Four experiments were conducted to explore the possible involvement of expl
icit memory in an indirect memory test in which white noise accompanying ol
d sentences was judged to be quieter than white noise accompanying new sent
ences (Tacoby, Allan, Collins, & Larwill, 1988). Experiment 1 established t
hat this effect lasted up to 1 week. Experiment 2 found that a group of amn
esic patients showed a noise effect that was marginally above chance and no
t significantly less than that of their matched controls after a delay of o
ne day. Effects of time pressure at test (Experiment 3) and divided attenti
on at study (Experiment 4) suggested that the memory processes mediating th
e noise effect were not automatic, although the possibility that the proces
ses involve enhanced fluency is also discussed.