Temporal auditory processing deficits are thought to play an important role in some pathologies of speech understanding difficulties. The purposes of the present study were to determine whether short silent gaps within a pure tone can be used to elicit mismatch negativity (MMN) as an objective measure of temporal resolution, and to investigate the relation between MMN and performance on a behavioral gap-detection task. The stimuli used for both tasks were 1-kHz sinusoids presented in a low-pass masking noise. Behavioral gap-detection thresholds were determined using an adaptive three-alternative forced-choice procedure. To elicit MMN, a series of deviant stimuli with varying gap durations was generated and presented in an oddball paradigm among standard stimuli without a gap. Only the gaps larger than the mean behavioral gap-detection threshold evoked a clear MMN. There was no clear relation between psychoacoustic and MMN thresholds.