In most previous experiments on the ability to identify sound intensity, the stimulus set has been chosen such that the range of intensities is many times larger than the value of the “just‐noticeable difference” in intensity that has been derived from the results of intensity discrimination experiments. In such cases, the resolution obtained in the identification experiment is much worse than would be expected merely on the basis of the pairwise discriminability of the stimuli. In this paper, we report the results of a study designed to explore the prediction [Durlach and Braida, J. Acoust. Soc. Amer.46, 372–382 (1969)] that this discrepancy between identification and discrimination does not occur if the range of intensities employed in the identification experiment is sufficiently small. Generally speaking, the results of this study support the prediction.