Zurek (1980) measured listeners' sensitivities to interaural disparities conveyed by a 5-ms "probe" segment embedded within a 50-ms burst of otherwise diotic broadband noise [P. M. Zurek, J. Acoust. Soc. Am. 67, 952-964 (1980)]. He found that thresholds for interaural time delay (ITD) and interaural intensitive difference (IID) were markedly elevated when the onset of the probe segment occurred between 1 and 5 ms after the onset of the burst. Zurek postulated that this occurred because the leading portion of the noise briefly inhibited sensitivity to subsequent binaural information. If such inhibition were the primary factor responsible for the elevation in thresholds, then the omission of the portion of the noise trailing the probe segment would be expected to have little, if any, influence on performance. In order to test this hypothesis, listeners' sensitivities to ITD and IID were measured using a paradigm similar to that employed by Zurek. The results revealed that the omission of either the leading or the trailing portions of the diotic noise led to substantial reductions in threshold ITDs and IIDs. The data were successfully accounted for by a model based upon a combination of a temporal window with an equivalent rectangular duration of approximately 10 ms and a weighting function representing a brief loss of binaural sensitivity just after the onset of a sound.