The multiple looks model holds that all of the nerve fibers have similar integration times, but that at some point in the brain, a kind of memory operates. The outputs of the nerve fibers (the buckets) are stored for 200 or 300 ms, while the listener scans their contents. If you are trying to detect a sound, you add up the contents of all the buckets. If you are trying to detect a gap, you look in the bucket where the gap was supposed to have occurred to so if there is less  of a response in it.The buckets in our analogy are referred to as ÒlooksÓ.