(ek’ spō nen, sh∂ āt) 1. To introduce time-dependent gain in processing, that is, to multiply input values by ekt where k may be either a positive or negative real number. The value of k is often determined arbitrarily or empirically. Often refers to additional gain adjustment after removal of gain effects during recording and correction for spherical divergence. 2. To magnify variations in input data so that the output is proportional to a constant raised to an exponent of the input value.