Conclusions and theory
In Chapter 3 several theories of attention were reviewed. How do they stand up in the light of the succeeding chapters?
To begin, it is unlikely that we can at present expect to construct a unified theory of attention. We have seen that there are many different kinds of behaviour which are included under the concept, and that there are strong reasons for thinking that several of them bear little real relation to one another, other than in name. Even when we restrict attention to the experiments on selection between simultaneously presented messages, or between input and output, there is evidence of selectivity occurring at several different levels of complexity, from loudness to language. It would be naive to expect such different types of classifying necessarily to be controlled by a single mechanism. Moreover there is no obvious reason to expect hearing and vision to be controlled in the same way. Visual signals tend to be spatially extended but of short duration, while characteristically auditory signals take a long time and consist of information presented sequentially. The estimated sampling times which we have seen in visual work seem to be much longer than in hearing. Even if there were a single control system sharing capacity between hearing and vision there might be a whole hierarchy of sampling times and switching times which would differ from sense to sense and perhaps even within senses.