FULL REPORT: R000222575 - Mechanisms of Prospective Memory and Ageing


Back to Research Grants

Back to Homepage


Prospective memory has been distinguished from retrospective memory (remembering past information, e.g., someone’s name or address) and refers to one’s ability to remember to carry out an intended action in future (e.g., remembering to take a medication, keep an appointment, make a phone call, etc.). It is a relatively new but rapidly growing area of research which is partly due to its importance and relevance to memory problems experienced by people in everyday life. Indeed, when people complain about their memory they usually are more concerned with prospective than retrospective memory failures (Meacham, 1977). Old people and patients with frontal lobe damage are considered to be particularly vulnerable to prospective memory failures as they apparently lack self-initiating processes involved in prospective remembering (Craik, 1986). Therefore, it is not surprising that ageing has been a major focus in prospective memory research.


Despite considerable progress in the last decade research on prospective memory has also produced some theoretical controversies which revolve around the number and characteristics of different retrieval processes in prospective memory and the effects of ageing on these processes. One controversy refers to the precise number and characteristics of different types of prospective memory. According to Einstein and McDaniel (1996) there are only two distinct types of prospective memory tasks: event- and time-based tasks. However, according to a model proposed by Kvavilashvili and Ellis (1996) one should also distinguish activity-based prospective memory tasks (remembering to do something after finishing certain activity) which, unlike event- and time-based tasks, do not require an interruption of an on going activity and therefore should be easier to remember than event- and time-based tasks. One aim was to test this prediction by studying the effects of type of task on young and old participants. We had two hypotheses. Firstly, that the highest performance rate would be achieved in activity-based tasks. Secondly, that old people’s performance on activity- and event-based tasks would be comparable to that of young people but reliably worse on a time-based task.

Another controversy, refers to discrepant findings with respect to effects of ageing on event-based tasks (i.e., doing something in response to a certain cue event such as posting a letter when seeing a post box). While Einstein and colleagues have consistently failed to observe any age differences on these tasks significant age effects have been reported by Maylor (1993, 1996a) and Mäntylä (1993). Einstein and his colleagues, unlike other researchers, almost always made age adjustments to the difficulty of the background task in which the prospective memory task is embedded. So one possible way to resolve this controversy was to conduct a study where the difficulty of background activity would be manipulated in young and old participants. It was expected that increased difficulty (i.e., high cognitive load) of background activity would be more detrimental for older than for young participants. We also expected to replicate Einstein et al.’s (1995) finding that when the difficulty of background task was age adjusted, performance on event-based task would show no age effects.

Finally, the third controversy concerns discrepant findings with respect to effects of ageing on time-based tasks (e.g. making a phone call at 2 p.m.). While no age effects have been obtained in naturalistic studies conducted outside the laboratory (Maylor, 1996b), significant effects have been obtained in a laboratory study by Einstein et al. (1995).



The overall objective of the project was to resolve these controversies within a unified theoretical and experimental framework by studying the effects of ageing in a variety of distinct retrieval conditions. Originally, two studies were planned. The specific aims of Study 1 were


1. To resolve current debates about the main forms of prospective memory tasks and identify whether elderly have particular difficulties with any of the three forms of prospective memory that have been discussed in previous research.

2. To assess whether elderly, in comparison to young participants, are especially vulnerable to prospective memory forgetting in conditions of enhanced cognitive loads.


The specific aim of Study 2 was to investigate the effects of external cues (reminders) and task interruption in order to enhance our understanding of retrieval mechanisms in prospective memory, and to identify ways of assisting elderly in carrying out time-based prospective memory tasks.

In line with the typical way of conducting research in cognitive ageing, we intended to use only two groups of participants: elderly (aged 60 and over) and young people (aged 18-25). However, shortly before the commencement of the project in October 1998 our attention was drawn to studies conducted by Maylor (1998) and Mäntylä and Nilsson (1997) indicating that old people in their 60’s may be reliably better on prospective memory tasks than those in their 70’s, with accelerated decline starting only after 75 years of age. These findings seemed important to us for the following reasons. Firstly, they can explain a large variability in prospective memory scores that often exists within a group of old people with a wide age range (60 to 80 and above) (e.g., see Park et al., 1997). Secondly, and more importantly, these findings indicate that by collapsing old participants of various ages into one single group interesting age effects on prospective memory can be masked.

In the light of above considerations it seemed fully justified to conduct Study 1 not on one group of old participants but on four groups covering the 5-year age bands: 61-65, 66-70, 71-75 and 76-80. This design provided the opportunity to enhance the potential impact of the research by determining the onset of decline in different types of prospective memory tasks. However, it also meant that we needed to recruit larger numbers of old participants (at least 144 instead of 96 as planned originally) which would make it impossible to conduct Study 2 given our time constraints Therefore, it was decided to drop Study 2 and to compensate the loss of its objectives by adding new objectives to Study 1.

Specifically, by turning Study 1 into a large scale investigation where in addition to a main prospective memory task young and old participants were also administered several other prospective and retrospective memory tasks as well as various additional tests of cognitive functioning and personality variables, it became possible to address two new objectives:


3. To establish the onset of age decline in several prospective memory tasks as well as various measures of general cognitive functioning.

4. To compare the onset of age effects in retrospective and prospective memory tasks in order to establish the similarities and differences between these two forms of memory.


Both objectives are highly relevant and important for current prospective memory research. Thus, basic information about the onset of decline in different types of prospective memory tasks is largely missing due to a current practice in ageing research to use only one group of old participants covering a wide age range (from 60 to 80 or above). Another important question in prospective memory research has always been the relationship between prospective and retrospective memory and whether they share common underlying mechanisms. By comparing the onset of age effects (and effect sizes) in various prospective and retrospective memory tasks it should be possible to provide some answers to this issue as well.

In addition to four main objectives specified above, the rich data set obtained in Study 1 would enable us to address several other interesting secondary issues such as


5. Correlational analyses to determine how diverse prospective memory tasks are related to each other and other measures of cognitive and emotional functioning

6. Activation/inhibition processes involved in remembering two different intentions simultaneously

7. The nature and characteristics of mind-popping in young and old participants (see method section below)


Study 1 was successfully completed and the statistical analyses of data has shown that we were able to meet all four objectives (for details see Results Section).

This report comprises a full account of the method followed by results and their discussion in order to show how the analyses that have already been completed enable us to meet the first four objectives. There is then a brief account of how planned additional analyses will meet objectives 5 -7. Finally, we will briefly cover such issues as activities, outputs and dissemination, impacts and future research plans and priorities.




All participants took part in one experimental session which lasted 1.5 to 2 hours. At the end of the first hour participants had a break of 10-15 minutes. Tasks administered to the participants are described below. Task order is detailed in the Appendix 1.

Prospective Memory tasks:

(1) Main Prospective Memory Task: A modified version of Einstein et al. (1995; Exp. 3) paradigm was used. The task consisted of asking participants to remember to type in 6 numbers (1, 2, 3, 4, 5, and 6), as if making a phone call to a friend, while they were engaged in answering general knowledge questions presented on a computer. In the event-based condition participants had to type in the numbers when they received a question about a telephone; in the time-based condition they had to do this once every 3 minutes; and in the activity-based condition after they finished each block of questions. Half of the participants were presented with questions at a normal rate (low cognitive load or low time pressure) and the other half at a fast rate (high cognitive load or high time pressure). Following Einstein et al. (1995), normal rate for young and elderly was defined as 12 and 15 sec per question respectively. The fast rate was 9 sec per question for young and 12 sec for elderly. Prospective memory performance was scored as a proportion of correct, on time responses out of six target opportunities which in each condition occurred once in every 3 minutes during a 19 minute task of general knowledge questions.

(2) Colour Task (event-based): At the beginning of the testing session, participants were told that whenever they were given a task printed on a coloured sheet of paper, they were to write the name of the colour on that sheet of paper. There were three opportunities to make this response throughout the session. Performance was scored as proportion of correct responses.

(3) Red Pen Task (event-based): At the beginning of the second half of the session, participants were told that if at some stage in the session they were asked to copy a geometric figure, they were to request a red pen with which to draw it. After 10-15 minutes they were asked to copy interlocked pentagons (a sub task of Mini Mental State Examination) which was a cue event for remembering the red pen task.

(4) Time and Date Task (activity-based, naturalistic): At the end of the session, the participants were given a booklet of questionnaires and were asked to take them home and fill them in. A prospective memory task was to write the time and date at which they completed the questionnaires in the top left hand corner (Dobbs & Rule, 1987).


Retrospective Memory Tasks

(1) Recognition Memory: Eighteen sets of four words were presented from which the participants had to identify the one word in each set which had occurred in the Spot-the-Word test , SCOLP2. This was performed as part of the tests of general cognitive functioning described below.

(2) Recall of three words: This was a sub task of the Mini-Mental State Examination and involved a delayed recall of three words.

(3) Paired Associate Learning: Participants were presented either with 10 (for all old age groups) or 20 (for young age group) pairs of unrelated words, and were then shown the first word of each pair, in the same order, and asked to recall the second word. Old participants received half as less items because the pilot study revealed that they performed at floor in the 20 item condition.


General Cognitive Functioning

(1) The Speed of Comprehension Test (SCOLP1) - a sub test of the Speed and Capacity of Language-Processing Test (SCOLP) (Baddeley et al. (1992) consists of 100 sentences which subjects have to rate as true or false. A score is a total number of correctly rated sentences in 2 minutes.

(2) Spot-the-Word Test (SCOLP2) - a vocabulary sub test of SCOLP. A score is a number of real words identified from a total of 60 real word - non word pairs.

(3) Mini-Mental State Examination (MMSE)(Folstein, et al. 1975) - A brief test of general cognitive functioning used in our study partly as a screening measure for elderly participants. A score , MMSE - recall comprising total score (max.=30) minus recall of three words score (see above), was used as a test of general functioning, since we wished to distinguish between memory and other cognitive functions.

(4) General Knowledge Questions - This was the computer based task into which our main prospective memory task was embedded (see above). A score was a proportion of correctly answered questions.


Affective/Personality Variables

The Hospital Anxiety and Depression Scale (HADs) (Zigmund & Snaith, 1983) - This questionnaire is sub-divided into two scales measuring state anxiety (generalised) and depression, as experienced over the last few days. Scores of 16 and above indicate a serious level of symptoms, and participants scoring at this level were excluded from the final data set.


Demographic and Health Information

(1) Each participant was asked to rate their health on a 5-point (1=poor; 5=excellent) scale as well as their health state in comparison to their peers (1=worse; 3=same; 5=significantly better)

(2) The following demographic information was also obtained: years in full time education, current or most recent occupation, marital status, gender.


Additional Variables/Tasks

In addition to the tasks described above several other objective and self-report measures were obtained from the participants. Only the two most theoretically interesting are described here, due to report length constraints. These are: (1) Mind-Popping Questionnaire (MPQ) (Kvavilashvili & Mandler, 1998) and (2) Schizotypal Personality Scale (STA) (Claridge and Brooks, 1984)

Mind-popping or involuntary semantic memory is defined by Kvavilashvili and Mandler (1998) as involuntary occurrence in one’s mind of the contents of semantic memory (e.g., words, images, familiar tunes, etc.). Very little is currently known about the characteristics and prevalence of involuntary semantic memories in wider population (as opposed to psychology undergraduates), and we had a very good opportunity to obtain some information on these issues by administering a short 5 item Mind-Popping Questionnaire (MPQ) to our participants. Another questionnaire that participants were asked to fill in with MPQ was STA - a sub scale of the Psychotic Traits Questionnaire (STQ). This was necessary to assess the correctness of our theoretical ideas about the possible relationship between the occurrence of involuntary memories and high scores on STA. An additional reason for including the MPQ into our study was an interesting similarity that exists between remembering intentions and involuntary semantic memories - both seem to occur spontaneously, often without any obvious internal or external cues.



A total of 257 young and old participants were individually tested. From the entire number of these participants, 34 were excluded from the final analysis: Sixteen were used in pilot tests; 4 were tested but excluded by the screening criteria (mainly due to scoring above the cut off point 14 on Anxiety Scale); and 14 failed to comply sufficiently well with the instructions for the main prospective memory task. The remaining 223 participants comprised 72 young (age 18 - 30), and at least 36 mentally and physically healthy participants in each of the four older age groups of 61-65, 66-70, 71-75 and 76-80. Detailed information about subject numbers, gender, and mean ages is presented in Appendix 2.

Both young and old participants were recruited from the local community in order to minimise differences due to background and education levels. The most successful recruitment method that we can highly recommend to other researchers turned out to be sending out letters (N=1489) to all patients on the local GP’s register who fell within appropriate age categories. This approach resulted in the recruitment of 185 volunteers (with 4% response rate in young and 17% - in old participants). The remaining 72 participants were recruited by

(a) advertising and the distribution of leaflets in the local newspaper (26 young, 2 old)

(b) advertising within the university for non-academic staff (15 young)

(c) advertising in a local young people’s advice centre, the local job centre and college of further education (9 young)

(d) friends of participants (4 young, 16 old)

Old participants were normal healthy community residing elderly who did not report experiencing any problems with vision, hearing or physical mobility. There was no reliable difference on self-rated health between young and old participants. All old participants were also well above the cut off point of 24 on Mini Mental State Examination. In all age groups the reported levels of anxiety and depression were in the normal range, i.e., well below the cut off point of 14 (Fig. 3 - bottom panels). Performance on the vocabulary test was reliably better for older participants, indicating that any superiority in performance of the younger group was not likely to be due to different crystallised cognitive ability (Fig.3 - middle panel).



An Apple Power Macintosh 7500/100 and 4400/160 was used for just two tasks: the main prospective memory task of typing in six digits and the associated background task of answering General Knowledge Questions; and the Paired Associate Learning task. Detailed instructions/explanations and practice was provided prior to the commencement of these tasks to ensure that older people were not disadvantaged by having less computer experience. All other asks were either paper and pencil tasks or orally presented.




Extensive analyses of results has already taken place and several other analyses, addressing the subsidiary objectives of the study, will be conducted shortly. We used the methods of analyses of variance and co-variance in order to look at the effects of age (and other variables where appropriate) on a number of variables described in the method section. However, with the main Prospective Memory Task we have also used more sophisticated methods of analyses (such as logit regression, Cochran-Manten-Haenszel - a repeated measures contingency table analysis) which take into account the fact that scores on most of the prospective memory tasks are not normally distributed. As will be shown below this allowed us to detect more subtle effects of ageing, presentation rate (cognitive load) and type of prospective memory task. The results will be reported in four sections. These will refer to findings obtained for the main prospective memory task, other prospective memory tasks, three retrospective memory tasks, and several other non-memory tasks.


(1) Main Prospective Memory task.

(a) Effects of Age. The analyses conducted on the scores of the main prospective memory task revealed a reliable effect of age so that young participants performed better in all three types of tasks than old participants (Fig.1: top panel). However, more detailed analyses showed that young group reliably differed only from the oldest age groups of 71-75 and 76-80. There was no statistically reliable difference between young and the elderly who were in the age bands of 61-65 and 66-70 (Fig.2: left top panel). Moreover, the latter were also performing significantly better than group of 76-80. These findings have several important implications. Firstly, they show that a decline in prospective memory in old age starts at a relatively late point. Secondly, they highlight the importance of grouping old participants into age bands instead of treating them as a homogeneous group. Finally, they cast some light on the possible reasons of obtaining discrepant results in different studies with respect to age effects on prospective memory. Thus, age effects can be attenuated (or perhaps even eliminated) in those studies which recruit elderly participants predominantly in their 60’s whereas reliable age effects are likely to be obtained in those studies who recruit participants predominantly in their 70’s and above.

(b) Effect of Type of Task. Another important finding refers to the effect of type of task. In line with our predictions, performance on activity-based task, was reliably better than on event-based task (Fig.1: top panel). Moreover, age effects were generally weakest for activity based-task. There was no reliable decline before age 70, with 88% remembering all the time. Even among people aged 71-80, 65% remembered all the time. Furthermore, performance on activity-based task was all or none. Less than 3% of participants had partial recall (1-5 recalls from 6 opportunities). By contrast, partial recall rates were 66% in time-based, and 32% in event-based tasks. Taken together, these findings provide support for the essential distinction between action- and event-based prospective memory postulated by Kvavilashvili and Ellis (1996).

However, in each age group performance on time-based task was also significantly better than on event-based task. This finding contradicts those obtained by Einstein et al. (1995). In their experiment, performance on time-based task was reliably worse than on event-based-task in elderly participants. The inspection of the top panel of Fig. 1 reveals that the main effect of task in our study was due to low performance levels on event-based condition. One possible reason for this low performance is that target event, i.e., question about the ‘telephone’ was much less distinctive than the question about the ‘president’ used by Einstein et al. (1995). Thus, questions about the president (N=6) were supplied with four possible answers which mostly were the names of presidents. Therefore if the participants did not notice a word president in the question itself they were likely to be reminded about the task by actually reading the names of the presidents presented underneath that question. In the present study, however, the target word ‘telephone’ occurred only in the question and not in the four possible answers. This line of reasoning, if correct, raises important issues for researchers who in future will want to compare performance on event-, activity- and time-based tasks.

(c) Other effects. Overall there was no main effect of presentation rate (cognitive load). This could be either due to a fact that the fast presentation rates employed in the study were not sufficiently taxing participants cognitive recourses or to the nature of General Knowledge Questions. Indeed in the majority of cases participants either immediately knew the correct answer or did not know it at all.

There were no reliable two or three way interactions in an ANOVA with the three explanatory factors of type of task, presentation rate and age group. However, the more sensitive logistic regression analysis showed that the two interactions involving presentation rate suggested by the bottom two panels in Figure 1 were highly reliable. Specifically, presentation rate interacted with age so that fast presentation reliably impaired prospective memory performance in the group of oldest participants (76-80) only (Fig. 1 - bottom panel). Fast presentation also reliably impaired performance in the time-based prospective memory task only (Fig. 1 - middle panel). Thus very old people are likely to be particularly vulnerable to prospective forgetting in time based tasks with high cognitive load.


(2) Other Prospective Memory Tasks

Separate analyses were conducted on all other prospective memory tasks that were administered to participants throughout the session and outside the laboratory in the home. No age effect was obtained for the date and time task. This finding replicates and extends some of the earlier studies which showed that naturalistic prospective memory is age independent (Maylor, 1996b; Rendell & Thompson, 1999). In contrast, the decline in performance on colour task and red pen task started earlier than in the main prospective memory task, namely at age 66-70 (Fig. 2: middle and bottom left panels). Inspection of the graphs also shows that generally the lowest performance levels were found in the colour task and the highest in the red pen task. The latter had a relatively unfamiliar cue - the words "geometric figure" - which apparently stood out from the otherwise familiar spoken words and activities involved in Mini Mental State Examination. On the other hand, the cue event in the colour task was much less distinctive. In fact in order to notice a cue a participant had to switch his/her attention from the on going task to an irrelevant feature of the environment. It has been suggested that event-based tasks which require such attention switching can be particularly affected by ageing (Maylor, 1996a; 1998).


(3) Retrospective Memory Tasks

(a) Recognition. Although recognition is a retrospective memory task that is supposed to be minimally affected by ageing (Craik, 1986), a significant age effect was obtained in the present study (Fig. 2: top right panel). Moreover, highly reliable difference existed between young and a group of young elderly of 61- 66. There was no further decline in performance until age 76-80.

(b) Recall of Three Words. A highly significant age effect was obtained also for this retrospective memory task, and again, the significant decline in performance was present in the youngest age group of elderly participants (Fig. 2: right middle panel).

(c) Paired Associate Learning. A statistically reliable age effect was obtained in this task, although the effect size was much smaller than in recognition and recall tasks (Fig. 2: bottom right panel). However, the interpretation of the age effects obtained in this task should be qualified by the fact that young participants received twice as many items to learn as old participants (20 and 10, respectively).


(4) General Effects of Ageing

The collection of various cognitive and affective measures throughout the session enabled us to also examine the age effects on these variables. These effects are presented in Fig 3. Two important findings emerged:





The aim of this section is to show how the results meet four main objectives specified in the beginning of this report.


Objective 1 - Controversy about main forms of prospective memory tasks.

Our results have unambiguously shown that performance on the activity-based task is quite different from event- and time-based tasks, and suggest that different processes are involved in these 3 types of tasks. Following the model of Kvavilashvili and Ellis (1996) we believe that the superiority of activity-based condition was due to the fact that participants had to remember to type in the numbers during the 15 sec rest interval which ensued at the end of each block of questions whereas participants in the other two conditions had to interrupt the question answering task in order to carry out the same action. However, the failure to replicate Einstein et al.’s (1995) age by task interaction (in event- and time-based conditions) has also shown that establishing the precise number of different types of prospective memory tasks is complex as levels of performance on event-based tasks vary with respect to the characteristics of target events and the type of attentional demand. Therefore, future research based on our results should concentrate on studying variables such as presence of and/or characteristics of cue events, task interruption, and attention switching, which are apparently crucial for distinguishing among different kinds of prospective memory tasks.


Objective 2 - Controversy about presence/absence of age effects in event-based tasks.

We addressed this issue by manipulating the presentation rate (cognitive load) of general knowledge questions in our main prospective memory task. However, this manipulation did not have any effect on performance on the event-based task. Therefore it did not turn out to be helpful in explaining the absence and presence of age effects in event-based tasks observed in different studies. However, several other findings obtained in the project with respect to remembering different event-based tasks seem to provide some solutions to this controversy. Performance levels in our main event-based task using the target word "telephone" was apparently much less distinctive than when using the word "president"(Einstein et al., 1995). On the other hand, performance levels on another event-based Red Pen task was very high apparently due to unfamiliarity and high distinctiveness of cue word "geometric figure" (cf. McDaniel & Einstein, 1993). These findings suggest that one of the possible reasons for obtaining discrepant findings is that different studies may utilise cue events of different familiarity and distinctiveness. Another possible reason, as revealed by our results, is that some event-based tasks require more attentional resources than others (Maylor, 1996a). For example, the strongest age effect in our study was obtained for the event-based Colour Task in which participants had to switch their attention from an ongoing task to some irrelevant cues in the environment.


Objective 3 - Onset of ageing in prospective memory and other cognitive tasks; and


Objective 4 - Comparison of age effects in prospective and retrospective memory

Due to the important decision to divide the older participants into four age groups, it was possible to show in detail how diverse psychological processes were differently affected by age. A major finding that emerged from the study is that the effect of age on prospective memory depends critically on how old an 'old' person is. Although overall prospective memory declines with age the decline starts at relatively later stage and its onset varies in different prospective memory tasks. In addition, most of the obtained age effects were reduced and/or disappeared when the effects of general cognitive functioning and personality variables (e.g., anxiety and depression) were taken into account. On the other hand, very large effects of age were obtained for retrospective memory tasks of Recognition and Recall and the substantial decline was already present in the youngest group of elderly (61-65). These different patterns of ageing revealed for prospective and retrospective memory tasks indicate that the former is much less disrupted by age than the latter and provides further support to existing theoretical distinction between these two types of memory.




In addition to the analyses reported above we also intend to carry out further analyses on the obtained data to address the subsidiary objectives (5), (6) and (7). Firstly, in order to further investigate how different prospective memory tasks are related to each other and other measures of cognitive and emotional functioning we intend to carry out Factor Analyses (objective 5). Secondly, we want to explore (probably using a repeated measures contingency table analysis) how remembering/forgetting one prospective memory task may affect the retrievability of another intention that has to be remembered at the same time (objective 6). It is possible to address this interesting issue because when participants were asked to copy a geometric figure they had to remember to ask for the red pen. However, the geometric figure was presented on the coloured paper which was the third cue-event occasion for the event-based Colour Task introduced in the beginning of the session (see Appendix 1). Finally, we intend to analyse the data obtained in the questionnaires of mind-popping and STA in order to investigate the nature and characteristics of involuntary semantic memories and their relation to age and prospective memory (objective 7).



The present project provided us with (i) interesting insights into various theoretical and methodological issues of studying different types of prospective memory and ageing, and (ii) invaluable experience in recruiting and conducting a study on elderly participants. Future research that will follow from this project will concentrate on studying those variables that are crucial for distinguishing among different kinds of prospective memory tasks and will also address the objectives of Study 2 which were in our initial proposal. The practical implications for enhancing recall of important prospective memory tasks such as taking medicine will also be investigated.




Baddeley, A. D., Emslie, H., & Nimmo-Smith, I. (1993). The speed and capacity of language processing (SCOLP) test. Bury St. Edmunds, Suffolk: Thames Valley Test CO.

Craik, F. I. M. (1986). A functional account of age differences in memory. In F. Klix & H. Hagendorf(Eds.), Human Memory and cognitive capabilities: Mechanisms and performances (pp. 409-422). Amsterdam: Elsevier.

Dobbs, A. R., & Rule, B. G. (1987). Prospective memory and self-reports of memory abilities in older adults. Canadian Journal of Psychology, 41, 209-222.

Einstein, G. O. & McDaniel, M. A. (1996). 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. Mahwah, NJ: Erlbaum

Einstein, G. O., McDaniel, M. A., Richardson, S. L., Guynn, M. J., & Cunfer, A. R. (1995). Aging and prospective memory: Examining the influences of self-initiated retrieval processes. Journal of Experimental Psychology: Learning, Memory & Cognition, 21, 996-1007.

Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). "Mini-Mental State": A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatry Research, 12, 189-198.

Kvavilashvili, L., & Ellis, J. (1996). Varieties of intentions: Some distinctions and classifications. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and applications. Mahwah, NJ: Erlbaum.

Kvavilashvili, L., & Mandler, G. (1998, September). Mind-popping: A naturalistic study of involuntary semantic memories. Paper presented at the BPS Cognitive Section Annual Conference, University of Bristol.

Mäntylä, T. (1993). Priming effects in prospective memory. Memory, 1, 203-218.

Mäntylä, T., & Nilsson, L.G. (1997). Remembering to remember in adulthood: A population-based study on aging and prospective memory. Aging, Neuropsychology, and Cognition, 4, 81-92.

Maylor, E. A. (1993). Age and prospective memory. The Quarterly Journal of Experimental Psychology, 42A, 471-493.

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. Mahwah, NJ: LEA.

Maylor, E.A. (1998). Changes in event-based prospective memory across adulthood. Aging, Neuropsychology, and Cognition, 5, 107-128.

McDaniel, M. A., & Einstein, G. O. (1993). The importance of cue familiarity and cue distinctiveness in prospective memory. Memory, 1, 23-41.

Meacham, J. A. (1977). Soviet investigations of memory development. In R. V. Kail & J. W. Hagen (Eds.), Perspectives on the development of memory and cognition . Hillsdale, NJ: LEA

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.

Rabbitt, P. (1993). Does it all go together when it goes? The Quarterly Journal of Experimental Psychology, 46A, 385-434.

Rendell, P., & Thompson, D. (1999). Aging and prospective memory: Differences between naturalistic and laboratory tests. Journal of Gerontology: Psychological Sciences, 54B, 256-269.

Zigmund, A., & Snaith, R. (1983). The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavia, 63, 361-370.





Order of Presenting the Tasks and Instructions during the Session


1. Introduction

2. Signing the consent form

3. Filling in personal detail sheet

4. Instructions for event-based prospective memory task (Colour Task)

5. Instructions for the General Knowledge Questions (GKQ)Task

6. Short practice session for GKQ Task (8 Questions)

7. Instructions for the Main Prospective Memory Task

-for event-based condition

-for activity-based condition

-for time-based condition

8. Filler task: SCOLP1

9. Filler task: SCOLP2

10. Filler task: Recognition Memory test (Presented on green paper)

11. General Knowledge Questions Task &emdash; Main Session (19 mins)

12. Probing those who forgot the Main Prospective Memory Task on all six occasions

13. General Knowledge Questions task (final short session with 8 questions)

14. Estimate how long it took to answer these questions

15. Hospital Anxiety and Depression Scale (Presented on yellow paper)


                  COFFEE BREAK


16. Instructions for event-based prospective memory task (Red Pen Task)

17. Paired Associate Learning Task

18. Mini Mental State Examination

- sub item - Recall of 3 words

- sub item - Draw a geometric figure (interlocked pentagons) (Presented on peach paper)

19. Probing those who forgot either the Red Pen or Colour Prospective Memory Tasks

20. Instructions for Mind-popping Questionnaire and STA

21. Instructions for a naturalistic activity-based prospective memory task (Time & Date Task)

22. Signing a receipt for £10

23. Obtaining a consent for Subject Participation Pool




Age and Gender Characteristics of Participants


Age group


Total no. of participants




Mean age

(SD in parentheses)










23.50 (3.60)










63.11 (1.47)










68.51 (1.45)










72.64 (1.40)










77.47 (1.28)











Figure 1. Probability of recall in main prospective memory task as a function of age, task and presentation rate.

Panels: Top = Task by age; Middle = Rate by task; Bottom = Rate by age



Figure 2. The effect of Age on Memory Performance.

Panels: Left = Prospective Memory; Right = Retrospective Memory



Figure 3. Effect of Age on non-Memory Tasks.

Panels: Top = Cognitive Function; Middle = General Knowledge; Bottom = Affective State.

Date of this Document: Jan 01

Ownership of this Document: Lia Kvavilashvili