查看文章

This chapter explains how to measure visual effects. Psychophysical methods are usually described in a historical context, starting with Weber, Fechner, and Wundt in the 1800sand the development of the theoretical foundations; here we take a practical approach, focusing on what is most useful to know. Drawing conclusions about visual perception is difficult-not all questions are answerable. Psychophysics only considers questions that can be answered by measuring an observer's performance of a visual task. The art of psychophysical measurement is to channel one's curiosity into designing a question that retains the motivating interest and yet can be convincingly answered by measuring task performance. This chapter describes those tasks and measures that have proven to be most useful in vision research, and explains what kinds of question they answer.

Consider the complications in what might seem the simplest question,'Do you see it?'One can simply present visual signals and put the question directly to the observer. But, on reflection, are we really interested in whether the observer says' yes', or are we interested instead in whether the observer can prove that he or she has seen the signal, eg by correctly identifying it or locating it in time or space? In either case, when we collect the responses we find that the answer is probabilistic: in practice one measures the probability of each kind of allowed response to the signal. But then what does one do with these probab ilities? Such complications need to be carefully considered on an experiment-by-experiment basis, but we will share with the reader the guidance offered by existing theory …