Notes For PM Chapter - University Of Warwick

Copy and paste this link to your website, so they can see this document directly without any plugins.



Keywords

that, with, task, from, were, performance, their, memory, Commentary, older, more, ongoing, this, adults, data, prospective, they, tasks, than, which, participants, young, (this, there, deficits, study, McDaniel, would, have, aging

Transcript

Commentary 1
Running head: PROSPECTIVE MEMORY THROUGH THE AGES
Maylor, E. A. (2008). Commentary: Prospective memory through the ages. In M.
Kliegel, M. A. McDaniel, & G. O. Einstein (Eds.), Prospective memory: Cognitive,
neuroscience, developmental, and applied perspectives (pp. 217-233). New York:
Erlbaum.
Commentary: Prospective Memory Through the Ages
Elizabeth A. Maylor
University of Warwick, UK
Address for correspondence
Elizabeth A. Maylor
Department of Psychology
University of Warwick
Coventry CV4 7AL
UK
e.a.maylor@warwick.ac.uk
Acknowledgments
Thanks to Richard Darby, Victoria Holman and Joanne Rodger for help with data
collection and to Stian Reimers for useful discussion.
Commentary 2
Commentary: Prospective Memory Through the Ages
The number of published articles on prospective memory has increased
markedly over the past 15 years (see Figure 1 of Marsh, Cook, & Hicks, 2006). A
recent Web of Science search revealed that almost 40% of 421 articles that included
the term “prospective memory” (PM) were concerned with some aspect of age,
whether childhood or adult age (the majority on the latter). My task here is to
summarise, comment upon, and discuss general themes and unresolved issues arising
from, the chapters in this volume on development (Kvavilashvili, Kyle, & Messer),
aging (Phillips, Henry, & Martin; McDaniel, Einstein, & Rendell), and the lifespan
(Kliegel, Mackinlay, & Jäger). The composition of this section reflects not only the
continued interest (and controversy) surrounding PM across adulthood (which was
similarly present in the previous volume; Brandimonte, Einstein, & McDaniel, 1996)
but also the growing evidence on the early development of PM and the need to take an
integrative approach to understand the processes and mechanisms that drive change in
PM across the lifespan (cf. Bialystok & Craik, 2006; Graf & Ohta, 2002).
The Development of PM
Kvavilashvili et al.‟s (this volume) comprehensive review of the development
of PM in children begins by noting the paucity of studies in this area (an unusual case
of development lagging aging research). They attribute this to the (mistaken, they
argue) assumption that developmental work is unlikely to tell us anything new about
PM and also to the undoubted challenges in collecting PM data from young children.
(To these explanations, one could perhaps add that it may be easier to justify the study
of aging PM to funding bodies because of the obvious importance of everyday PM
tasks, such as remembering to take medication and pay bills on time, to living
independently in old age.) Nevertheless, there clearly has been some success in
Commentary 3
designing PM paradigms suitable for children to address these methodological issues,
many of which have been encountered previously in the aging literature (see Maylor,
1993b, 1996b; Uttl, 2005, for summaries). For example, it seems to me that the failure
of the “feeding the dog” scenario described by Kvavilashvili et al. illustrates the need
to avoid the PM requirement becoming a vigilance or monitoring task in the sense that
it occupies working memory or conscious awareness throughout the retention interval
(see Graf & Uttl, 2001, on the distinction between PM “proper” and
vigilance/monitoring).
Also in common with the aging literature, it seems that the data from
developmental studies of PM are somewhat inconsistent, which Kvavilashvili et al.
(this volume) attribute in part to different policies on whether or not ongoing task
difficulty should be adjusted to match the demands on younger and older children.
However, the emerging picture appears to be of quite well developed PM in preschoolers, with relatively modest improvement with increasing age thereafter. This
contrasts with much stronger developmental trends for retrospective memory (RM),
but I would argue that such a comparison raises the question of the reliability of PM
measures in children. If reliability is low, this limits the amount of systematic
variance available to be associated with age. Therefore, a priority for future research
should be a study, along the lines of Salthouse, Berish and Siedlecki‟s (2004)
investigation of PM across adulthood, in which children of different ages are
administered multiple PM and RM tasks (and other measures, such as executive
functioning) to first establish their construct validity and only then to compare their
developmental sensitivity.
In their concluding remarks, Kvavilashvili et al. (this volume) suggest that
further insights might be gained from applying current theoretical models of PM to
Commentary 4
development, in particular, McDaniel and Einstein‟s (2000) multiprocess framework.
This would predict stronger developmental trends for PM tasks requiring more
strategic resource-demanding monitoring but weaker developmental trends for PM
tasks relying on more automatic processing (for similar arguments with respect to
aging, see McDaniel et al., this volume). Preliminary evidence consistent with this
framework as applied to development comes from a study that was briefly described
in Maylor, Darby, Logie, Della Sala and Smith (2002). Children aged 6-11 years (n =
200) were presented with a series of photographs of their teachers and the children
were asked to name each of them (ongoing task). In addition, they were to indicate if
the teacher was wearing glasses or if there was a plant in the picture (PM task).
Although the glasses were visually much less prominent than the plant, the glasses
occurred within the focus of attention for the ongoing task whereas the plant occurred
outside the focus of attention (cf. Hicks, Cook, & Marsh, 2005). Responding to the
glasses as the PM cue was therefore assumed to be more dependent on automatic
processes and less dependent on strategic monitoring processes than responding to the
plant.
Almost every child successfully named both the PM target teachers (who were
deliberately chosen to be the most familiar to the children). As can be seen in Figure
1, there was general improvement in PM performance with increasing age but this
was less striking for the glasses than for the plant, consistent with the predictions from
the multiprocess framework. McGann, Defeyter, Reid and Ellis (2005) reported
similar trends in a study of 4-7 year olds in which PM target salience was manipulated
by increasing the size of the PM stimulus relative to the non-PM stimuli. Clearly, as
Kvavilashvili et al. (this volume) point out, this represents a promising approach for
future developmental investigations.
Commentary 5
Kvavilashvili et al. (this volume) suggest that whereas in laboratory-based
studies PM performance generally improves with development, in a naturalistic study
of children who were asked to remind their caregivers of various everyday tasks, there
was no improvement between the ages of 2 and 4 years and PM success rate was high
throughout (Somerville, Wellman, & Cultice, 1983). Thus there is an intriguing hint
of a developmental PM paradox (though not a complete cross-over of developmental
effects inside vs. outside the laboratory) that deserves replication and further
investigation. But it would surely be surprising to find younger children performing
better than older children on PM tasks outside the laboratory. At least in this respect,
development is not the mirror-image of aging, for which there is substantial evidence
of an age PM paradox such that young adults outperform older adults on laboratorybased PM tasks, whereas exactly the reverse is the case outside the laboratory (Henry,
MacLeod, Phillips, & Crawford, 2004; Rendell & Thomson, 1999). However, a
complete explanation for the age PM paradox remains surprisingly elusive.
The Age PM Paradox
Phillips et al. (this volume) provide a critical examination of factors that have
commonly been held responsible for the age PM paradox and in particular for the
unexpected positive effects of old age on PM performance in naturalistic studies.
These data have been too readily dismissed in the past largely because of a lack of
experimental control over participants‟ use of memory aids, ongoing activities, and so
on. Note, for example, the comment that “although the study of behavior in context is
important, it cannot be a replacement for the systematic study of behavior under
laboratory conditions” (p. 184, Maylor 1996b). However, Phillips et al. argue
convincingly that it is still important to understand fully the reasons behind the often
Commentary 6
superior PM performance of older adults outside the laboratory and they helpfully
outline the missing crucial studies required to test them.
It seems clear to me that older adults are more highly motivated to succeed in
naturalistic PM tasks like phoning the experimenter than are young adults, though
why is less clear (but note that the majority of young adults in the naturalistic studies
of PM included in Henry et al.‟s, 2004, meta-analysis were undergraduate students
who may have other priorities). For example, Patton and Meit‟s (1993) older subjects
“indicated that the task was more important to them than it was to younger subjects,
confirming the importance of motivation” (p. 175). Rendell and Thomson‟s (1999)
young adults‟ poor PM performance was attributable to their failure to keep an
electronic Organizer with them at all times. Comments from Rendell and Craik‟s
(2000) participants suggested that their older adults took the naturalistic task more
seriously than did their young adults. And Kvavilashvili and Fisher‟s (in press) older
adults reported “reliably higher levels of (intrinsic) motivation both before and after
completion of the task than younger adults.”
Age differences in motivation may also interact with other factors discussed
by Phillips et al. (this volume), particularly the use of memory aids. I find it hard to
imagine, for example, an undergraduate student turning one item of furniture upside
down in every room of the house as a reminder to make a phone call to the
experimenter, which was the external cue adopted by one older participant in
Maylor‟s (1990) study. This case (albeit extreme) makes the point that previous
categorisations of memory aids as internal, external or conjunction have overlooked
the obvious fact that all cues are not equal. As Phillips et al. suggest, we should be
wary of the simplistic view that older adults outperform the young in naturalistic PM
tasks because they are more likely to adopt external reminders. Instead, I suggest that
Commentary 7
we focus on the nature of the cues adopted, the effort involved in setting them up,
their potential effectiveness, how they interact with the structure of the person‟s life,
how practised the person is in using that particular cue, and so on. An additional
question is whether a strategy can be imposed or trained, as illustrated by the
following instruction to preachers on a Primitive Methodist Plan of 1857: “….as
forgetfulness is not tolerated, he is desired to mark out his appointments and read
them over once a week.”
Phillips et al.‟s (this volume) interesting discussion of ecological validity (see
their Table 1) concludes that task setting is most critical in determining whether there
are positive or negative age effects on PM, the former being observed “when tasks are
set in the day-to-day environment of the participant”. It is not clear whether this
would include, for example, the population-based study by Huppert, Johnson and
Nickson (2000) in which a random sample of people aged 65+ were interviewed in
their places of residence by a trained experimenter. Performance on a simple PM task
administered during the session declined dramatically and linearly with increasing
age. However, although the task setting was the everyday environment of the
participant, the PM task was still under the control of the experimenter (and the same
would apply to the web-based study to be described later, despite the physical absence
of an experimenter in that case) – see Phillips et al.‟s section on Participant Control.
The hypothetical scenario posed by Phillips et al. (this volume) is especially
sobering in that it highlights just how little we have learned from decades of both
naturalistic and laboratory work on aging and PM about older people‟s actual
performance in everyday PM tasks. These would include not only those habitual PM
tasks usually listed in the introductory sections of articles and grant proposals
(remembering to take medication, pay the bills, etc) but also “tasks that crop up
Commentary 8
unexpectedly during the day” (P268, Rendell & Thomson, 1999) such as
remembering to set the video to tape a television programme, turn off the bath taps or
the electric blanket, prompt a friend who has asked you to remind them of something
when they leave, phone someone back in an hour when they will be available, and so
on. As Phillips et al. note, we simply do not know whether aging data from
naturalistic- or laboratory-based tasks are more appropriate to everyday PM
situations. This seems a disappointing state of affairs but one that could be remedied
by following the recommendations for future research outlined in their chapter,
including systematic manipulations of relevant factors across both naturalistic and
laboratory settings, and also direct observational studies. It would for instance be
interesting to discover the extent of collaboration and reliance on others in prospective
remembering across adulthood (see Schaefer & Laing, 2000).
Aging PM in the Laboratory
The age PM paradox is of course a generalization that ignores the fact
highlighted by McDaniel et al. (this volume) that age-related deficits are not always
observed in laboratory-based PM tasks. Thus it is now recognised that PM tasks vary
in their attentional demands, with age differences greater under more demanding
conditions consistent with McDaniel and Einstein‟s (2000) multiprocess framework
(see Henry et al., 2004; Maylor et al., 2002). In their chapter, McDaniel et al. explore
in detail a distinction between focal and nonfocal cues in event-based PM tasks, the
former eliciting more automatic spontaneous retrieval than the latter because of their
“overlap with the information constellation relevant to performing the ongoing task”.
This is reminiscent of an earlier notion of “task appropriate processing” (p. 78,
Maylor, 1996a), which suggested that a key to understanding age differences in PM
was in terms of the relationship between the type or level of stimulus processing
Commentary 9
required to perform the ongoing task and stimulus processing required to perform the
PM task (see also Maylor et al., 2002). Although McDaniel et al. argue that these two
ideas are not the same, it seems to me that they are conceptually very similar and
indeed make identical predictions. Taking their example of making lexical decisions
as the ongoing task and indicating a member of the animal category as the PM task,
McDaniel et al. define this as task appropriate (because both are semantic tasks) but
nonfocal (because deciding if a letter string is a word “does not require processing the
semantic features necessary to make a category determination”). Instead, I would
define this as task inappropriate because the processing required to decide if a letter
string is a word is insufficient for a participant to realise that it is a member of a
particular category. Maylor (1996a) similarly defined Mäntylä‟s (1993) condition in
which the ongoing task was word association and the PM cue was any member of a
particular category (e.g., liquids) as task inappropriate because a shift was required in
the level of processing from the generation of a semantically related word to the
categorization of that word as a member of a specific group. A task appropriate
version of McDaniel et al.‟s example would be lexical decision as the ongoing task
and indicating the word “cat” as the PM task. Thus, I would claim that all the
nonfocal cases in McDaniel et al.‟s Table 1 are task inappropriate and all the focal
cases are task appropriate. In other words, in all the nonfocal cases, some additional
(self-initiated) processing, beyond that required by the ongoing task, is necessary for
the PM cue to be detected as such.
Terminology aside, the data are reasonably consistent in showing smaller agerelated deficits with focal than with nonfocal cues (e.g., Maylor et al., 2002;
McDaniel et al., this volume; Rendell, McDaniel, Forbes, & Einstein, in press;
Salthouse et al., 2004). However, the data are less consistent on the question of
Commentary 10
whether age deficits can be eliminated altogether with focal cues. For example,
Salthouse et al. observed age deficits in their PM tasks with focal cues (namely,
“drawing classification” and “concept identification”) and several of my own attempts
to replicate Einstein and McDaniel‟s (1990) classic findings of no age deficits with
focal cues have failed (Maylor, unpublished data). Some reasons for these
discrepancies will be discussed later.
One important possibility identified by McDaniel et al. (this volume) that has
been ignored in past studies is that perhaps older adults may be able to achieve an
equivalent level of PM performance to that of the young but only by sacrificing
ongoing task performance. Investigating this obviously requires monitoring ongoing
task performance both with and without the PM task to obtain a measure of PM costs
(and this should probably be routine in future PM studies). McDaniel et al. present
data with focal cues suggesting that the absence of age-related deficits in PM is not
accompanied by greater costs to the ongoing task in older adults. These results are
striking but they do raise a couple of concerns. The first is the extent to which
performance is at or close to ceiling – note that although focal PM performance was
not actually 100%, PM may be subject to a “functional measurement ceiling” (p.
1143, Salthouse et al., 2004) that is effectively lower than 100%. (This ceiling issue
also complicates the interpretation of Age x Condition interactions, where condition
might be focal/nonfocal, salient/nonsalient, event-based/time-based, etc.)
The second point is that in addition to considering ongoing task performance
more closely as advocated by McDaniel et al. (this volume), we should examine PM
performance in greater detail. Traditionally, only accuracy is considered. But
evidence that more might be learned from other measures, reaction time (RT) in
particular, comes from an unpublished study I conducted in collaboration with Maria
Commentary 11
Brandimonte. The PM task was embedded within an ongoing letter-matching task.
Participants were presented with strings of five letters and were asked to decide
whether the second and fourth letters in each string were the same or different (lettermatching task). They were additionally instructed that if one or both of these letters
was the letter “B,” then they should press the spacebar before making the usual
same/different response (PM task). PM targets occurred twice in each of 8 blocks of
65 trials. There were 26 young and 34 older participants with mean ages of 21 and 70
years, respectively, although the data from four participants (1 young; 3 older) were
excluded from the analysis because they did not perfectly recall every element of the
instructions when questioned at the end of the experiment. On the letter matching
task, both age groups were highly accurate (around 97% correct) but young adults
were faster than older adults (M RTs of 839 and 1267 ms, respectively, an age-related
slowing factor of 1.5). For the PM task, young adults were numerically but not
significantly more successful than older adults (63 and 57% correct, respectively),
consistent with a classification of the PM cue as focal. However, the PM responses of
young adults were considerably faster than those of older adults (M RTs of 1225 and
2337 ms, respectively, an age-related slowing factor of 1.9). This particularly marked
slowing of older adults‟ PM responses (also observed by West & Craik, 1999, though
using a nonfocal PM cue) could reflect greater difficulty in inhibiting the ongoing task
response, slower retrieval of the PM action required, and so forth. Whatever the
explanation, the point here is that PM performance in focal tasks may not be entirely
indistinguishable between young and older adults and that we may be missing more
subtle and potentially interesting age-related deficits in PM by generally focusing on
percent correct as the dependent variable.
Commentary 12
McDaniel et al.‟s (this volume) most intriguing result is the absence of reliable
age-related deficits in PM for nonfocal cues (requiring strategic monitoring), which
was associated with disproportionate costs to ongoing task performance in older
adults. In other words, for whatever reason, their older adults apparently chose to
prioritize PM performance at the expense of ongoing task performance. Why older
adults in previous studies of nonfocal PM tasks (e.g., Maylor, 1998) apparently chose
not to do this is a puzzle (but see discussion later on task instructions) that should be
addressed in future research.
It is suggested by McDaniel et al. (this volume) that their laboratory-based
findings can help to reconcile the age PM paradox. Thus, they argue that age deficits
will not occur in naturalistic PM tasks when the cue is focal because performance
relies on spontaneous retrieval processes that are preserved in old age. Nor will they
necessarily occur in naturalistic PM tasks when the cue is nonfocal because ongoing
task performance can be sacrificed or adjusted to take the PM task into account.
However, aside from the difficulties in extrapolating from the laboratory to
naturalistic settings (see Phillips et al., this volume), such arguments fail to explain
the positive effects of aging observed in naturalistic PM tasks. Nonetheless, McDaniel
et al.‟s provocative discussion opens up some interesting new lines of enquiry.
McDaniel et al. (this volume) close by mentioning other factors besides the
focal-nonfocal distinction that may influence the extent of age effects in laboratorybased PM tasks, one of which is time of day. Significant effects of time of day were
reported by Leirer, Decker Tanke and Morrow (1994) in a naturalistic study of older
adults who were required to simulate taking medication at specified times each day.
PM performance was best in the morning, a result they attributed to the morning
hours being less busy (see Rendell & Thomson, 1999, for a similar result). If time of
Commentary 13
day effects were found in laboratory-based PM tasks, they would presumably require
a different explanation because ongoing activity would be controlled.
Time of day has received some recent attention in the aging literature (see
Yoon, May, & Hasher, 2000, for a review) because whereas most young adults
describe themselves as “neutral” or “evening” types, most older adults are “morning”
types. Moreover, cognitive performance for young adults tends to be better when
tested in the afternoon than in the morning, with precisely the reverse for older adults.
For example, May, Hasher and Stoltzfus (1993) observed substantial age deficits in
recognition memory in the late afternoon (optimal for young but not older adults) but
no age deficits in the morning (optimal for older but not young adults). Therefore, a
potential explanation for the conflicting effects of aging on laboratory-based PM tasks
in the literature is that studies have been conducted at different times of the day, with
those showing no age deficits conducted in the morning and those showing age
deficits conducted in the afternoon. Unfortunately, such information is not usually
reported in laboratory-based PM studies.
Preliminary evidence from young adults in the laboratory comes from an
unpublished study in which 94 undergraduate students were tested either in the
morning or in the afternoon. The ongoing task was either to name famous people from
their photographs or to provide their occupations (half assigned to each condition) and
the PM task was to indicate those wearing glasses (cf. Maylor, 1996a, 1998). Figure 2
shows correct performance on the ongoing and PM tasks (name and occupation data
were combined) as a function of time of day. There was no overall effect of time of
day but there was a significant interaction with task (p < .05) indicating that PM
performance was relatively better in the morning and ongoing performance was
relatively better in the afternoon. The reason for this is unclear – one possibility is that
Commentary 14
there are differential and independent effects of time of day on retrieval from
semantic memory (ongoing task) and PM; another possibility is that participants adopt
different strategies or tradeoffs in the allocation of attentional resources between the
ongoing and PM tasks in the morning and afternoon. Whatever the explanation, these
data suggest that time of day should be considered in future studies (though time of
day effects may be more apparent with nonfocal than with focal PM cues),
particularly those involving different age groups.
Lifespan Changes in Complex PM
In contrast to the rather weak developmental improvements in PM reviewed
by Kvavilashvili et al. (this volume) and the absence of age deficits in PM obtained by
McDaniel et al. (this volume), Kliegel et al. (this volume) report a wealth of data
showing marked inverted-U-shaped changes in performance across the lifespan.
Importantly, their task is a complex planning task with presumably high demands on
executive functioning, in which the PM component as I have previously understood it
– for example, “the requirement to remember to perform an action at some point in
the future…in the absence of any prompting by the experimenter” (p. 175, Maylor,
1996b) – is only one element (i.e., intention initiation) of the intention formationretention-initiation-execution process. Kliegel et al.‟s paradigm was designed to
explore lifespan changes in each of these phases in a complex PM task. Impressive
data and combined plots from 555 participants over a wide age range (6-84 years)
show that growth, stability then decline (in some cases accelerating) are evident in all
phases but particularly for initiation and execution. Furthermore, an increase in the
need for inhibitory control in the execution phase exaggerated age effects across the
lifespan.
Commentary 15
At least as implemented in Sections 2 and 3 of Kliegel et al. (this volume), the
task‟s emphasis is on how participants form a plan to follow a set of arbitrary rules,
retain the plan, and then later carry it out without breaking the rules (see their
Footnotes 1, 2 and 4). However, I would argue that the paradigm fails to capture what
is perhaps most relevant to PM and that is the formation, retention and execution of a
plan for initiating an intention. Thus, in view of the PM instructions to self-initiate the
multitask set on encountering the PM cue (i.e., a request to write date of birth at the
top of a questionnaire – see Footnote 3), of most interest is what participants of
different ages do to ensure that they will remember, whether their methods are
effective, and so forth. In fact, Section 4 begins to address this potential criticism by
introducing conditions that include specifically instructing participants to consider
planning aids that would target the intention initiation component. Nonetheless, it
remains of interest to discover the extent to which participants of different ages focus
on intention initiation or execution in their spontaneous plans and how these then
relate to success or otherwise in each phase.
Another possible reservation concerning Kliegel et al.‟s (this volume)
paradigm is that it may not be relevant to performance outside the laboratory (though
this criticism also applies to other laboratory PM tasks, of course). For example,
Kliegel et al. consistently observed that older adults formed less elaborate plans in
comparison with young adults. However, while others have also found age deficits in
formulating and executing plans with novel tasks, it seems that if the planning task is
more familiar/ecologically valid, age differences disappear (see Phillips et al., this
volume). This is not to deny the potential value of Kliegel et al.‟s paradigm for
addressing questions such as the role of various explanatory mechanisms in each of
Commentary 16
the phases, but it does suggest some caution in interpreting the data and drawing
conclusions from the paradigm.
General Themes, Unresolved Issues, and an Internet Study
One common theme running through these chapters is the age-complexity
effect, which refers to the tendency for age (and developmental) differences in
performance to increase with the complexity of the task (see Salthouse, 1991, for
discussion). Age-complexity is related to the reduced processing resources view
whereby older adults and young children have fewer processing resources or less
attentional capacity than young adults; they are therefore particularly disadvantaged in
complex tasks that are more demanding and less automatic (Hasher & Zacks, 1979).
Although this view enables us to make relative predictions such as lifespan changes
should be more dramatic for nonfocal than for focal PM cues, it is less helpful in
making absolute predictions about whether any particular PM task will show age
differences. The argument becomes circular if we find a reliable age difference in PM
performance and therefore conclude that retrieval was not automatic. A possible
added complication is that sometimes PM tasks that appear more complex result in
superior performance. For example, Maylor (1993a) asked participants to name
famous faces and to circle the trial number if the person had a beard and cross out the
trial number if the person was smoking a pipe. Maylor (1996a) asked participants to
name famous faces and to circle the trial number if the person was wearing glasses.
PM performance in the first block of trials was much higher with the former more
complex instructions than with the simpler instructions (68% vs. 42% correct). It
seemed that greater effort was made to encode the complex instructions, which may
also have been mentally rehearsed more often subsequently because of their perceived
difficulty. A related argument was made by Henry et al. (2004) in categorising 6Commentary 17
event and 12-event PM conditions as requiring high and low strategic demands,
respectively, on the grounds that the greater frequency of PM cues would
“presumably maintain activation of the PM task” (p. 29). In short, the age-complexity
notion may not necessarily prove to be so helpful in the context of PM.
Another emerging theme is the somewhat inconsistent nature of the PM
findings in both the developmental and aging literatures. Of course, some contrasting
data patterns have led to interesting proposals like the age PM paradox that should
continue to inspire researchers. But other inconsistencies raise the general issue of the
reliability/validity of PM measures across the lifespan. In the only large-scale study of
several different laboratory event-based PM tasks, Salthouse et al. (2004) found that
although their PM tasks showed both convergent and discriminant validity, little of
the variance in each task was associated with what they had in common (unlike, e.g.,
RM). They therefore concluded with a note of caution to researchers trying to make
inferences about the construct of PM based on results from a single task. No such
equivalent data exist for naturalistic PM tasks, which represents a gap that urgently
needs to be filled.
Inconsistent results also highlight both the logical and methodological
problems associated with the investigation of PM. Thus, single binary measures of
success/failure on laboratory tasks such as remembering to ask for a red pen at the
appropriate moment (designed to simulate everyday “crop-up” tasks mentioned
earlier) are noisy, coarse indices of PM ability. Increasing the number of PM trials
introduces other complications – for example, it risks the task becoming one of
vigilance (Uttl, 2005); also, performance on the first PM trial may be influenced
differently by factors such as aging to performance on subsequent PM trials (Maylor,
1996b).
Commentary 18
A novel way of producing a more finely-grained index of PM performance
from a single PM response was recently described by Graf, Uttl and Dixon (2002).
Participants are shown the PM cue (e.g., a picture of a helicopter) and instructed to
stop performing the ongoing task (e.g., categorising letters presented in the center of
the computer screen) when the PM cue occurs in one of the corners of the screen.
Each ongoing task trial is accompanied by four pictures of different sizes. The PM
cue when it first appears is small and if the participant fails to respond, it reappears a
few trials later but slightly larger. This procedure continues until the participant
responds, the dependent variable being the size of the cue at that point. Using this
simple method, Uttl (2006) observed significant age-related deficits in PM with both
visual and auditory cues, older adults requiring larger and louder PM cues,
respectively, before responding. It would seem quite straightforward and potentially
interesting to extend this paradigm to the study of PM development.
An alternative solution to the problem of noisy, coarse data from single-event
PM tasks is to compensate by increasing the numbers of participants in the study. This
could be achieved by collecting data via the Internet. With the enormous recent
growth in access to the Internet has come a rapidly expanding literature reporting
psychological experiments conducted online (see Birnbaum, 2004; Reips, 2002, for
reviews). There are many obvious advantages of such methodology. For example, it
can save researchers both time and money as once an experiment is set up, it can be
run concurrently on large numbers of unpaid volunteers. These generally represent a
wider demographic than the usual undergraduate population and hence the results
may be more generalisable. Experimenter bias can be avoided because Internet
experiments run automatically. To these can be added a couple of particular
advantages with respect to research on aging, namely, participants are not required to
Commentary 19
travel for testing, and older adults are probably less anxious when tested in their own
familiar environment.
Of course, there are also some obvious disadvantages associated with Internet
experimentation. These include the possibility of a biased sample as although most
people now have access to computers, not everyone will have the appropriate
software installed for downloading and running experiments. People may not be
honest, for example, in answering questions about demographics (age, education,
gender, etc). More importantly, they may not always understand the instructions and,
unlike laboratory research, there is no opportunity to check and provide further
instructions if necessary. Internet studies provide no control over the conditions under
which the experiment is conducted – uncontrolled factors include monitor size, hand
positions, distractions, noise, time of day, and so on. Also, there is no control over the
state of participants who, for example, may be tired, intoxicated, or not wearing their
glasses.
However, in view of the large numbers of people who can be tested in Internet
studies, researchers can ensure the integrity of their data by taking a conservative
approach to the datasets they allow to enter the analysis. In general, this methodology
tends to produce effects that account for only a small proportion of the variance but
are highly significant. The effects may be more generalisable to real world situations
if they emerge from experiments conducted on diverse samples under poorly
controlled conditions. It is therefore argued that the considerable advantages more
than outweigh the disadvantages, particularly as evidence suggests that web-based
studies can reliably replicate laboratory findings (see Buchanan & Smith, 1999;
Gosling, Vazire, Srivastava, & John, 2004; McGraw, Tew, & Williams, 2000).
Commentary 20
Access to the Internet is now widespread in schools; older adults are also
increasingly being encouraged to use the Internet although home access decreases
with age (Cutler, Hendricks, & Guyer, 2003; Selwyn, Gorard, Furlong, & Madden,
2003). Thus there is currently enormous potential for conducting lifespan research
online. Recent published examples include a study on self-esteem by Robins,
Trzesniewski, Gosling and Potter (2002) with 326,641 individuals aged between 9
and 90 years, and a study of task switching by Reimers and Maylor (2005) with 5,271
participants between 10 and 66 years.
In collaboration with Robert Logie and the British Broadcasting Corporation
(BBC), Internet data are currently being collected on a set of memory experiments
that includes a simple PM task. Toward the beginning of the session, participants
view a screen informing them that later in the test, they will see a smiley face and they
have to remember to click on the smiley face when it appears. This PM cue is
presented in the top right-hand corner of the screen that provides feedback to the
participant after the tests have been completed. There is no time limit imposed on
viewing either the instruction screen or the feedback screen. Preliminary results from
1,199 UK volunteers aged between 16 and 77 years from the first five days of data
collection are presented in Figure 3.
1
The point-biserial correlation between exact age
and PM success was weak but highly significant, rpb = -.201, p < .001. Thus, although
the task setting was the participants‟ own familiar environment, there was
approximately linear age-related decline in PM performance from young adulthood.
Internet methodology may therefore be a promising avenue to pursue in the future,
particularly with respect to the investigation of PM in neglected age groups such as
adolescents and 20-40 year olds.
Commentary 21
Returning to the issue of discrepancies in the literature, a significant source of
variance across PM studies may lie in the exact wording of the task instructions,
which are not always reported but perhaps should be routinely included in
Appendices. As noted by Phillips et al. (this volume), age deficits can vary in the
laboratory depending on the relative emphasis in the instructions on the ongoing vs.
PM tasks. Performance may also be influenced by whether the PM task is described
explicitly as a test of memory as shown by Gao, Cuttler and Graf (2005) who
administered a neuropsychological test battery to 141 community-living adults. Prior
to starting the tests, the experimenter unplugged the phone to prevent disruptions and
asked the participant to remind the experimenter to plug the phone back when testing
was completed. Half of the participants were assigned to an “informed” condition in
which they were told that the phone task was designed to assess their memory,
whereas the other half were in a “naïve” condition in which they were not told that the
phone task was a memory test. Informed participants were significantly more
successful in remembering to remind the experimenter than were naïve participants
(approximately 59% and 28%, respectively).
The importance of the precise wording of instructions with respect to aging is
highlighted by intriguing data from a study of RM by Rahhal, Hasher and Colcombe
(2001; see also Desrichard & Köpetz, 2005). When the instructions emphasised the
memory component of the task (e.g., “You will be tested on your memory of this
information in phase two”), age deficits were observed. However, when the
instructions were phrased more neutrally (e.g., “You will be tested on this information
in phase two”), there were no age differences. Although task instructions are unlikely
to account entirely for discrepancies in the PM literature,
2
they surely deserve serious
consideration in the design of future aging (and perhaps developmental) studies.
Commentary 22
From a lifespan perspective, several important issues remain largely
unresolved. First, as mentioned in more than one chapter, all our present conclusions
are based on cross-sectional data so there is obviously a need to replicate PM findings
longitudinally. Second, there are at least some hints that PM may be more strongly
related to RM and other aspects of cognition in childhood and old age than in
adulthood, consistent with the differentiation-dedifferentiation view of intellectual
development across the lifespan (see Li, Lindenberger, Hommel, Aschersleben, Prinz,
& Baltes, 2004) but clearly further data are required. Third, although there appear to
be some parallels between the two ends of the lifespan in terms of mechanisms and
processes underlying the development and aging of PM (changes in executive
functioning, inhibitory control, speed of processing, etc), there may also be some
differences (cf. Craik & Bialystok, 2006). For example, it seems unlikely that the
significant contribution to PM performance from sensory functioning found in old age
(Uttl, 2006) will be replicated in development.
In summary, over recent years there has been considerable progress in
research on PM in both development and aging. Studies are now more theory-driven
and beginning to benefit from improved methods and insights from wider domains.
However, a nagging question relevant to all these chapters is whether any of the
findings would particularly surprise a developmental/aging RM researcher. The age
PM paradox perhaps stands out as the most unexpected; it would perhaps further
surprise an RM researcher to learn that there is still no completely satisfactory
explanation for it.
Commentary 23
Footnotes
1
Participants were in fact randomly assigned by the computer program to one
of four conditions in this PM experiment; for present purposes, the data have been
combined across the different conditions to produce Figure 3.
2
For example, compare the PM instructions in a study with no age deficits in
which participants were told that the experimenters “had a secondary interest in their
ability to remember to do something in the future” (p. 719, Einstein & McDaniel,
1990), and a study with age deficits in which the experimenter told participants “If
you see a person wearing glasses, then I want you to put a circle around the number of
that slide” (p. 75, Maylor, 1996a).
Commentary 24
References
Bialystok, E., & Craik, F. I. M. (Eds.) (2006). Lifespan cognition: Mechanisms of
change. New York: Oxford University Press.
Birnbaum, M. H. (2004). Human research and data collection via the Internet. Annual
Review of Psychology, 55, 803-832.
Brandimonte, M., Einstein, G. O., & McDaniel, M. A. (Eds.) (1996). Prospective
memory: Theory and applications. Mahwah, New Jersey: Erlbaum.
Buchanan, T., & Smith, J. L. (1999). Using the Internet for psychological research:
Personality testing on the World Wide Web. British Journal of Psychology,
90, 125-144.
Craik, F. I. M., & Bialystok, E. (2006). Cognition through the lifespan: Mechanisms
of change. Trends in Cognitive Sciences, 10, 131-138.
Cutler, S. J., Hendricks, J., & Guyer, A. (2003). Age differences in home computer
availability and use. Journal of Gerontology Series B: Psychological Sciences
and Social Sciences, 58, S271-S280.
Desrichard, O., & Köpetz, C. (2005). A threat in the elder: The impact of taskinstructions, self-efficacy and performance expectations on memory
performance in the elderly. European Journal of Social Psychology, 35, 537552.
Einstein, G. O., & McDaniel, M. A. (1990). Normal aging and prospective memory.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 16,
717-726.
Gao, J., Cuttler, C., & Graf, P. (2005, November). Prospective memory, instructions
and personality. Poster presented at the annual meeting of the Psychonomic
Society, Toronto, Canada.
Commentary 25
Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust webbased studies? A comparative analysis of six preconceptions about Internet
questionnaires. American Psychologist, 59, 93-104.
Graf, P., & Ohta, N. (Eds.) (2002). Lifespan development of human memory.
Cambridge, Massachusetts: The MIT Press.
Graf, P., & Uttl, B. (2001). Prospective memory: A new focus for research.
Consciousness and Cognition, 10, 437-450.
Graf, P., Uttl, B., & Dixon, R. (2002). Prospective and retrospective memory in
adulthood. In P. Graf & N. Ohta (Eds.), Lifespan development of human
memory (pp. 257-282). Cambridge, Massachusetts: The MIT Press.
Hasher, L., & Zacks, R. T. (1979). Automatic and effortful processes in memory.
Journal of Experimental Psychology: General, 108, 356-388.
Henry, J. D., MacLeod, M. S., Phillips, L. H., & Crawford, J. R. (2004). A metaanalytic review of prospective memory and aging. Psychology and Aging, 19,
27-39.
Hicks, J. L., Cook, G. I., & Marsh, R. L. (2005). Detecting event-based prospective
memory cues occurring within and outside the focus of attention. American
Journal of Psychology, 118, 1-11.
Huppert, F. A., Johnson, T., & Nickson, J. (2000). High prevalence of prospective
memory impairment in the elderly and in early-stage dementia: Findings from
a population-based study. Applied Cognitive Psychology, 14, S63-S81.
Kvavilashvili, L., & Fisher, L. (in press). Is time-based prospective remembering
mediated by self-initiated rehearsals? Role of incidental cues, ongoing
activity, age and motivation. Journal of Experimental Psychology: General.
Commentary 26
Leirer, V. O., Decker Tanke, E., & Morrow, D. G. (1994). Time of day and
naturalistic prospective memory. Experimental Aging Research, 20, 127-134.
Li, S.-C., Lindenberger, U., Hommel, B., Aschersleben, G., Prinz, W., & Baltes, P. B.
(2004). Transformations in the couplings among intellectual abilities and
constituent processes across the life span. Psychological Science, 15, 155-163.
Mäntylä, T. (1993). Priming effects in prospective memory. Memory, 1, 203-218.
Marsh, R. L., Cook, G. I., & Hicks, J. L. (2006). An analysis of prospective memory.
In B. H. Ross (Ed.), The Psychology of Learning and Motivation, Volume 46
(pp. 115-153). San Diego: Elsevier Academic Press.
May, C. P., Hasher, L., & Stoltzfus, E. R. (1993). Optimal time of day and the
magnitude of age differences in memory. Psychological Science, 4, 326-330.
Maylor, E. A. (1990). Age and prospective memory. Quarterly Journal of
Experimental Psychology, 42A, 471-493.
Maylor, E. A. (1993a). Aging and forgetting in prospective and retrospective memory
tasks. Psychology and Aging, 8, 420-428.
Maylor, E. A. (1993b). Minimized prospective memory loss in old age. In J. Cerella,
J. Rybash, W. Hoyer, & M. L. Commons (Eds.), Adult information
processing: Limits on loss (pp. 529-551). San Diego: Academic Press.
Maylor, E. A. (1996a). Age-related impairment in an event-based prospective
memory task. Psychology and Aging, 11, 74-78.
Maylor, E. A. (1996b). Does prospective memory decline with age? In M.
Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory:
Theory and applications (pp. 173-197). Mahwah, New Jersey: Erlbaum.
Maylor, E. A. (1998a). Changes in event-based prospective memory across adulthood.
Aging, Neuropsychology, and Cognition, 5, 107-128.
Commentary 27
Maylor, E. A., Darby, R. J., Logie, R. H., Della Sala, S., & Smith, G. (2002).
Prospective memory across the lifespan. In P. Graf & N. Ohta (Eds.), Lifespan
development of human memory (pp. 235-256). Cambridge, Massachusetts: The
MIT Press.
McDaniel, M. A., & Einstein, G. O. (2000). Strategic and automatic processes in
prospective memory retrieval: A multiprocess framework. Applied Cognitive
Psychology, 14, S127-S144.
McGann, D., Defeyter, M. A., Reid, C., & Ellis, J. A. (2005, July). Prospective
memory in children: The effects of age and target salience. Paper presented at
the Second International Conference on Prospective Memory, Zurich.
McGraw, K. O., Tew, M. D., & Williams, J. E. (2000). The integrity of Webdelivered experiments: Can you trust the data? Psychological Science, 11,
502-506.
Patton, G. W. R., & Meit, M. (1993). Effect of aging on prospective and incidental
memory. Experimental Aging Research, 19, 165-176.
Rahhal, T. A., Hasher, L., & Colcombe, S. J. (2001). Instructional manipulations and
age differences in memory: Now you see them, now you don‟t. Psychology
and Aging, 16, 697-706.
Reimers, S., & Maylor, E. A. (2005). Task switching across the life span: Effects of
age on general and specific switch costs. Developmental Psychology, 41, 661671.
Reips, U.-D. (2002). Standards for Internet-based experimenting. Experimental
Psychology, 49, 243-256.
Commentary 28
Rendell, P. G., & Craik, F. I. M. (2000). Virtual Week and Actual Week: Age-related
differences in prospective memory. Applied Cognitive Psychology, 14, S43S62.
Rendell, P. G., & Thomson, D. M. (1999). Aging and prospective memory:
Differences between naturalistic and laboratory tasks. Journal of Gerontology:
Psychological Sciences, 54B, P256-P269.
Rendell, P. G., McDaniel, M. A., Forbes, R. D., & Einstein, G. O. (in press). Agerelated effects in prospective memory are modulated by ongoing task
complexity and relation to target cue. Aging, Neuropsychology, and Cognition.
Robins, R. W., Trzesniewski, K. H., Gosling, S. D., & Potter, J. (2002). Global selfesteem across the life span. Psychology and Aging, 17, 423-434.
Salthouse, T. A. (1991). Theoretical perspectives on cognitive aging. Hillsdale, New
Jersey: Erlbaum.
Salthouse, T. A., Berish, D. E., & Siedlecki, K. L. (2004). Construct validity and age
sensitivity of prospective memory. Memory & Cognition, 32, 1133-1148.
Schaefer, E. G., & Laing, M. L. (2000). „Please, remind me…‟: The role of others in
prospective remembering. Applied Cognitive Psychology, 14, S99-S114.
Selwyn, N., Gorard, S., Furlong, J., & Madden, L. (2003). Older adults‟ use of
information and communications technology in everyday life. Ageing and
Society, 23, 561-582.
Somerville, S. C., Wellman, H. M., & Cultice, J. C. (1983). Young children‟s
deliberate reminding. Journal of Genetic Psychology, 143, 87-96.
Uttl, B. (2005). Age-related changes in event cued prospective memory proper. In N.
Ohta, C. M. MacLeod, & B. Uttl (Eds.), Dynamic cognitive processes (pp.
273-303). Tokyo: Springer.
Commentary 29
Uttl, B. (2006). Age-related changes in event-cued visual and auditory prospective
memory proper. Aging, Neuropsychology, and Cognition, 13, 141-172.
West, R., & Craik, F. I. M. (1999). Age-related decline in prospective memory: The
roles of cue accessibility and cue sensitivity. Psychology and Aging, 14, 264272.
Yoon, C., May, C. P., & Hasher, L. (2000). Aging, circadian arousal patterns, and
cognition. In D. Park & N. Schwarz (Eds.), Cognitive aging: A primer (pp.
151-171). Hove, East Sussex: Psychology Press.
Commentary 30
Figure Captions
Figure 1. Proportion of children in each of five age groups (mean ages 6.5-10.5 years;
n = 40 per group) who responded successfully to each of two PM targets (a teacher
wearing glasses; a plant in the picture) while naming their teachers from photographs
(ongoing task). From Maylor (unpublished data).
Figure 2. Mean correct performance from 94 undergraduate students, half of whom
were tested in the morning and half in the afternoon, for the ongoing task (providing
the names or occupations of famous people from their photographs) and the PM task
(circling the trial number of those people wearing glasses; n = 8 out of 120). Error
bars indicate 1 standard error of the mean. From Maylor (unpublished data).
Figure 3. Mean proportion of participants in each of seven age groups from 16-77
years (n = 232, 343, 283, 188, 111, and 42, respectively) who responded correctly to a
single PM target event in an Internet study of memory run from the BBC‟s Science
and Nature website. Error bars indicate 1 standard error of the mean. Preliminary
results from UK participants in the first five days of data collection (Logie & Maylor,
unpublished data).
Commentary 31
0.0
0.2
0.4
0.6
0.8
1.0
6.5 7.5 8.5 9.5 10.5
Age (years)
P
M
C
o rr e c t (p ro p o rt io n )
Glasses
Plant
Commentary 32
0.0
0.2
0.4
0.6
0.8
1.0
Ongoing PM
Task
C
o rr e c t (p ro p o rt io n )
Morning
Afternoon
Commentary 33
0.0
0.2
0.4
0.6
0.8
1.0
16-19 20-29 30-39 40-49 50-59 60-77
Age (years)
P
M
C
o rr e c t (p ro p o rt io n )

PDF Document reader online

This website is focused on providing document in readable format, online without need to install any type of software on your computer. If you are using thin client, or are not allowed to install document reader of particular type, this application may come in hand for you. Simply upload your document, and Docureader.top will transform it into readable format in a few seconds. Why choose Docureader.top?

  1. Unlimited sharing - you can upload document of any size. If we are able to convert it into readable format, you have it here - saved for later or immediate reading
  2. Cross-platform - no compromised when reading your document. We support most of modern browers without the need of installing any of external plugins. If your device can oper a browser - then you can read any document on it
  3. Simple uploading - no need to register. Just enter your email, title of document and select the file, we do the rest. Once the document is ready for you, you will receive automatic email from us.

Previous 10

Next 10