These experiments examine two aspects of the automatic influences on m
emory as measured by target responding in the exclusion condition with
in the process dissociation framework, In Experiment 1, we examine the
extent to which congruency between study and test contexts affects au
tomatic processes in memory, In Experiment 2, we investigate qualitati
ve differences in consciously controlled and automatic processing as i
ndexed by metamemory judgements, In both experiments, a process-dissoc
iation procedure was used to separate automatic and consciously contro
lled uses of memory in a stem completion task, In the study phase of E
xperiment 1, subjects read a passage from one of two directed perspect
ives. The subsequent stem completion task, which subjects performed wh
ile mindful of the study perspective, contained (a) old words congruen
t with the directed perspective, (b) old words congruent with a differ
ent (non-directed) perspective, and (c) words that had not been presen
ted, Estimates of automatic influences for words congruent with the di
rected perspective were found to be greater than estimates for words i
ncongruent with the directed perspective, These results provide eviden
ce for the automatic or unconscious influences of meaning on task perf
ormance, which is uncontaminated by the influence of consciously contr
olled recollection which may occur in indirect memory tests, In Experi
ment 2, judgements of learning made prior to retrieval under inclusion
and exclusion instructions were found to be different for consciously
controlled and automatic processes, suggesting that memory as measure
d by the opposition (exclusion) procedure is involuntary and unconscio
us, with prospective monitoring of performance not sensitive to eventu
al performance. (C) 1997 Elsevier Science B.V.