Continuous Loop Averaging Deconvolution (CLAD) is a recently developed mathematical theory and algorithm that allows to deconvolve averaged electrophysiological signals obtained at high stimulation (Delgado and Özdamar (2004) ). It assumes that the individual unit responses are linearly combined to form quasi-steady state responses. One limitation of this algorithm is related to the fact that the length of the recording window has to be equivalent to the duration of the response. In this study, we have developed a modified deconvolution averaging algorithm to derive deconvolved responses when the length of the recording is longer than the duration of the response. The algorithm uses the Least Mean Square (LMS) optimization method to find the deconvolved recording that best fits the acquired data. The method has been applied to study the evoked activity in 40 Hz auditory responses.