1. The effects of olivocochlear (OC) feedback on signal processing in the cochlea were studied by comparing responses seen with and without a contralateral noise or by comparing responses seen before and after cutting the OC bundle (OCB). Adding and subtracting a contralateral noise is a convenient, reversible way of changing the level of OC feedback; however, it fully reveals only the contribution of the contralaterally responsive efferent fibers. Cutting the OCB can reveal the full contribution of all fibers in the OCB; however, the manipulation can only be performed once per experiment. 2. The amplitude of the compound action potential (CAP), recorded from anesthetized or decerebrate cats in response to tone pips, could be increased by addition of contralateral noise at moderate sound pressure levels. These enhancement phenomena were most easily demonstrable when the tone pips were masked by ipsilateral broadband noise; however, in some animals CAP enhancement was seen in the absence of ipsilateral maskers. Enhancement-in-quiet may arise because of internal masking from animal-generated noise. All contralateral-noise enhancement disappeared when the OCB was cut. 3. Enhancement effects of contralateral noise could be seen in both simultaneous and forward-masking paradigms. Enhancement was largest for high-frequency tone pips (8-16 kHz) and could be demonstrated over a wide range of tone-pip levels and ipsilateral-masker levels. Suppression of CAP by the contralateral noise was often seen for lower tone-pip frequencies (2-8 kHz) and lower tone-pip intensities. These trends may be understood in the context of known properties of OC peripheral effects and known properties of physiological masking. 4. Cutting the OCB resulted in a decrease in CAP amplitudes to masked tone pips. When CAP was measured to tone pips presented in equilevel, binaural noise, OCB section resulted in a decrease in CAP amplitudes equivalent to at least a 6-dB decrease in signal-to-noise ratio. Such antimasking effects of an intact OCB were seen in both simultaneous and forward-masking paradigms. 5. Present evidence suggests that all these antimasking effects can be explained on the basis of activation of the medial OC fibers to the outer hair cells. By suppressing responses to continuous noise backgrounds, the OC reflex may enhance responses to transient masked stimuli by decreasing the level of adaptation in auditory nerve fibers. Such effects of the OC reflex should improve discrimination of transient signals presented in a continuous noise background.