Eventually a threshold is reached where everyone hears just one of two rapidly presented sounds.
(click) (click)
-----------------------|-|--------------------------
(time)
This is interesting (possibly), but does it have any relevance to speech and language, you may ask. No, not really, but the next step does. The principle is the same for distinguishing the order of rapidly presented series of different stimuli.
Speech, itself, is nothing but sequences of different stimuli (phonemes), and perceiving their order is more than half of the decoding game. If the phonemes are far enough apart in time, you can perceive their order.