Resistive and elastic loads added to inspiration are readily detected, and detection latencies vary as a function of load magnitude and load type. In the present study, we recorded endogenous event-related potentials (i.e., N2 and P3) to the detection and classification of large (15.0 cmH2O.1-1.s and 70.0 cmH2O/l) and small (1.45 cmH2O.1-1.s and 19.0 cmH2O/l) loads equated for subjective magnitude in 14 men (mean age 21.14 yr). In blocks of trials comprised of either large or small loads, subjects made a button-press response upon detecting a load and then classified the load as resistive or elastic. Loads were presented briefly (for approximately 200 ms) early in inspiration and at the same level of inspiratory pressure. For loads of comparable magnitude, subjects detected equivalent numbers of resistive and elastic loads but could not discriminate reliably between load types. On the other hand, the latency of N2 was shorter to larger than to smaller loads, to resistive than to elastic loads, and to correct than to incorrect load classifications. The latency of P3 was affected similarly by load magnitude and load type. These findings demonstrate that event-related potentials are elicited by brief presentations of resistive and elastic loads and that N2 and P3 latencies vary reliably as a function of load magnitude and load type. Most importantly, event-related potential latencies are sensitive to load type and to classification accuracy even when resistive and elastic loads are not distinguishable subjectively.