Four groups of adults learned letters of the American manual alphabet with
augmented information (demonstrations) provided by a model. The model was p
resented either before (proactive) or after (retroactive) attempted perform
ances of the letter handshapes, thereby providing different learning contex
ts. Acquisition performance was enhanced when the demonstrations were prese
nted immediately prior to performance, and degraded when the demonstrations
followed performance. Same day and delayed (48 h) retention tests without
augmented information revealed a large drop in performance for the proactiv
e groups and a slight improvement in performance for the retroactive groups
. These findings are compared to related effects which suggest that augment
ed information has both negative and positive influences on learning. (C) 1
999 Elsevier Science B.V. All rights reserved.