load compensation in respiratory disease. The aim of the study was to examine and compare the recruitment pattern of the diaphragm and parasternal intercostal muscles when the respiratory system was loaded using two methods. Twelve subjects performed incremental inspiratory threshold loading up to 50% of their maximal inspiratory pressure, and 10 subjects underwent incremental, steady-state hypercapnia to a maximal inspired CO2 of 5%. The diaphragmatic electromyogram (EMGdi) was measured using a multipair oesophageal catheter, and the parasternal intercostal muscle EMG (sEMGpara) was recorded from bipolar surface electrodes positioned in the second intercostal space. The EMGdi and sEMGpara were analysed over the last minute of each increment of both protocols, normalized using the peak EMG recorded during maximal respiratory manoeuvres and expressed as EMG%max. The EMGdi%max and sEMGpara%max increased in parallel during the two loading methods, although EMGdi%max was consistently greater than sEMGpara%max in both conditions, inspiratory threshold loading [bias (SD) 9 (3)%, 95% limits of agreement 4-15%] and hypercapnia [bias (SD) 6 (3)%, 95% limits of agreement -0.05 to 12%]. Inspiratory threshold loading resulted in more pronounced increases in mean (SD) EMGdi%max [10 (7)-45 (28)%] and sEMGpara%max [5.3 (3.1)-40 (28)%] from baseline compared with EMGdi%max [7 (4)-21 (8)%] and sEMGpara%max [4.7 (2.3)-10 (4)%] during hypercapnia, despite comparable levels of ventilation. These data support the use of sEMGpara%max, as a non-invasive alternative to EMGdi%max recorded with an invasive oesophageal electrode catheter, for the quantification of neural respiratory drive. This technique should make evaluation of respiratory muscle function easier to undertake and therefore more readily acceptable in patients with respiratory disease, in whom transduction of neural respiratory drive to pressure generation can be compromised.