Double dissociation between meaningful and meaningless gesture reproduction in apraxia.

A model of the normal processing of gestures to account for the complexity of the different patterns of apraxia has been proposed by Rothi et al. (1991; 1997) and recently refined by Cubelli et al. (2000). This model, in analogy with the dual-route models of reading (e.g., Coltheart et al., 1993) encompasses three processing components1: (i) A “lexical route” which supports recognition (action “input lexicon”), identification (action semantic system) and production (action “output lexicon”) of familiar gestures; (ii) A visuo-motor conversion mechanism which turns visual information into motor programmes, upholding the reproduction of all seen gestures; (iii) A short-term memory workspace, on which both the lexical and the nonlexical route converge, storing the motor programmes to be executed. According to the functional architecture of this model, the reproduction of visually presented known gestures can be accomplished either by means of the lexical route or through the visuo-motor conversion system. A deficit in the reproduction of familiar gestures could then derive solely from the impairment of the gestural buffer. Such a deficit should cut across both meaningful and meaningless gestures alike, and should be yoked to an analogous disturbance of elicited, non-imitative gesture production. The malfunctioning of the visuo-motor conversion mechanism would give rise to a selective deficit of the imitation limited to meaningless gestures. On the other hand, the deficit of the lexical route would hamper spontaneous production of learned actions though imitation of both meaningful and meaningless gestures would be made possible via the visuo-conversion mechanism. In summary, whereas the model can easily account for the selective deficit in the imitation of meaningless gestures, it may run into difficulties should isolated impairment of meaningful gesture reproduction be observed, which is the pattern reported in this paper.