Is written language production more difficult than oral language production? Probably, yes. But why?

Several experiments were conducted in order to test the impact of low-level activities involved in writing on the performance of higher-level activities also involved in writing. Three assumptions were made: (1) the capacity of working-memory is limited, (2) every component of writing has a cognitive load, and (3) every increase in the load devoted to the activity of one component would lead to a decrease in the remaining resources available for the other components. These low-level activities are more resource-consuming in children than in adults because children have not yet automated these activities. So, it was hypothezised that the difficulties encountered by children in dealing with the low-level activities would have a negative impact on the performance of higher activities.

To test that hypothesis, a serial recall paradigm was used. Adults and children were asked to recall series of words, either orally or in writing. The results showed that: (1) serial recall of children, but not of adults, was weaker with writing as compared to speaking; (2) the slowness of writing relative to speaking did not explain these results; and (3) the difficulties of graphic and of orthographic transcription were partially responsible for these results.

These results are discussed in the general framework of production models.