Procesamiento fonológico en la escritura manual de palabras aisladas

Abstract

In the present manuscript, we provide striking evidence about the involvement of phonology in a range of writing tasks (copy, spelling-to-dictation, associated-pairs) when adult writers produce known words. Interestingly, multiple phonological units seem to be functional during handwriting, as syllables and graphemes. Furthermore, we demonstrate that he frequency of a phoneme have an impact in the duration of its corresponding grapheme. Altogether, the evidence presented here also confirms the impact of central high-order (phonological) variables in the duration of a written response. This fact supports the idea that writing starts as soon as the initial segments of the response have been processed, and rules out the affirmation that a word does not begin to be produced until the whole word has been processed at the central (abstract) levels. Evidence obtained in these experiments strongly support the claim that phonology is systematically retrieved during the normal handwriting process, even when writing well-known words. Phonology is retrieved during handwriting, and it is used at a sublexical level to strengthen the orthographic (lexical) information kept in the orthographic working memory. Later on the process, the written response is produced syllable-by-syllable, indicating that the phonological loop plays a role in the response segmentation. In the case of low-frequency syllables, which lack of a holistic motor pattern, the motor programs for individual letters have to be retrieved. This process produces an increase of the cognitive load demanded by complex graphemes. More studies are necessary to establish how these effects vary when complex graphemes are embedded in high-frequency syllables. However, we have been able to confirm the involvement of at least two different units of phonological origin during handwriting production: syllables and graphemes. Multiple units seem to be used to program the writing movements, in line with Van Galen's (1991) proposal. Finally, this pattern of results fits a model of written production in which levels of processing are simultaneously active, but engaged with different segments of the response. When processing demands are increased at a certain level of processing, concurrent processes can be also affected. We propose a psycholinguistic model of handwriting production which integrates the evidence reported here and most of the previous literature.