Prospective Memory: A Neuropsychological Study

Neuropsychology
1999, Vol. 13, No. 1, 103-110
Copyright 1999 by the American Psychological Association, Inc.
0894-4105/99/S3.00
Prospective Memory: A Neuropsychological Study
Mark A. McDaniel
Elizabeth L. Glisky and Susan R. Rubin
University of New Mexico
University of Arizona
Melissa J. Guynn
Barbara C. Routhieaux
University of New Mexico
University of Arizona
To examine the neuropsychology of prospective remembering, older adults were divided
preexperimentally into 4 groups on the basis of their scores on 2 composite measures: one
assessing frontal lobe function and the other assessing medial temporal lobe function. The
groups reflected the factorial combination of high and low functioning for each neuropsychological system, and they were tested on an event-based laboratory prospective memory task.
High-functioning frontal participants showed better prospective remembering than lowfunctioning frontal participants. There was no significant difference in prospective memory
performance attributable to medial temporal functioning. The results support the theoretical
notion that frontal lobe processes play a key role in prospective remembering. Discussion
focuses on the particular components of prospective memory performance that frontal lobes
might mediate.
A kind of memory task that is pervasive in daily activities
is remembering to perform some intended action at a
particular point in the future (in keeping with the literature,
we label this prospective memory). Examples of this kind of
task are remembering to give a colleague a message when
you next see her or remembering to buy bread when you
pass the store on the way home. Despite the pervasiveness of
prospective memory in everyday life, there has been little
experimental research on this topic (see Brandimonte,
Einstein, & McDaniel, 1996), especially in contrast to the
plethora of work published on retrospective memory. In
terms of theoretical treatments of memory in general,
currently popular theoretical approaches incorporate considerations of neuropsychological systems that subserve various kinds of memory performance (e.g., see Schacter &
Tulving, 1994). Neuropsychological models of memory,
however, also fail to consider the processes and systems that
underlie prospective memory.
The purpose of this article is to sketch some preliminary
theoretical approaches concerning the neuropsychology of
prospective memory, and to present an experiment to inform
these approaches. In this article we focus on event-based
prospective remembering. Event-based prospective memory
tasks are those for which some environmental event signals
the appropriateness of the intended action (e.g., remembering to give a friend a message when you encounter the
friend; Einstein & McDaniel, 1990). By contrast, in timebased prospective memory tasks, a particular time or a
particular amount of elapsed time signals the appropriateness of the intended action (a 2:00 PM dentist appointment;
take cookies out of oven in 10 min). Time-based prospective
memory tasks have been associated with significant agerelated deficits and generally lower performance than observed in event-based tasks (Einstein, McDaniel, Richardson, Guynn, & Cunfer, 1995; Park, Hertzog, Kidder, Morrell,
& Mayhorn, 1997). Because the present approach required
exclusive testing of older adults, low performances with a
time-based prospective memory task could have diminished
the experimental sensitivity of influences of the neuropsychological systems discussed next (cf. McDaniel, Glisky,
Routhieaux, & Remers, 1994). Accordingly, this initial study
used event-based prospective memory as the experimental
task.
When speculations concerning the neuropsychology of
prospective memory are offered, albeit infrequently, the
typical notion is that prospective remembering depends
largely on prefrontal systems (cf. Bisiacchi, 1996; Burgess
& Shallice, 1997; Glisky, 1996; West, 1996). There are many
formal aspects to a prospective memory task that appear to
articulate with putative frontal lobe functions. For example,
in prospective memory tasks there is no external agent
prompting the person to initiate a search of memory.
Therefore, one hypothesis is that prospective memory
requires monitoring the environment for the event or marker
(or cue) that signals the appropriateness of the intended
activity (Bisiacchi, 1996; Burgess & Shallice, 1997). In one
neuropsychological model of cognition, these kinds of
processes are mediated by a "supervisory attentional sys-
Mark A. McDaniel and Melissa J. Guynn, Department of
Psychology, University of New Mexico; Elizabeth L. Glisky, Susan
R. Rubin, and Barbara C. Routhieaux, Department of Psychology,
University of Arizona.
This project was supported in part by National Institute on Aging
Grants AG05627 and AG09195. We thank Alysha Teed and Laurel
Remers for their assistance in testing participants, and we appreciate helpful comments from Art Shimamura on an earlier version of
this article. A subset of the data reported herein was presented at the
Cognitive Aging Conference, Atlanta, Georgia, April 1994.
Correspondence concerning this article should be addressed to
Mark A. McDaniel, Department of Psychology, University of New
Mexico, Albuquerque, New Mexico 87131. Electronic mail may be
sent to mcdaniel@unm.edu.
103
104
McDANIEL, GLISKY, RUBIN, GUYNN, AND ROUTHIEAUX
tern," a system intimately linked to frontal functioning
(Burgess & Shallice, 1997; Shallice & Burgess, 1991; see
also Cohen & O'Reilly, 1996). Further, some theorists have
suggested that once the general significance of the environmental event (cue) has been noticed, retrospective memory
is then probed to attempt retrieval of the intended action
(McDaniel, 1995; Einstein & McDaniel, 1996b; McDaniel
& Einstein, 1993). To the extent that such retrieval processes
are strategic and voluntary, they may involve prefrontal
systems (Moscovitch, 1994; Shimamura, Janowsky, & Squire,
1991). Retrieval of the intended action may also depend on
the activation level of the memory representation of the
intended action, with activation level related to planning
initiated at the time of intention formation (Mantyla, 1996).
Such planning processes have been linked to frontal lobe
structures (cf. Burgess & Shallice, 1997). Finally, having
activated the intended activity at the appropriate time, the
person must then interrupt or inhibit ongoing actions, and
organize and execute a sequence of several responses
(including the prospective memory response; Guynn, McDaniel, & Einstein, in press; McDaniel, Robinson-Riegler,
& Einstein, 1998). These kinds of processes are also
presumed to be supported by frontal structures (Petrides &
Milner, 1982; Shimamura, 1994).
In sum, there is a general theoretical basis on which to link
prospective remembering with prefrontal functioning. Up to
this point, however, there have been very few attempts to
empirically establish such linkages. To our knowledge the
only published work suggesting a relationship between
frontal functioning and prospective memory consists of
several single case studies. Cockburn (1995) described a
patient with bilateral frontal infarcts who displayed prospective memory failures. Shallice and Burgess (1991) tested 3
patients with evidence of focal frontal lobe damage on
planning tasks. The patients performed poorly relative to
normal participants on the planning tasks. To the extent that
the tasks involved some prospective memory components
(as argued by Burgess & Shallice, 1997), then these results
may implicate frontal involvement in prospective remembering. In a pilot study reported by Bisiacchi (1996; conducted
by Sgaramella, Zattin, Bisacchi, Verne, & Rago, 1993) 3
closed-head-injury patients with frontal lobe involvement
were tested on laboratory prospective memory tasks. The
findings were mixed, with 2 patients completely failing to
remember to perform the prospective memory activity and
the other patient remembering on 100% of the trials.
Moreover, when considered as a group, it is unclear whether
the average performance level obtained for the frontal
patients would be impaired relative to various control patient
groups with unimpaired frontal function. Thus, the idea that
prefrontal neuropsychological systems play a role in prospective memory remains virtually untested. The current study
was conducted to provide an initial experimental test of that
idea.
It should be noted that it is not a foregone conclusion that
prospective memory, especially the kind of event-based task
studied herein, would be primarily frontally based. Another
neuropsychological system that could play a primary role in
event-based prospective memory is the medial-temporal
system, including the hippocampus (for purposes of exposition we subsequently refer to this system as the hippocampal
system). For example, in Moscovitch's (1994) neuropsychological model of memory, the hippocampal system supports
conscious recollection through an external cue that automatically interacts with information that has been previously
associated with that cue. If the cue is good enough, then the
hippocampal system rapidly, obligatorily, and with few
cognitive resources delivers to consciousness the information associated with the cue. That is, the system responds
reflexively to cues—it cannot conduct a memory search if
cues are initially ineffective.
One hypothesis, then, is that prospective remembering is a
reflexive process that is either successfully stimulated by the
presence of the event or fails if the event is an ineffective cue
(McDaniel et al., 1998). This hypothesis is consistent with
participants' reports concerning their introspections (for
event-based laboratory prospective memory tasks like those
used in the present experiment) that the intended action
appears to "pop into mind" on presentation of the event
signaling that the action should be performed (Einstein &
McDaniel, 1990, 1996a). Thus, the idea here is that an
involuntary associative memory system (hippocampal) may
be more intimately involved in event-based prospective
memory than a system responsible for self-initiated retrieval
activity (frontal).
To inform the alternative theoretical views introduced
above, we selected older adults that varied in their performance on neuropsychological tests presumed to assess
hippocampal functioning and frontal functioning. Specifically, we sampled older adults whose performance on the
neuropsychological tests placed them in one of four cells
representing the factorial combination of high and low
hippocampal functioning crossed with high and low frontal
functioning. The participants were then given an eventbased laboratory prospective memory task patterned on that
used by Einstein et al. (1995). We reasoned that if this kind
of prospective memory task involved planful, strategic,
frontally associated retrieval processes (e.g., Bisiacchi,
1996), then high-frontal-functioning adults should perform
better than low-frontal-functioning adults and there should
be no modulating influence of hippocampal functioning.
That is, for both low- and high-hippocampal participants,
there should be significant enhancement of prospective
memory associated with high frontal functioning.
In contrast, on the view that event-based prospective
remembering is predominantly a hippocampally (associative
system) mediated process (Guynn et al., in press; McDaniel
et al., 1998), there should be main effects of the hippocampal
variable, with the high-hippocampal-functioning participants performing better than the low-hippocampal-functioning participants, but no effect of the frontal variable. Of
course, it is possible that the hippocampal and frontal
systems cooperate to support the processes involved in
prospective memory. If so, then an interaction between the
hippocampal and frontal variables could emerge, with a
more complex framework than those sketched above being
implicated.
One other feature of our experiment requires comment.
PROSPECTIVE MEMORY
As stated at the outset, the proposal that prefrontal systems
are involved in prospective memory is based on the perspective that there are many component processes in prospective
memory that are potentially related to frontal functioning
(cf. Bisiacchi, 1996). These include monitoring for the event
and identifying it as a significant signal, probing memory for
the intended action once the event has been identified, and
coordinating the interruption of the ongoing task along with
the execution of the prospective memory activity. As an
initial attempt to more precisely identify the particular
processes in prospective remembering that are associated
with prefrontal systems (if any), we manipulated the saliency of the target event. Past work has shown that making
the target event more salient significantly enhances prospective memory performance, presumably because a more
salient target event (hereafter termed the cue) would not be
as dependent on self-initiated monitoring and more directly
signals the significance of the event (Brandimonte & Passolunghi, 1994; Einstein & McDaniel, 1990; McDaniel &
Einstein, 1993). We reasoned that if frontal processes were
primarily involved in monitoring, then decrements in prospective memory for the low-frontal group relative to the
high-frontal group would be especially pronounced in the
low-salient cue condition (a Frontal X Cue Salience interaction). If, on the other hand, frontal-related decrements in
prospective memory were as robust for the high-salient as
for the low-salient cue condition, then the frontal effects
might more fruitfully be interpreted as reflecting problems in
probing memory once the cue has been identified as such, or
coordinating the execution of various activities at time of
retrieval, or both.
Method
Participants and Design
Forty-one older adults between the ages of 66 and 85 years
(M = 74 years) were selected from a larger pool of elderly research
participants at the University of Arizona's Amnesia and Cognition
Unit. All members of the pool are healthy, community-dwelling
volunteers without dementia, depression, or any history of neurological, psychiatric, or substance abuse problems. Those who
participated in this study were paid $5/hr. All of the individuals in
the pool have been categorized along two dimensions on the basis
of a factor analytic study of neuropsychological test performance
(Glisky, Polster, & Routhieaux, 1995). The analysis revealed two
independent factors: one based on tests of memory function, which
we have referred to as a Hippocampal factor, and the other based on
tests associated with frontal function, which we have labeled a
Frontal factor. Although we cannot be sure that the factors reflect
actual functioning of these particular brain regions, previous
studies have shown that they are predictive of performance on tasks
presumed to be associated with these structures (Glisky et al.,
1995). The Hippocampal factor is based on a composite measure
derived from Logical Memory I, Verbal Paired Associates I, and
Visual Paired Associates II from the Wechsler Memory Scale—
Revised (WMS-R; Wechsler, 1987) and Long-Delay Cued Recall
from the California Verbal Learning Test (Delis, Kramer, Kaplan,
& Ober, 1987). The Frontal factor is based on a composite measure
derived from the number of categories achieved on the modified
Wisconsin Card Sorting Test (Hart, Kwentus, Wade, & Taylor,
1988), the total number of words generated on the Controlled Oral
105
Word Association Test (Spreen & Benton, 1977), Mental Arithmetic from the Wechsler Adult Intelligence Scale—Revised
(WAIS-R; Wechsler, 1981), Mental Control and Backward Digit
Span from the WMS-R. The composite scores represent average z
scores relative to a normative population of 100 older adults from a
previous factor analytic study. Variance attributable to age was
removed prior to the factor analysis.
Of the 41 participants in the present study, 21 had z scores above
the mean and 20 had z scores below the mean on the frontal
composite measure. All z scores were greater than the absolute
value of .08, so that no participant was at or nearly at the mean for
either composite measure. Eleven of the participants in the
high-frontal group and 10 in the low-frontal group had hippocampal scores above the mean. The remainder of the participants in
each frontal group (10 from each group) had hippocampal scores
below the mean. Thus, there were 10 participants in three of the
cells formed by the combination of the two composite measures of
frontal lobe and hippocampal function and 11 participants in the
fourth cell. A separate two-factor (frontal, hippocampal) betweensubjects analysis of variance (ANOVA) for each composite measure confirmed that the high- and low-functioning frontal groups
differed significantly from each other with respect to the frontal
composite score, F( 1,37) = 117.71, MSB = 0.10, p< .05,but the
hippocampal groups did not differ on the frontal score (F < 1).
Similarly, the high- and low-functioning hippocampal groups
differed significantly on the hippocampal score, F(l, 37) = 87.39,
MSB = 0.16, p < .05, but the frontal groups did not differ on that
score (F < 1). Further, there were no significant interactions (see
Table 1).
Other characteristics of the four groups are also shown in Table
1. Between-subjects ANOVAs (2 X 2) on the demographic variables revealed no group differences in education or the MiniMental State Examination (Folstein, Folstein, & McHugh, 1975).
Age varied significantly across the cells, with the high-frontal-highhippocampal and low-frontal-low-hippocampal groups being
slightly older on average than the remaining two groups, F( 1, 37) =
5.96, MSB = 23.52, p < .05, for the interaction. No significant
group differences were observed in intelligence as assessed by
Performance IQ on the WAIS-R. The Performance IQ measure was
chosen because none of its subtests were included in the frontal or
hippocampal composite scores. In addition, because several of the
older adults were over the age of 74, the Mayo Older Americans
Normative Studies norms (Ivnik, Malec, & Smith, 1992) were used
to calculate the IQ scores for all of the sample. For Verbal IQ, the
Vocabulary subtest of the WAIS-R indicated that scores were
significantly lower for low-frontal than for high-frontal participants, F(l, 36) = 11.4, MSB = 56.51, p < .05, and for
low-hippocampal relative to high-hippocampal participants, F(l,
36) = 9.54, MSB = 56.5\,p < .05 (for 1 participant there was no
vocabulary score). These main effects were qualified by an
interaction showing that the low-frontal-low-hippocampal group
showed lower vocabulary scores than the other three groups, F( 1,
36) = 4.18, MSB = 56.51, p< .05.
Materials and Procedure
The ongoing task in this study was a multiple choice test of
general knowledge, facts, and trivia. Two different sets of 172
questions were selected from materials used in Einstein et al.
(1995, Experiment 3) and from popular trivia games. Half of the
participants received Set 1 and half received Set 2. Each question
consisted of one or two sentences followed by four alternative
choices. Participants indicated their answers by pressing one of
four keys on the computer keyboard (the / g, h, or j keys, which
106
McDANIEL, GLISKY, RUBIN, GUYNN, AND ROUTHIEAUX
Table 1
Mean Characteristics of Experimental Groups
Frontal functioning
Low
High
Participant
characteristics
High
hippocampal
Age
PIQ
Educ
Vocab
MMSE
Frontal" score
Hippocampala score
73.0
120.6
16.1
62.1
29.0
.45
.59
Low
hippocampal
High
hippocampal
Low
hippocampal
69.9
116.0
15.8
59.6
28.7
.63
-.78
69.1
113.8
15.0
58.9
28.3
-.47
.44
73.4
110.0
14.4
46.7
28.0
-.59
-.53
Note. PIQ = Performance IQ on the Wechsler Adult Intelligence Scale—Revised (Wechsler,
1981); Educ = years of education; Vocab = vocabulary; MMSE = Mini-Mental State Examination
(Folstein, Folstein, & McHugh, 1975).
"See text for explanation of this score.
were labeled a, b, c, d to correspond to the multiple choices).
Embedded in this task were eight target questions about presidents
in which the word president appeared (Set 1) or eight target
questions about states in which the word state appeared (Set 2). The
target questions appeared approximately every 5 min in the course
of the general knowledge task.
Salience of the target words was also manipulated: Targets were
either highlighted by using bold-faced type (i.e., made salient), or
they appeared normally (i.e., indistinguishable from other words in
terms of display format). Highlighting was blocked such that either
the first four or the last four targets were highlighted in counterbalanced order across question sets.
Participants received instructions concerning the general knowledge task and were told that from time to time they would see
questions about presidents (Set 1) or states (Set 2) and at that time
they should respond by pressing the F8 key on the computer. This
constituted the prospective memory task. After these instructions
were given and before the start of the general knowledge task,
participants were engaged in one of two retrospective memory
tasks (the tasks were varied for purposes not central for present
concerns). These retrospective memory tasks were included to
introduce a delay between the prospective memory instructions and
the general knowledge task (following Einstein & McDaniel,
1990). Half of the participants received a computerized version
(i.e., presented visually) of the Rey Auditory Verbal Learning Test
(Rey, 1964). This test consisted of the presentation of 15 unrelated
concrete nouns presented at the rate of 1 every 2 s, followed by an
immediate test of free recall. This study-test sequence was
repeated five times. Words were presented by computer and recall
was written. The test took approximately 12 min to complete. The
other half of the participants received a recognition memory test in
which 56 unrelated words were presented on the computer screen at
the rate of 1 every 2 s. Immediately after presentation, participants
were given a paper and pencil recognition test on which the 56
studied words were randomly mixed with 56 new words. Participants were instructed to circle all of the old items that they could
remember and to rate their degree of confidence in their decision
(1 = very confident, 2 = fairly confident, 3 = unsure, but I think
I've seen it before). This task required approximately 10 min. After
the recall or recognition task, participants began the general
knowledge multiple choice task. No additional instructions concerning the prospective memory task were provided at this time.
General knowledge questions were presented on the computer
one at a time for 12s, during which the participants were required
to read the question and make a keypress indicating their choice of
answer. Feedback was provided. It took approximately 40 min to
complete this task.
Results
The rejection level for all statistical tests was set at .05.
We first report the analyses associated with the question
answering task and the other retrospective memory tasks,
and these data are displayed in Table 2. The analyses of the
prospective memory data follow.
Retrospective Memory Tasks
The proportions of correct responses on the question
answering task were tabulated and submitted to a 2 X 2
between-subjects ANOVA, with frontal functioning (high,
low) and hippocampal functioning (high, low) as variables
(see Table 2 for the means). This analysis indicated that high
frontals answered significantly more questions correctly
Table 2
Question Answering, Recall, and Recognition Performance
Dependent measure and
hippocampal functioning
Frontal functioning
High
Low
.50
.47
.44
.36
.75
.68
.70
.62
.64
.24
.59
.47
.70
.62
.64
.54
.16
.31
.12
.14
3
Question answering
High
Low
2
Recall
High (n = 4)
Low (n = 6)
Recognition scoreb
High (n = 6)
Low (n — 4)
Hits
High
Low
FAs
High
Low
Note. A participant in the high frontal-high hippocampal group
was excluded from the recognition means because of 100% hits
coupled with 100% FAs. FAs = false alarms.
"Proportion correct. b(Hits - FAs)/l - FAs.
PROSPECTIVE MEMORY
than low frontals, F(l, 37) = 7.34, MSB = 112.56 (percentage scores were used in the ANOVA), and that high
hippocampals tended to answer more questions than low
hippocampals, F(l, 37) = 3.24, p < .08. There was no
significant interaction between the frontal and the hippocampal factors.
The recognition and recall tests were also scored. Recall
scores represent the proportion of correct responses. To
obtain a measure of recognition that considers both the hit
and false alarm (FA) rates we computed the following
recognition scores: (Hits - FAs)/l - FAs (Table 2 gives the
means for recall, recognition scores, hits, and FAs). Because
each test was administered to half of the participants, the
number of observations is small. Even so, a 2 X 2
between-subjects ANOVA of the recognition scores found
that high hippocampals performed significantly better than
low hippocampals, F(l, 16) = 4.67, MSB = .07, whereas
there was no significant effect of frontal functioning (F < 1)
nor an interaction. A 2 X 2 between-subjects ANOVA of the
recall data revealed no significant effects.
Prospective Memory
Table 3
Prospective Memory Performance as a Function of
Neuropsychological Condition and Cue Salience
High
Nonsalient
Salient
M
Low
Nonsalient
Salient
M
MSE = .17. There were no other significant effects in the
overall ANOVA.
Because frontal status significantly affected performance
on the task in which the prospective memory task was
embedded (question answering), it could be the case that the
effect of frontal status on prospective memory was due in
part to differential difficulty across groups in the ongoing
activity (cf. Einstein, Smith, McDaniel, & Shaw, 1997). To
statistically factor out any influence on prospective memory
of the question answering task per se, we computed an
analysis of covariance on the prospective memory scores
with question answering performance as the covariate and
frontal and hippocampal status as independent variables.
The adjusted means still showed diminished performance
for the low-frontal (.44) relative to the high-frontal group
(.66), a difference that was marginally significant, F(l, 36) =
3.16, MSE = .13, p < .09. In contrast, there was little
variation in prospective memory as a function of the
hippocampal variable (F < 1; low-hippocampal = .51 and
high-hippocampal = .59).
Discussion
For the high-salient cue and the low-salient cue conditions, we computed the proportion of times that each
participant remembered to press the response key out of four
possible tries (yielding two scores per participant). Table 3
provides the means. These scores were included in a 2 X
2 X 2 mixed ANOVA, with the frontal and hippocampal
status as between-subjects variables and cue salience as the
within-subject variable. There was a significant main effect
of frontal status, with the high-frontal group (M = .72)
performing significantly better than the low-frontal group
(M = .38), F(l, 37) = 8.20, MSB = .29. Though the
high-hippocampal group showed a nominal advantage over
the low-hippocampal group (.63 vs. 46, respectively), this
difference was not significant, F(l, 37) = 1.94,p < .20.
A significant main effect of cue salience emerged, such
that high-salient cues produced better prospective memory
than low-salient cues, F(l, 37) = 12.12, MSE = .05. Cue
salience did not interact with frontal status (F < 1), and
comparisons confirmed that frontal status significantly affected prospective memory for highly salient cues, F(\, 37) =
7.78, MSE = .18, and less salient cues, F(l, 37) = 6.13,
Hippocampal functioning
and cue salience
107
Frontal functioning
High
Low
.70
.91
.81
.38
.52
.45
.52
.72
.62
.22
.38
.30
This experiment provides initial evidence regarding the
neuropsychological systems that are involved in prospective
remembering. Older adults with lower than average scores
on neuropsychological tests associated with frontal functioning showed diminished prospective memory performance
relative to older adults with higher than average scores on
frontal tests. This pattern is in line with theoretical speculation that prefrontal systems subserve significant processes in
prospective memory. The findings are not as clear-cut with
regard to hippocampal involvement in prospective memory.
Older adults with low scores on neuropsychological tests
implicating hippocampal functioning showed decreased prospective memory performance compared to older adults with
high scores on the hippocampal tests, but this difference was
not statistically significant. We first consider the implications of the differences between the low- and high-frontal
groups, and then discuss the outcome associated with the
hippocampal variable.
The present support for prefrontal involvement in prospective memory is indirect, as it rests on the assumption that the
factor scores are reflective of the functioning of the corresponding brain regions. Though the tests that comprised the
frontal factor have been found to be sensitive to frontal
dysfunction (cf. Glisky et al., 1995), it is possible that their
common variance is associated with cognitive processes that
are partly or wholly dependent on other unspecified brain
regions. Though this possibility cannot be categorically
rejected, there is cause to consider the neuropsychologically
defined frontal variable to be a reasonable first approximation to illuminating frontal function. In this regard, most of
the present participants (78%) were included in two studies
that examined item and source memory for sentential
material (Glisky et al., 1995, and an unpublished study). The
results consistently showed a dissociation such that the
participants with the low-frontal factor scores demonstrated
significant decline in source memory but not item memory
108
McDANIEL, GLISKY, RUBIN, GUYNN, AND ROUTHIEAUX
relative to the participants with high-frontal factor scores.
Because source memory but not item memory is implicated
in frontal functioning (Janowsky, Shimamura, & Squire,
1989; Shimamura & Squire, 1987), this pattern converges
with the present assumption that the Frontal factor as
presently denned was reflective of frontal lobe function.
Given the theoretical consensus that frontal systems play
a role in prospective remembering (Bisiacchi, 1996; Glisky,
1996; Guynn, McDaniel, & Einstein, in press) and in light of
the present evidence converging on that consensus, some
speculation concerning the role of frontal systems in prospective memory seems warranted. One theoretical view of
prospective memory is that it involves processes that
monitor the environment for "markers" that signal the
appropriateness of performing the intended action (cf.
Bisiacchi, 1996; Burgess & Shallice, 1997). Frontal lobes
might be involved in these putative monitoring processes. If
so, then one might expect a greater degree of prospective
memory decrements for low-frontal-functioning individuals
with less salient cues (markers) than for salient cues. The
reasoning here is that less salient cues would demand more
monitoring and so should be more likely to show decrements
if monitoring were compromised. This logic is supported by
findings of Einstein, McDaniel, Manzi, and Cochran (reported by McDaniel, Guynn, & Einstein, 1997), in which a
divided attention task (which should hamper monitoring)
produced declines in prospective memory for nonsalient
cues but not for salient cues (with saliency manipulated by
means of perceptual features, as in the present experiment).
In opposition to the above expectation, the deficits in
prospective memory for low-frontal participants were as
robust for the salient cues as for the nonsalient cues. Note
that the absence of an interaction between frontal functioning and cue salience is not due to a weak manipulation of cue
salience; high-salient cues increased prospective memory
39% over low-salient cues. Thus, it may be that frontal lobes
are not closely tied to the kinds of monitoring processes
required for prospective memory (if such processes are
indeed involved), at least for the present range of frontal
functioning examined or for the particular locus of dysfunction reflected by the present neuropsychological assessment.
Probing memory for cue significance, the intended action,
or both may require strategic retrieval processes, processes
that have been linked to frontal lobes (Moscovitch, 1994;
Shimamura, 1994). The idea here is that low-frontal participants in the present experiment may have displayed reduced
prospective memory performance because they were compromised in their ability to retrieve the requisite information
concerning the cue. This interpretation is consonant with the
finding that low-frontal participants showed a significant
decline in retrieving the correct information to the general
knowledge questions as well.
Countervening the above interpretation is the observation
that the retrieval of the intended activity (once the cue is
noticed) should not be particularly difficult. In line with this
claim, in another study with a similar kind of intended
(prospective memory) action (press an F# key on the
keyboard) paired with a highly salient cue, prospective
memory performance was nearly perfect (over 90%) for
older adult participants (McDaniel et al., 1997). This suggests that given that the cue is noticed, the intended action is
readily retrieved. Further support for this assertion comes
from a divided attention condition in which a secondary task
was added to the ongoing activity (reading text passages)
during the prospective memory retrieval phase. There was
no decline in performance for older adults in the high salient
cue condition, again suggesting that once the cue is noticed
very little strategic processing is needed for retrieving the
intended action.
These considerations suggest another formulation. Retrieval of the intended action could depend in part on the
level of resting activation of the intended action (Mantyla,
1996; see also Marsh, Hicks, & Bink, 1998) that would
influence the success of an associative retrieval process
triggered by the prospective memory cue (Guynn et al., in
press; McDaniel & Einstein, 1993; McDaniel et al., 1998).
The level of resting activation would be a function of the
quality of the initial encoding or planning processes (Mantyla, 1996), and this encoding process has been posited to be
supported by frontal subsystems (Burgess & Shallice, 1997).
The present finding of improved prospective remembering
with increased frontal functioning (as assessed by the
neuropsychological tests) for both low and high-salient cue
conditions is consistent with this idea that frontal processes
might be involved in forming a coherent, highly activated
encoding of the anticipated prospective memory cue and the
intended action.
Frontal processes may be implicated in one final component of performing a prospective memory task. Once the
intended action has been retrieved, successful prospective
memory performance requires that participants maintain
ongoing goals in working memory (associated with the
ongoing activity as well as with the intended action), and
interrupt and initiate execution of the intended prospective
memory action (cf. Guynn et al., in press; McDaniel et al.,
1998). Neuropsychological studies using tasks that require
participants to organize a sequence of responses to be
executed show a severe and across the board (e.g., using
different stimulus materials) deficit for patients with frontal
lobe excisions (see Petrides & Milner, 1982). Accordingly,
the present decline in prospective memory performance due
to low frontal functioning may reflect deficits in the executive (or working memory) processes that are required for
interrupting ongoing activity and scheduling the execution
of the prospective memory task.
More specifically, for low-frontal participants the demands of processing and attempting to answer the general
knowledge-trivia questions could completely capture their
attention such that even if the intended action were retrieved,
these participants would not have the available inhibitory
mechanisms to interrupt the question answering activity and
switch resources to execute the competing activity. This
interpretation of frontal involvement in prospective memory
coincides with Shallice and Burgess's (1991) finding that
patients with frontal lesions are significantly impaired in
inhibiting performance on one activity in order to initiate
another intended activity within a certain time frame.
The impact of the question answering task is potentially
PROSPECTIVE MEMORY
more complicated. Because low-frontal participants did not
perform as well on this task as the high-frontal participants,
it is possible that question answering posed more difficulty
or processing demands to low-frontal participants. If so, then
the effects of frontal functioning on prospective memory
could be due to differential difficulty of the ongoing activity
across the low- and high-frontal groups (cf. Einstein et al.,
1997). Although this possibility cannot be completely ruled
out, when performance on the question answering task was
statistically factored out there remained an advantage in
prospective memory for high- versus low-frontal participants.
Another feature of the experiment raises an additional
potential explanation of the frontal effect on prospective
remembering. Because the event-based cues that signalled
the appropriateness of the intended action occurred regularly
at 5-min intervals, participants could have used time cues to
perform the prospective memory task. If so, then the reduced
prospective remembering for the low-frontal participants
might be related to difficulties in time monitoring associated
with low frontal functioning.1 Clearly, more evidence would
be needed to compel this interpretation that participants
relied on time cues rather than the target cues and that time
perception is mediated by frontal systems.
Finally, it is worth noting that there was a 17% advantage
in prospective remembering for high-hippocampal relative
to low-hippocampal participants. Though this difference was
not statistically significant, the magnitude of the difference
favoring the high-hippocampal participants approximated
that reflected in the significant cue-salience effect (a withinsubjects manipulation). The absence of statistical significance for the hippocampal effect may represent some lack of
precision associated with the neuropsychological groupings.
Thus, it remains possible that the nominal prospective
memory advantage in the high hippocampal group reflects
some hippocampal involvement in prospective remembering. For instance, the hippocampal system could be involved
in retrieval of the intended action upon encounter of the
prospective memory cue (Guynn et al., in press; McDaniel et
al., 1998). This possibility is consistent with general views
that assume that hippocampal systems play a role in
associative memory retrieval (Moscovitch, 1994), and suggests that even more robust hippocampal effects might be
observed for prospective memory tasks in which the retrieval component is more challenging (i.e., having to
remember different responses to different cues; e.g., Robinson-Riegler, 1994).
In sum, the present experiment represents the first largescale study (using more than several participants) of which
we are aware to examine the influence of frontal systems and
hippocampal systems on prospective remembering. The
findings support the hypothesis proposed by several researchers that frontal functioning should play a key role in
prospective memory performance (e.g., Bisiacchi, 1996;
Burgess & Shallice, 1997; Glisky, 1996; Guynn et al., in
press; West, 1996). The particular components of prospective memory that are supported by frontal mechanisms still
need to be identified, as well as the particular frontal
subsystems that articulate with prospective memory performance. The current results seem most consistent with the
109
idea that executive functions served by frontal processes
associated with the current neuropsychological assessment
procedures are involved in prospective memory performance. These executive functions might include encodingplanning the prospective memory task and inhibiting ongoing activity and organizing the execution of the intended
activity. Convergent with this idea, planning and organizational activities have been related to mid-dorsal lateral
frontal activity (Petrides, 1995), a region also thought to be
reflected in the neuropsychological frontal tests used here
(Milner, 1963). The data also suggest, though less strongly,
that hippocampal functioning influences prospective memory,
perhaps in the service of retrieving the intended activity
during presentation of the target event.
1
We thank a reviewer for raising this possibility.
References
Bisiacchi, P. S. (1996). The neuropsychological approach in the
study of prospective memory. In M. Brandimonte, G. O.
Einstein, & M. A. McDaniel (Eds.), Prospective memory:
Theory and applications (pp. 297-318). Mahwah, NJ: Erlbaum.
Brandimonte, M., Einstein, G. O., & McDaniel, M. A. (Eds.).
(1996). Prospective memory: Theory and applications. Hillsdale,
NJ: Erlbaum.
Brandimonte, M. A., & Passolunghi, M. C. (1994). The effect of
cue-familiarity, cue-distinctiveness, and retention interval on
prospective remembering. Quarterly Journal of Experimental
Psychology: Human Experimental Psychology, 47(A), 656-687.
Burgess, P. W., & Shallice, T. (1997). The relationship between
prospective and retrospective memory: Neuropsychological evidence. In M. A. Conway (Ed.), Cognitive models of memory (pp.
247-272). Cambridge, MA: MIT Press.
Cockburn, J. (1995). Task interruption in prospective memory: A
frontal lobe function? Cortex, 31, 87-97.
Cohen, J. D., & O'Reilly, R. C. (1996). A preliminary theory of the
interactions between prefrontal cortex and hippocampus that
contribute to planning and prospective memory. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective
memory: Theory and applications (pp. 267-295). Mahwah, NJ:
Erlbaum.
Delis, D. C., Kramer, J., Kaplan, E., & Ober, B. A. (1987). The
California Verbal Learning Test. San Antonio, TX: The Psychological Corporation.
Einstein, G. O., & McDaniel, M. A. (1990). Normal aging and
prospective memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 16, 717-726.
Einstein, G. O., & McDaniel, M. A. (1996a). Prospective memory:
Remembering a forgotten topic. In D. Hermann, C. McEvoy, C.
Hertzog, P. Hertel, & M. Johnson (Eds.), Basic and applied
memory research: Vol. 2. Practical applications (pp. 79—94).
Mahwah, NJ: Erlbaum.
Einstein, G. O., & McDaniel, M. A. (1996b). Retrieval processes in
prospective memory: Theoretical approaches and some new
empirical findings. In M. Brandimonte, G. O. Einstein, & M. A.
McDaniel (Eds.), Prospective memory: Theory and applications
(pp. 115-142). Hillsdale, NJ: Erlbaum.
Einstein, G. O., McDaniel, M. A., Richardson, S. L., Guynn, M. J.,
& Cunfer, A. R. (1995). Aging and prospective memory:
Examining the influence of self-initiated retrieval processes.
110
McDANIEL, GLISKY, RUBIN, GUYNN, AND ROUTHIEAUX
Journal of Experimental Psychology: Learning, Memory, and
Cognition, 21, 996-1007.
Einstein, G. O., Smith, R. E., McDaniel, M. A., & Shaw, P. (1997).
Aging and prospective memory: The influence of increased task
demands at encoding and retrieval. Psychology and Aging, 12,
479^88.
Folstein, M. E, Folstein, S. E, & McHugh, P. R. (1975). MiniMental State: A practical method for grading the cognitive state
of patients for the clinician. Journal of Psychiatric Research, 12,
189-198.
Glisky, E. (1996). Prospective memory and frontal lobes. In M.
Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.),
Prospective memory: Theory and applications (pp. 249-266).
Hillsdale, NJ: Erlbaum.
Glisky, E., Polster, M. R., & Routhieaux, B. C. (1995). Double
dissociation between item and source memory. Neuropsychology, 9, 229-235.
Guynn, M. J., McDaniel, M. A., & Einstein, G. O. (in press).
Remembering to perform actions: A different type of memory?
In H. D. Zimmer & R. L. Cohen (Eds.), Memory for action: A
distinct form of episodic memory.
Hart, R. P., Kwentus, J. A., Wade, J. B., & Taylor, J. R. (1988).
Modified Wisconsin Card Sorting Test in elderly normal, depressed and demented patients. Clinical Neuropsychologist, 2,
49-56.
Ivnik, R. J., Malec, J. E, & Smith, G. E. (1992). Mayo's Older
Americans Normative Studies. WAIS-R norms for ages 56-97.
Clinical Neuropsychologist, 6, 1-30.
Janowsky, J. S., Shimamura, A. P., & Squire, L. R. (1989). Source
memory impairment in patients with frontal lobe lesions.
Neuropsychologia, 27, 1043-1056.
Mantyla, T. (1996). Activating actions and interrupting intentions:
Mechanisms of retrieval sensitization in prospective memory. In
M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.),
Prospective memory: Theory and applications (pp. 93-113).
Hillsdale, NJ: Erlbaum.
Marsh, R. L., Hicks, J. L., & Bink, M. L. (1998). The activation of
completed, uncompleted, and partially completed intentions.
Journal of Experimental Psychology: Learning, Memory, and
Cognition, 24, 350-361.
McDaniel, M. A. (1995). Prospective memory: Progress and
processes. In D. L. Medin (Ed.), The psychology of learning and
motivation (Vol. 33, pp. 191-221). San Diego, CA: Academic
Press.
McDaniel, M. A., & Einstein, G. O. (1993). The importance of cue
familiarity and cue distinctiveness in prospective memory.
Memory, 1, 23^1.
McDaniel, M. A., Glisky, E. L., Routhieaux, B. C., & Remers, L. A.
(1994, April). The role of frontal lobe functioning in temporallybased and cue-based prospective remembering in older adults.
Paper presented at the Cognitive Aging Conference, Atlanta, GA.
McDaniel, M. A., Guynn, M. J., & Einstein, G. O. (1997,
November). Prospective memory: Two views and some new
data. Paper presented at the 38th Annual Meeting of the
Psychonomic Society, Philadelphia.
McDaniel, M. A., Robinson-Riegler, B., & Einstein, G. O. (1998).
Prospective remembering: Perceptually driven or conceptuallydriven processes? Memory & Cognition, 26, 121-134.
Milner, B. (1963). Effects of different brain lesions on card sorting.
Archives of Neurology, 9, 90-100.
Moscovitch, M. (1994). Memory and working with memory:
Evaluation of a component process model and comparisons with
other models. In D. L. Schacter & E. Tulving (Eds.), Memory
systems 94 (pp. 269-310). Cambridge, MA: MIT Press.
Park, D. C., Hertzog, C., Kidder, D. P., Morrell, R. W, & Mayhorn,
C. B. (1997). Effect of age on event-based and time-based
prospective memory. Psychology and Aging, 12, 314-327.
Petrides, M. (1995). Functional organization of the human frontal
cortex for mnemonic processing: Evidence from neuroimaging
studies. Annals of the New York Academy of Sciences (pp.
85-96).
Petrides, M., & Milner, B. (1982). Deficits on subject-ordered tasks
after frontal- and temporal-lobe lesions in man. Neuropsychologia, 20, 249-262.
Rey, A. (1964). L'Examen Clinique en Psychologic [The Clinical
Examination in Psychology]. Paris: Press Universitaire de
France.
Robinson-Riegler, M. B. (1994). The recognition-recall hypothesis
of prospective memory. Unpublished doctoral dissertation, Purdue University.
Schacter, D. L., & Tulving, E. (Eds.). (1994). Memory systems 94.
Cambridge, MA: MIT Press.
Shallice, T, & Burgess, P. W. (1991). Deficits in strategy application following frontal lobe damage in man. Brain, 114, 727-741.
Shimamura, A. P. (1994). Memory and frontal lobe function. In
M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp. SOSSIS). Cambridge, MA: MIT Press.
Shimamura, A. P., Janowsky, J. S., & Squire, L. R. (1991). What is
the role of frontal lobe damage in memory disorders? In H. S.
Levin, H. M. Eisenberg, & A. L. Benton (Eds.), Frontal lobe
functioning and dysfunction (pp. 173-195). Oxford, England:
Oxford University Press.
Shimamura, A. P., & Squire, L. R. (1987). A neuropsychological
study of fact memory and source amnesia. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 464—473.
Spreen, O., & Benton, A. L. (1977). Neurosensory Center Comprehensive Examination for Aphasia (NCCEA) (Revised edition).
Victoria, British Columbia, Canada: University of Victoria
Neuropsychology Laboratory.
Wechsler, D. (1981). Wechsler Adult Intelligence Scale—Revised
manual. San Antonio, TX: The Psychological Corporation.
Wechsler, D. (1987). Wechsler Memory Scale—Revised manual.
San Antonio, TX: The Psychological Corporation.
West, R. L. (1996). An application of prefrontal cortex function
theory to cognitive aging. Psychology Bulletin, 120, 272-292.
Received January 14, 1998
Revision received April 27, 1998
Accepted April 28, 1998