ABSTRACT
The problem of speech perception by ear and by eye has been studied within the framework of falsification and strong inference. The issues that we have addressed seem fundamental to developing a psychological understanding of the phenomenon. The methods of information integration and mathematical model testing appear to be ideally suited for addressing many issues. Although the experiments have been reasonably successful in providing answers to some of the questions, future work will be necessary in order to resolve other questions. On the basis of the research, perceiving speech by ear and by eye is described within the context of a model representing a general theory of perceptual recognition. This model provides an algorithm for evaluating and integrating multiple sources of information in pattern classification.
