Academic journal article Memory & Cognition

Impact of Phonology on the Generation of Handwritten Responses: Evidence from Picture-Word Interference Tasks

Academic journal article Memory & Cognition

Impact of Phonology on the Generation of Handwritten Responses: Evidence from Picture-Word Interference Tasks

Article excerpt

The degree to which phonological codes constrain handwriting is at present controversial. Two experiments used a picture-word interference paradigm in which participants wrote down the names of pictures while attempting to ignore visual distractor words presented at various time intervals (SOAs). Distractors could be orthographically and phonologically related, orthographically related only, or unrelated. We found an exclusive effect of phonology at an early SOA, and orthographic priming at a later SOA. In a second experiment, we showed that the effect of phonology was diminished when writers engaged in concurrent articulatory suppression. The results suggest a role of phonology in the generation of handwritten words that is to some extent dependent on situational circumstances.

Over the last few decades, much work has been carried out to investigate the processes and mechanisms underlying spoken word production. As a result, detailed computational accounts of speaking have been brought forward (e.g., Dell, Schwartz, Martin, Saffran, & Gagnon, 1997; Levelt, Roelofs, & Meyer, 1999). Relatively less work has been devoted to an understanding of written production. Much of the existing work has come from two streams of research. First, writing has been investigated as a special type of skilled motor behavior, and from such a perspective, the way in which graphemes are converted into overt written output (e.g., allographic selection, size control, muscular adjustments, etc.) has been explored in considerable detail (see, e.g., the framework proposed by van Galen, 1991). Second, several studies, mainly from a neuropsychological perspective, have investigated spelling tasks-that is, the conversion of auditory input into orthographic codes (see, e.g., Houghton & Zorzi, 2003, for a computational model of spelling). Given that handwriting, much like speaking, typically serves to express meaning, it could be argued that the generation of written codes should be studied in close parallel to spoken production.1 In the experiments below, we apply an experimental paradigm popular in the speech production field, and we use it to investigate a central theoretical issue in handwriting.

A theoretical account of handwriting must account for how orthographic codes are accessed from conceptual knowledge. A central debate in this field concerns the role of phonological codes in this process. Early theoretical accounts (e.g., Geschwind, 1969; Luria, 1970) characterized handwriting as entirely dependent on the prior retrieval of phonological codes. According to such "phonological mediation" theories, in order to write a word, one would first have to retrieve its phonological format (i.e., covertly name the word), and these sound-based codes would subsequently be converted into graphemic codes. Claims of this type are consistent not only with the common introspective experience of how written codes are generated (Hotopf, 1980), but also with the observation of phonologically conditioned spelling errors (e.g., homophone substitutions such as there for their, or quasi-homophone substitutions such as dirth for dearth; Aitchison & Todd, 1982). Some experimental evidence also supports the phonological mediation hypothesis, such as the finding that spelling errors generated by brain-impaired patients and normal writers occur more frequently on picture names with inconsistent rather than consistent spelling (Aitchison & Todd, 1982). This finding implies online competition between two orthographic codes for inconsistent items and hence provides evidence of phonological involvement in writing. Additionally, many neuropsychological patients with writing disorders exhibit comparable impairments in spoken and written language production (Basso, Taborelli, & Vignolo, 1978; Head, 1926; Hécaen & Angelergues, 1965; Luria, 1966), as the phonological mediation hypothesis predicts.

However, the phonological mediation account has largely fallen out of favor, because more recently, numerous neuropsychological studies have demonstrated dissociations between spoken and written production, such as in cases of acquired brain damage in which patients could, for instance, name a picture, but not write its name, or vice versa. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.