The present experiment was carried out to test the hypothesis that the use
of the written mode increases the working memory load. Second and fourth gr
aders were orally presented with series of unrelated words which they were
required to recall in sequence. Each subject had to recall five different l
ists in the following conditions: oral alone, written, oral with a concurre
nt task (tapping, categorization, drawing). Participants recalled more word
s in the oral condition than in either the written mode or the `oral and ca
tegorization' conditions. Moreover, second graders performed better in the
oral mode than in the `oral and drawing condition'. This trend was not sign
ificant with older children. Finally, the tapping task did not affect child
ren's performance. The results are consistent with our cognitive load hypot
hesis.