Control processes in typing

Recent studies have used errors in speech and typing as a source of evidence for the hypothesis put forward by Lashley (1951) that response is hierarchically organized. The present study continues this line of inquiry on typing, using both the form of the error and its latency as data. Earlier models to account for the latency of typing have postulated two stages of output processing and with a slight modification to the notion of a stage it is shown that they can also account for the error data. Suggestions are made on the nature of codes represented in each stage and on the aetiology of errors. An analysis of response latency in the vicinity of errors supports a conclusion that error detection is very rapid and the (inferred) incidence of detection suggests that it is based upon a comparison process rather than an estimate of response likelihood.