Abstract

Prospective memory (PM) refers to the ability to remember to do something in the future. This study examined the relationship between three PM subtypes, and intelligence and retrospective memory (RM) in schizophrenia patients and healthy controls. The study sample comprised 110 schizophrenia patients and 110 healthy controls matched according to age, sex, and level of education. The patients' clinical condition was evaluated with the Brief Psychiatric Rating Scale. Time-, event-, and activity-based PM and RM (immediate and delayed Logical Memory subtests of the Wechsler Memory Scales-Revised), executive functioning (Design Fluency Test, Tower of London-4 disk, and Wisconsin Card Sorting Test), and intelligence (Raven's Progressive Matrices) tests were administered to all participants. Correlation analyses showed time- and event-based PM to be significantly associated with RM in both the patients and controls, but with intelligence only in the patients. After controlling for covariates, only time-based PM was associated with RM in the controls and only event-based PM with intelligence in the patients. In schizophrenia, PM deficit may arise from the impairments of the retrospective components of memory.

Introduction

Prospective memory (PM) is defined as the ability to remember to perform actions in the future (Einstein & McDaniel, 1990). PM deficits can lead to a highly disorganized life because remembering to perform certain tasks in the future is vital to the establishment of a structured life (Altgassen, Kliegel, Rendell, Henry, & Zollig, 2008). PM has three subtypes (Einstein & McDaniel, 1990; Kvavilashvili & Ellis, 1996): Time-based (remembering to perform an action at a specific time), event-based (remembering to perform an action when an external cue appears), and activity-based (executing an intention after finishing another task).

Previous studies exploring PM in psychiatric disorders have mainly focused on schizophrenia (Kumar, Nizamie, & Jahan, 2005,, 2007; Shum, Ungvari, Tang, & Leung, 2004; Ungvari, Xiang, Tang, & Shum, 2008; Woods, Twamley, Dawson, Narvaez, & Jeste, 2007). Evidence suggests that PM deficits in schizophrenia constitute a primary deficit (Kumar et al., 2005; Wang et al., 2008) that involves multiple cognitive processes such as executive functions, retrospective memory (RM), and IQ (Altgassen et al., 2008; Kliegel, Martin, McDaniel, & Einstein, 2002; Shum et al., 2004; Wang et al., 2008). However, little is known about the relationships between the PM subtypes and RM and intelligence after adjusting for possible covariates. Wang et al. (2008) examined 54 schizophrenia patients and 54 healthy controls, and found that in the patients IQ was significantly associated with time- and event-based PM, whereas RM measured by Wechsler Memory Scales-Revised (WMS-R) Logical Memory-Delayed Recall was associated only with event-based PM. However, they did not control for the confounding effects of other variables (e.g., age, education, executive functions, length of illness, etc.), and the results for the controls were not reported.

Exploring the relationship between PM and other components of cognition, such as RM and intelligence, in both healthy controls and schizophrenia patients could have theoretical and clinical implications for research and clinical practice. Following the preliminary evidence that PM is associated with RM and IQ in schizophrenia (Wang et al., 2008), this study aimed to confirm this observation and extend it to healthy controls in a larger sample using multivariate analyses. It was hypothesized that performance on PM tasks would be independently associated with RM and intelligence in both schizophrenia patients and healthy controls.

Materials and Methods

Study Setting and Participants

This study was part of a large-scale project on neurocognitive deficits in schizophrenia patients, the methods of which have been described elsewhere in detail (Shum et al., 2004; Ungvari et al., 2008). Briefly, schizophrenia inpatients in the long-term rehabilitation unit of a university-affiliated hospital in Hong Kong were invited to participate if they were aged between 18 and 50 years, had a diagnosis of DSM-IV schizophrenia lasting at least 2 years, were taking antipsychotic drugs of less than 800 mg/day in chlorpromazine equivalents (Taylor, Paton, & Kerwin, 2005), had received at least a primary level of education, and were able to understand the study requirements. The exclusion criteria were electroconvulsive therapy conducted within the past year, concomitant administration of antidepressant or benzodiazepine drugs, benzhexol in doses exceeding 6 mg/day, a history of or current significant medical or neurological condition, or past or current significant drug/alcohol abuse. Healthy controls matched according to age, sex, and educational level were selected from the Hong Kong College of Technology.

The study protocol was approved by the Human Research and Ethics Committee of the Chinese University of Hong Kong-New Territories East Cluster. Written consent was obtained from all participants.

Outcome Measures and Assessment

Two subscores of the Brief Psychiatric Rating Scale (BPRS) were employed to measure the severity of psychotic symptoms: 1. Positive symptoms of conceptual disorganization, suspiciousness, hallucinatory behavior, and unusual thought content; 2. Negative symptoms of emotional withdrawal, motor retardation, blunted affect, and disorientation.

The following neuropsychological test battery was employed to measure intelligence, executive functioning, PM and RM: (a) Raven's Progressive Matrices for intelligence (RPM; Raven, 1965); (b) the Wisconsin Card Sorting Test (WCST; Heaton, Chelune, Talley, Kay, & Curtis, 1993), the Design Fluency Test (DFT; Jones-Gotman & Milner, 1977), and the Tower of London-4 Disk (TOL-4D; Shum et al., 2009; Tunstall, 1999) to measure executive functions; (c) the immediate and delayed Logical Memory subtests of the WMS-R (Wechsler, 1987) to assess RM; and (4) a set of computerized time-, event-, and activity-based PM tasks developed on the basis of the classical dual-task experimental paradigm (Einstein, McDaniel, Richardson, Guynn, & Cunfer, 1995).

A detailed account of the PM tasks used in this study has been provided elsewhere (Shum et al., 2004; Ungvari et al., 2008). Briefly, each task lasted for about 25 min. The participants were told that they would undertake an ongoing task by themselves, with the research assistant (RA) in the next room. The ongoing task was a three-choice general knowledge test presented on a computer screen. Participants were asked to read the questions appearing on the screen and then to respond by pressing the corresponding answer key on the keyboard. They were given feedback for each question and their cumulative score of correct answers was shown at the bottom of the screen. (a) In the time-based condition, participants were required to contact the RA every 5 min using an intercom to inform him of the cumulative correct score on the general knowledge task. To keep track of the time, they could press the space bar on the computer bringing a digital clock on the screen for 2 s. They were allowed to press the space bar at any time for reference as often as they liked. The measure obtained in this task was the percentage of correct prospective responses at 5, 10, 15, 20 and 25 min. Responses within ±20 s of these times were regarded as correct. (b) In the event-based condition, participants were required to contact the RA using the intercom whenever they saw the word “police” written in any position of the general knowledge questions of the ongoing task. Five questions with the target word were scheduled to appear at the 2nd, 8th, 12th, 16th, and 22nd minute. The measure obtained was the percentage of correct responses. (c) In the activity-based condition, the general knowledge task was broken into five 5-min subsections. At the end of each subsection, the computer screen turned blank and participants were required to contact the RA via the space bar to reactivate the computer. The measure obtained was the percentage of correct responses.

The neuropsychological tests were administered first, followed by the PM tasks, and the BPRS. Testing usually took 2–3 hr and was completed within 2 consecutive days for the patients and within a day for the controls.

Statistical Analysis

Data were analyzed using SPSS for Windows (Version 13.0). The associations of each type of PM tasks with intelligence and RM were analyzed using Spearman's rank correlation analyses. To control for the confounding influence of covariates that were significantly associated with PM types, partial correlation analysis was carried out. Based on previous results (Shum et al., 2004; Ungvari et al., 2008), the DFT (free response and fixed response conditions), and TOL total score (total score), age, and education were entered as covariates for the controls, and the DFT (free response and fixed response conditions), TOL total score, and WCST scores (total correct, total and perseverative errors, categories completed, and trials needed to complete the first category), age, length of illness, and use of anticholinergic medications were entered as covariates for the patients. The normality of distributions for the continuous variables was checked with the one-sample Kolmogorov–Smirnov test. Two-tailed tests were used in all analyses, with the significance level set at 0.05.

Results

Table 1 shows the basic socio-demographic and clinical factors separately for the 110 patients and the 110 healthy controls. Table 2 shows the relationship between intelligence and RM and performance on the three PM tasks based on the bivariate and partial correlation analyses. For the controls, performance on the time-based PM task was significantly associated with both WMS-R Logical Memory subtest scores (delayed and immediate), whereas that on the event-based PM task was associated only with the immediate WMS-R Logical Memory subtest score. After controlling for the confounding effects of previously reported significant covariates (Shum et al., 2004; Ungvari et al., 2008), only the association between the time-based PM task and the WMS-R Logical Memory scores (delayed and immediate) remained significant.

Table 1.

Demographic and clinical characteristics of schizophrenia patients and healthy controls

 Patients (n = 110)
 
Controls (n = 110)
 
 N N 
Men 72 65.5 73 66.4 
Use of anticholinergics 63 57.3 — — 
 Mean SD Mean SD 
Age (year) 31.8 7.4 31.5 7.5 
Education (year) 10.2 2.3 10.4 2.2 
Length of illness (year) 8.5 5.3 — — 
Time-based Prospective Memory 38.2 34.2 80.9 26.2 
Event-based Prospective Memory 75.3 31.9 93.0 12.8 
Activity-based Prospective Memory 92.6 17.2 a — 
Raven's Progressive Matrices 41.5 11.3 50.3 7.1 
Wechsler Memory Scale (Revised)-Logical Memory, delayed 16.8 8.6 29.2 7.0 
Wechsler Memory Scale (Revised)-Logical Memory, immediate 21.0 7.8 31.6 6.3 
Tower of London 19.2 3.8 21.6 2.2 
Design Fluency Test-free response 7.4 8.1 13.9 7.8 
Design Fluency Test-fixed response 7.1 5.6 15.8 7.4 
Wisconsin Card Sorting Test-total correct 61.4 17.4 70.8 10.6 
Wisconsin Card Sorting Test-total errors 55.6 28.4 28.8 20.4 
Wisconsin Card Sorting Test-perseverative errors 34.3 25.4 14.4 12.1 
Wisconsin Card Sorting Test-categories completed 3.2 2.4 5.3 1.5 
Wisconsin Card Sorting Test-trials needed to complete the first category 51.4 48.9 19.4 22.0 
 Patients (n = 110)
 
Controls (n = 110)
 
 N N 
Men 72 65.5 73 66.4 
Use of anticholinergics 63 57.3 — — 
 Mean SD Mean SD 
Age (year) 31.8 7.4 31.5 7.5 
Education (year) 10.2 2.3 10.4 2.2 
Length of illness (year) 8.5 5.3 — — 
Time-based Prospective Memory 38.2 34.2 80.9 26.2 
Event-based Prospective Memory 75.3 31.9 93.0 12.8 
Activity-based Prospective Memory 92.6 17.2 a — 
Raven's Progressive Matrices 41.5 11.3 50.3 7.1 
Wechsler Memory Scale (Revised)-Logical Memory, delayed 16.8 8.6 29.2 7.0 
Wechsler Memory Scale (Revised)-Logical Memory, immediate 21.0 7.8 31.6 6.3 
Tower of London 19.2 3.8 21.6 2.2 
Design Fluency Test-free response 7.4 8.1 13.9 7.8 
Design Fluency Test-fixed response 7.1 5.6 15.8 7.4 
Wisconsin Card Sorting Test-total correct 61.4 17.4 70.8 10.6 
Wisconsin Card Sorting Test-total errors 55.6 28.4 28.8 20.4 
Wisconsin Card Sorting Test-perseverative errors 34.3 25.4 14.4 12.1 
Wisconsin Card Sorting Test-categories completed 3.2 2.4 5.3 1.5 
Wisconsin Card Sorting Test-trials needed to complete the first category 51.4 48.9 19.4 22.0 

aControls had perfect performance on the activity-based task, therefore they were not listed.

Table 2.

Bivariate and partial correlation analyses of the association between the prospective memory (PM) subtypes and intelligence and retrospective memory (RM)

 Time-based PM
 
Event-based PM
 
Activity-based PM
 
 r p ra pa r p ra pa r p rb pb 
Controls 
 Raven's Progressive Matrices 0.17 0.08 0.09 0.37 −0.02 0.87 −0.11 0.28 — — — — 
 Wechsler Memory Scale (Revised)-Logical Memory, delayed 0.28 0.003 0.24 0.02 0.17 0.08 0.10 0.33 — — — — 
 Wechsler Memory Scale (Revised)-Logical Memory, immediate 0.28 0.003 0.24 0.01 0.19 0.047 0.11 0.26 — — — — 
 r p rc pc r p rc pc r p rc pc 
Patients 
 Raven's Progressive Matrices 0.38 <0.001 0.08 0.44 0.34 <0.001 0.23 0.023 0.17 0.08 0.07 0.47 
 Wechsler Memory Scale (Revised)-Logical Memory, delayed 0.23 0.02 0.09 0.40 0.22 0.02 0.12 0.25 0.07 0.46 −0.001 0.99 
 Wechsler Memory Scale (Revised)-Logical Memory, immediate 0.27 0.004 0.09 0.40 0.26 0.006 0.11 0.27 0.07 0.49 −0.04 0.73 
 Time-based PM
 
Event-based PM
 
Activity-based PM
 
 r p ra pa r p ra pa r p rb pb 
Controls 
 Raven's Progressive Matrices 0.17 0.08 0.09 0.37 −0.02 0.87 −0.11 0.28 — — — — 
 Wechsler Memory Scale (Revised)-Logical Memory, delayed 0.28 0.003 0.24 0.02 0.17 0.08 0.10 0.33 — — — — 
 Wechsler Memory Scale (Revised)-Logical Memory, immediate 0.28 0.003 0.24 0.01 0.19 0.047 0.11 0.26 — — — — 
 r p rc pc r p rc pc r p rc pc 
Patients 
 Raven's Progressive Matrices 0.38 <0.001 0.08 0.44 0.34 <0.001 0.23 0.023 0.17 0.08 0.07 0.47 
 Wechsler Memory Scale (Revised)-Logical Memory, delayed 0.23 0.02 0.09 0.40 0.22 0.02 0.12 0.25 0.07 0.46 −0.001 0.99 
 Wechsler Memory Scale (Revised)-Logical Memory, immediate 0.27 0.004 0.09 0.40 0.26 0.006 0.11 0.27 0.07 0.49 −0.04 0.73 

aPartial correlation analysis using DFT (free response and fixed response conditions), TOL (total score), age, and education as covariates.

bCorrelations between the activity-based PM task and intelligence and RM could not be calculated because the controls performed perfectly on this task.

cPartial correlation analysis using DFT (free response and fixed response conditions), TOL (total score), WCST (total correct, total errors, perseverative errors, categories completed, and trials needed to complete the first category), age, length of illness, and use of anticholinergic medication as covariates.

For the patients, performance on both the time- and event-based PM tasks was significantly associated with Raven's Progressive Matrices scores and WMS-R Logical Memory subtest scores (delayed and immediate). After controlling for the confounding effects of covariates, however, only the association between performance on the event-based PM task and the Raven's Progressive Matrices scores remained.

Discussion

We could only partially confirm the hypothesis that performance on PM tasks would have independent associations with RM and intelligence in both schizophrenia patients and healthy controls. Similar to earlier findings (Altgassen et al., 2008; Wang et al., 2008), both time- and event-based PMs were significantly associated with RM in the patients and controls in this study. Neuropsychological evidence (Einstein & McDaniel, 1996) has been found to demonstrate that PM also involves a retrospective component. The prospective component is the process of being aware of the appropriate moment that an intended action must be initiated and performed, whereas the retrospective component refers to the process of retrieving the intention content and the specific action that must be executed. The retrospective component of PM is mediated by RM capability (Kliegel, Jager, Altgassen, & Shum, 2008). Intelligence was associated with both the time- and event-based PM tasks in the patients, thus replicating previous findings (Wang et al., 2008). We hypothesize that this association was not observed in the controls perhaps because only non-verbal intelligence was measured by the RPM in this study. Therefore, the results might not have been the same if verbal intelligence had been included. The lack of significant findings on the activity-based PM task in both groups might be the result of a ceiling effect.

In the partial correlation analyses, the associations between time- and event-based PM and RM were eliminated in the patients, and only time-based PM remained associated with RM in the controls. Similarly, only event-based PM was associated with intelligence in the patient group. Morrison, O'Carroll, and McCreadie (2006) found that non-verbal intelligence measured by RPM declined over time in schizophrenia, which might account for the positive association between event-based PM and intelligence in the patients.

A key finding of this study is that time-based PM is independently associated with RM in healthy controls, but not in schizophrenia patients. This discrepancy might be due to impairment of the retrospective component of PM in schizophrenia patients, that is, their ability to retrieve the intention and the specific action that should be executed, leading to poorer PM performance. Woods et al. (2007) found that the intention formation and maintenance stages of PM were relatively intact in schizophrenia patients, and concluded that the PM impairment in schizophrenia might arise at the cue detection and intention retrieval stages. Wang et al. (2008) also suggested that PM deficit was most likely due to impairment of cue detection and intention retrieval. The lack of independent association between RM and PM in patients of this study lends support to their findings.

The findings of this study have clinical implications for schizophrenia patients with PM deficits. The impaired ability of retrieving the intention content and the corresponding action could be alleviated by providing cues in daily life; for example, having an automated phone call or text message to remind patients to take their medication.

Our findings suggest that some illness-related effects influence the associations of PM with RM and intelligence in schizophrenia patients and this is more prominent in time-based PM. A possible reason is that time-based PM involves more neuropsychological processes, that is, it requires self-initiated retrieval and the subsequent interruption of an ongoing activity, places more demand on the prefrontal cortex, and does not benefit from environmental cues for the retrieval of retrospectively encoded intentions when compared with event- and activity-based PM (Einstein et al., 1995; Kvavilashvili & Ellis, 1996).

In conclusion, no independent relationship was found between PM and RM, and only a weak association with intelligence, in schizophrenia patients. However, the lack of such an association does not necessarily mean that changes in RM and intelligence have no affect on PM; rather, it points to the complexity of the neuropsychological processes underlying PM.

Funding

This study was sponsored by a Hong Kong Research Grants Council Earmarked Research Grant.

Conflict of Interest

None declared.

Acknowledgements

The authors are grateful to Mr. Jacky Yuen for his assistance in data collection and to the late Dr. Jin Pang Leung for his contribution to setting up the study.

References

Altgassen
M.
Kliegel
M.
Rendell
P.
Henry
J.
Zollig
J.
Prospective memory in schizophrenia: The impact of varying retrospective-memory load
Journal of Clinical and Experimental Neuropsychology
 , 
2008
, vol. 
30
 (pg. 
777
-
788
)
Einstein
G. O.
McDaniel
M. A.
Normal aging and prospective memory
Journal of Experimental Psychology, Learning, Memory and Cognition
 , 
1990
, vol. 
16
 (pg. 
717
-
726
)
Einstein
G. O.
McDaniel
M. A.
Brandimonte
M.
Einstein
G. O.
McDaniel
M. A.
Retrieval processes in prospective memory: Theoretical approaches and some new empirical findings
Prospective memory: Theory and applications
 , 
1996
Hillsdale, NJ
Lawrence Erlbaum Associates
(pg. 
115
-
142
)
Einstein
G. O.
McDaniel
M. A.
Richardson
S. L.
Guynn
M. J.
Cunfer
A. R.
Aging and prospective memory: Examining the influences of self-initiated retrieval processes
Journal of Experimental Psychology. Learning, Memory, and Cognition
 , 
1995
, vol. 
21
 (pg. 
996
-
1007
)
Heaton
R. K.
Chelune
G. J.
Talley
J. L.
Kay
G. G.
Curtis
G.
Wisconsin Card Sorting Test (WCST) Manual: Revised and expanded
 , 
1993
 
Odessa, FI: Psychological Assessment Resources, Inc
Jones-Gotman
M.
Milner
B.
Design fluency: The invention of nonsense drawings after focal cortical lesions
Neuropsychologia
 , 
1977
, vol. 
15
 (pg. 
653
-
674
)
Kliegel
M.
Jager
T.
Altgassen
M.
Shum
D.
Kliegel
M.
McDaniel
M. A.
Einstein
G. O.
Clinical neuropsychology of prospective memory
Prospective memory: Cognitive, neuroscience, developmental, and applied perspectives
 , 
2008
Mahwah, NJ
Lawrence Erlbaum Associate
(pg. 
283
-
308
)
Kliegel
M.
Martin
M.
McDaniel
M. A.
Einstein
G. O.
Complex prospective memory and executive control of working memory: A process model
Psychologische Beiträge
 , 
2002
, vol. 
44
 (pg. 
303
-
318
)
Kumar
D.
Nizamie
H. S.
Jahan
M.
Event-based prospective memory in schizophrenia
Journal of Clinical and Experimental Neuropsychology
 , 
2005
, vol. 
27
 (pg. 
867
-
872
)
Kumar
D.
Nizamie
S. H.
Jahan
M.
Activity-based prospective memory in schizophrenia
Clinical Neuropsychologist
 , 
2007
, vol. 
19
 (pg. 
1
-
10
)
[PubMed]
Kvavilashvili
L.
Ellis
J.
Brandimonte
M.
Einstein
G. O.
McDaniel
M. A.
Varieties of intention: Some distinctions and classification
Prospective memory: Theory and applications
 , 
1996
Mahwah, NJ
Lawrence Erlbaum
(pg. 
23
-
52
)
Morrison
G.
O'Carroll
R.
McCreadie
R.
Long-term course of cognitive impairment in schizophrenia
British Journal of Psychiatry
 , 
2006
, vol. 
189
 (pg. 
556
-
557
)
Raven
J. C.
Guide to using the coloured progressive matrices, H.K.
 , 
1965
London
Lewis
Shum
D.
Gill
H.
Banks
M.
Maujean
A.
Griffin
J.
Ward
H.
Planning ability following moderate to severe traumatic brain injury: Performance on a 4-disk version of the Tower of London
Brain Impairment
 , 
2009
, vol. 
10
 (pg. 
320
-
324
)
Shum
D.
Ungvari
G. S.
Tang
W. K.
Leung
J. P.
Performance of schizophrenia patients on time-, event-, and activity-based prospective memory tasks
Schizophrenia Bulletin
 , 
2004
, vol. 
30
 (pg. 
693
-
701
)
[PubMed]
Taylor
D.
Paton
C.
Kerwin
R.
The Maudsley Prescribing Guidelines (2005–2006)
 , 
2005
Abingdon
Taylor & Francis
Tunstall
J.
Improving the utility of the Tower of London: A neuropsychological test of planning
 , 
1999
Brisbane, Australia
Griffith University
 
Unpublished MPhil thesis
Ungvari
G. S.
Xiang
Y. T.
Tang
W. K.
Shum
D.
Prospective memory and its correlates and predictors in schizophrenia: An extension of previous findings
Archives of Clinical Neuropsychology
 , 
2008
, vol. 
23
 (pg. 
613
-
622
)
Wang
Y.
Chan
R. C.
Hong
X.
Ma
Z.
Yang
T.
Guo
L.
, et al.  . 
Prospective memory in schizophrenia: Further clarification of nature of impairment
Schizophrenia Research
 , 
2008
, vol. 
105
 (pg. 
114
-
124
)
Wechsler
D.
Wechsler Memory Scale-Revised
 , 
1987
San Antonio, TX
Psychological Corporation
Woods
S. P.
Twamley
E. W.
Dawson
M. S.
Narvaez
J. M.
Jeste
D. V.
Deficits in cue detection and intention retrieval underlie prospective memory impairment in schizophrenia
Schizophrenia Research
 , 
2007
, vol. 
90
 (pg. 
344
-
350
)