Previous action potential (AP) tuning curve methods have used a reduction in amplitude of the probe‐elicited AP as an indication of tone‐induced masking. The reduction criterion used in different studies has varied from 25% to 100%. For low level probe stimuli, which elicit a low‐amplitude AP, this is a sensitive indicator. In contrast, for high‐amplitude AP responses elicited by high‐level stimuli, the required reduction in absolute terms is large, making it an insensitive indicator. AP tuning curves have been recorded using a sensitive method for detecting masker/probe interaction with a fixed criterion, unrelated to the unmasked AP amplitude. For each masking condition, a derived response was obtained by digitally subtracting the tone‐masked AP waveform from the unmasked response. Derived responses are generated if there are ANY changes in the AP waveform induced by the masker, including amplitude changes, latency changes, or even changes in AP morphology not necessarily associated with the major peaks. A fixed criterion (10 μV) of tone‐derived (TD) response was used as an indication of interaction of the responses to the masker and probe. Tuning curves generated by this method were compared with those generated by conventional amplitude reduction (AR) methods. TD tuning curves show different characteristics, especially with respect to increasing probe levels. They appear to give a good representation of the array of afferent fibers responding to a probe stimulus. In addition, frequency regions making minor contributions to the AP are better represented in TD tuning curves.