Written object naming, spelling to dictation, and immediate copying: Different tasks, different pathways?

Abstract

We report an investigation of cross-task comparisons of handwritten latencies in written object naming, spelling to dictation, and immediate copying. In three separate sessions, adults had to write down a list of concrete nouns from their corresponding pictures (written naming), from their spoken (spelling to dictation) and from their visual presentation (immediate copying). Linear mixed models without random slopes were performed on the latencies in order to study and compare within-task fixed effects. By-participants random slopes were then included to investigate individual differences within and across tasks. Overall, the findings suggest that written naming, spelling to dictation, and copying all involve a lexical pathway, but that written naming relies on this pathway more than the other two tasks do. Only spelling to dictation strongly involves a nonlexical pathway. Finally, the analyses performed at the level of participants indicate that, depending on the type of task, the slower participants are more or less influenced by certain psycholinguistic variables.

Keywords:

• Written naming
• Spelling to dictation
• Copying
• Individual differences
• Word frequency
• Phonology-to-orthography consistency

The authors wish to thank Mélanie Provost for her help in the collection of the data, and Marc Brysbaert and two anonymous reviewers for their constructive comments on a previous version of the paper.

This work was supported by a grant from the Institut Universitaire de France given to the first author.

Notes

¹The possibility of nonsemantically mediated lexical routes—that is, “direct” lexical routes—has been put forward in the literature (see Hillis, 2001) for both spelling to dictation (a direct link from the auditory word recognition system to the phonological output lexicon) and copying (a direct link from visual word recognition system to the orthographic output lexicon). It is also possible that visually presented stimuli might be copied by means of visual processes only, without the involvement of visual word recognition or phonological conversion. However, to avoid overcomplicating the model in Figure 1, we did not include these links. Finally, since to date there is no evidence in support of the existence of lexical links between the phonological and orthographic output lexicons (Bonin, Peereman, & Fayol, 2001), these bidirectional links are also not included in Figure 1.

²This elimination procedure was used in order to be able to use a similar pool of items, providing a sufficient numbers of observations, across all analyses. The reasons for excluding these 29 items were as follows: (a) There were technical errors for four words; (b) There were difficulties in naming or recognizing the pictures for 12 items; and (c) There were plausible phonological or orthographic errors for the remaining 13 items. Additional analyses showed that, on both imageability and cumulative frequency, the values for the 12 items that were difficult to name or to recognize were reliably lower than those for the remaining items (271) that were retained for analysis.

³These were: name agreement, image agreement, imageability, number of letters, cumulative frequency, and frequency trajectory in written naming; acoustic duration, initial PO consistency, and cumulative frequency in spelling to dictation; and number of letters and cumulative frequency in copying.

⁴Only independent variables for which there were significant random effects were kept in the models.

⁵We thank an anonymous reviewer for pointing this out to us.