A method for studying the evolution of naming error types in the recovery of acute aphasia: a single-patient and single-stimulus approach

In this study, we present a method for analysing the evolution of picture naming errors in the follow-up of single patients affected by acute aphasia. In particular, we have based our analysis on the presence of response type inconsistency, as patients often fail to give the same type of response to the same stimulus at a task repetition attempted after a short time. Due to the uncertain definition of the type of response associated to a given stimulus for each stage of the clinical course, the investigation of the factors underlying the transition between different types of response is a serious methodological challenge. The solution presented here is based on a multiple presentation of the same naming battery at different stages of the clinical course, on the estimation of the probability associated with each response type at each stage, and on the estimation of the transition probability between different response types from one clinical stage to another. The basic idea was to use the set of probabilities referred to above as single stimuli weights in the study of linear models; these permit to compare different types of responses and different types of transitions. We present the application of this method to the study of a single case, a woman affected by fluent aphasia examined twice in the first 2 weeks following stroke. Besides discussing empirical findings, we comment on the usefulness of this method for wider fields of inquiry. (C) 2003 Elsevier Ltd. All rights reserved.