Abstract
A classic problem in the study of perception is determining how much is perceived and remembered from a brief presentation. In early studies, this problem was addressed by using the full report paradigm in which several letters were displayed and an observer was asked to report their identity. In more recent studies, the problem has been redefined to distinguish between different processing limitations on performance. Other studies have measured performance for visual information such as the size and shape of objects rather than linguistic information such as the identity of letters. Here, the problem is further narrowed in two ways.