This article was downloaded by: [Gunnar Benjamin Scheibner]On: 06 February 2014, At: 15:28Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Aging, Neuropsychology, and Cognition:A Journal on Normal and DysfunctionalDevelopmentPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/nanc20

Memory control beliefs and everydayforgetfulness in adulthood: Theeffects of selection, optimization, andcompensation strategiesGunnar Benjamin Scheibner a & Janet Leathem aa Department of Psychology , Massey University , Wellington , NewZealandPublished online: 02 Nov 2011.

To cite this article: Gunnar Benjamin Scheibner & Janet Leathem (2012) Memory control beliefsand everyday forgetfulness in adulthood: The effects of selection, optimization, and compensationstrategies, Aging, Neuropsychology, and Cognition: A Journal on Normal and DysfunctionalDevelopment, 19:3, 362-379, DOI: 10.1080/13825585.2011.615905

To link to this article: http://dx.doi.org/10.1080/13825585.2011.615905

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

Aging, Neuropsychology, and Cognition, 2012, 19 (3), 362–379http://www.psypress.com/ancISSN: 1382-5585 print; 1744-4128 onlinehttp://dx.doi.org/10.1080/13825585.2011.615905

Memory control beliefs and everydayforgetfulness in adulthood: The effects ofselection, optimization, and compensationstrategies

Gunnar Benjamin Scheibner and Janet Leathem

Department of Psychology, Massey University, Wellington, New Zealand

ABSTRACT

Controlling for age, gender, education, and self-rated health, the present study usedregression analyses to examine the relationships between memory control beliefs andself-reported forgetfulness in the context of the meta-theory of Selective Optimizationwith Compensation (SOC). Findings from this online survey (N = 409) indicate that,among adult New Zealanders, a higher sense of memory control accounts for a 22.7%reduction in self-reported forgetfulness. Similarly, optimization was found to accountfor a 5% reduction in forgetfulness while the strategies of selection and compensa-tion were not related to self-reports of forgetfulness. Optimization partially mediatedthe beneficial effects that some memory beliefs (e.g., believing that memory declineis inevitable and believing in the potential for memory improvement) have on for-getfulness. It was concluded that memory control beliefs are important predictors ofself-reported forgetfulness while the support for the SOC model in the context ofmemory controllability and everyday forgetfulness is limited.

Keywords: Memory control beliefs; Selection; Optimization; Compensation; SOC;Everyday memory.

At the dawn of the twenty-first century, it is common knowledge that theworld’s population is rapidly aging. On a global scale, the observed popula-tion of older individuals (60+) rose from 205 million in 1950 to 606 millionin 2000 (United Nations [UN], 2002). The UN estimates that the number of

Address correspondence to: Mr Gunnar Benjamin Scheibner, Department of Psychology, MasseyUniversity, 24 King Street, Newtown, Wellington 6011, New Zealand. E-mail: gunnar.scheibner@gmail.com

© 2012 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 363

individuals in this age-category may reach the 2 billion mark by 2050. In NewZealand, the 2001 census reported that 450,426 individuals (12% of the entirepopulation) are aged 65 or over, and it is estimated that by 2026 this age-groupmay include almost 1 million people – a fifth of New Zealand’s total popu-lation (Statistics New Zealand, 2009). With exponential growth in the agingpopulation, it follows that scientific attention in the area of cognitive aginghas dramatically increased in recent years.

In general, research has shown that cognitive abilities decline withadvancing age (Colsher & Wallace, 1991; Evans et al., 1993; Schaie, 1996;Wilson, Becket, Bennet, Albert, & Evans, 1999). The onset of significant cog-nitive decline can usually be observed within the fifth decade of life – a timethat, by most accounts, lies within the lifespan period of midlife (Lachman,2001). Compared to other cognitive domains, memory is no exception tothe rule as longitudinal and cross-sectional research pinpoints the onset ofsignificant aging-related decline within the fifth decade of life (Larrabee,Trahan, Curtiss, & Levin, 1988; Schaie, 1996; West & Crook, 1990). Despitethese laboratory findings, high levels of everyday memory failures are fre-quently reported for adults of all ages even those who are considerablyyounger.

A staggering 40% of adults aged 25–85 in the Maastricht AgingStudy considered themselves to be forgetful (Ponds, Commissaris, & Jolles,1997). In line with the generally observed decline in cognitive function-ing, a systematic increase in the prevalence of forgetfulness was reportedwith 29% of the younger age group and 52% in the oldest age groupreporting forgetfulness (Commissaris, Ponds, & Jolles, 1998). Perceivedforgetfulness may have considerable consequences for everyday function-ing and well-being as higher levels of forgetfulness are associated withlower life satisfaction and well-being ratings, and an increased number ofdepression and anxiety symptoms (Mol, van Boxtel, Willems, Verhey, &Jolles, 2009; Mordechovich, Merims, & Giladi, 2010). Subjective mem-ory decline may also be related to mild cognitive impairment and even beindicative of an early stage of Alzheimer’s disease (Glodzik-Sobanska et al.,2007; Jessen et al., 2007; Treves, Verchovsky, Klimovitzky, & Korczyn,2005).

One widely held belief is that physiological deterioration in the brainresults in memory deficits that are inevitable and uncontrollable (Poon, 1985),and that memory will decline irrespective of one’s efforts to maintain itsfunction. In contrast, memory may also be viewed as a body of skills thatcan be developed and maintained over time. The current study examines thelatter conception of memory and aging – focusing on the beneficial effectsthat beliefs about memory may have on individual’s perceptions of everydayforgetfulness.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

364 GUNNAR BENJAMIN SCHEIBNER AND JANET LEATHEM

MEMORY CONTROL BELIEFS AND MEMORY PERFORMANCE

Throughout the aging process memory performance is known to be affectedby metacognitions. Hertzog and Hultsch (2000, p. 417) suggested three dis-tinct categories of metacognitions including ‘a) knowledge about cognitionand cognitive functions, b) the monitoring of the current state of the cogni-tive system, and c) beliefs about one’s own cognitions’. The current studyexamines metamemory – a specific type of metacognition that relates topeople’s knowledge, awareness, and control of their memory (Nelson &Narens, 1990). Among the most salient aspects of metamemory that mayhave an influence on performance are those about one’s beliefs about mem-ory control. Memory control beliefs are metacognitions that refer to whetheror not one believes that there is something that can be done to sustain orimprove memory (Lachman, 1991). Individuals’ beliefs about their controlover age-related forgetfulness may vary considerably. It has been shownthat older adults tend to believe that forgetfulness is due to irreversibleage-related causes, while younger adults tend to attribute forgetfulness tomemory-extrinsic factors such as tension and emotional problems (Pondset al., 1997), or lack of effort (Lachman, 1990). Thus, as people get older,their beliefs tend to shift from perceptions of forgetfulness as manageableand reversible to beliefs that memory deficits are inevitable, uncontrollable,and age-related.

A sense of control over memory is related to superior memory-testperformance. However, this relationship is mediated by memory strat-egy use, that is, individuals with strong beliefs in memory controllabilityare more likely to use strategies which in turn improve memory perfor-mance (Lachman & Andreoletti, 2006). The current study used the MemoryControllability Inventory (MCI; Lachman, Bandura, Weaver, & Elliot, 1995)to examine memory beliefs. The MCI is a psychometric instrument that exam-ines individuals’ beliefs about their memory ability and controllability acrossfour dimensions including beliefs about one’s current level of memory abil-ity (Preset Ability), confidence in the value memory improvement strategies(Potential for Improvement), beliefs in the controllability of memory by effortexpenditure (Effort Utility), and beliefs about uncontrollable aging-relatedmemory decline (Inevitable Decrement).

SELECTIVE OPTIMIZATION WITH COMPENSATION:ADAPTING TO AGING-RELATED CHANGE

The concept of successful aging has become a major theme within thecognitive aging literature. Lifespan developmental psychology posits thatsuccessful aging denotes a process that encompasses the entire lifespan(Freund, 2008). The framework of selective optimization with compensation

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 365

(SOC; Baltes & Baltes, 1990) defines successful aging as the relative max-imization and attainment of goals, and the minimization and avoidanceof undesired developmental outcomes. SOC suggests key processes (e.g.,selection, optimization, and compensation) that individuals may employ toadapt successfully to developmental changes. Briefly, the process of selec-tion involves developing, choosing, and committing oneself to specific goals.In the context of memory, individuals’ may differ in the degree to which theychoose to improve or maintain their memory function as they age. For exam-ple, person A on the one hand may view aging-related memory decline asa normal and acceptable fact of life and hence makes no effort to compen-sate for declining memory performance. Person B on the other hand mayworry about declining memory and therefore sets memory performance goals(i.e., I want to improve my memory for names). There are two subtypes ofselection: elective selection (i.e., focusing on aiming at desired states [e.g.,to get better at remembering names]) and loss-based selection (i.e., restruc-turing one’s goal system when faced with loss in goal-relevant means [i.e.,it’s ok to forget names at my age]). Optimization denotes the employment ofmeans aimed at goal achievement (e.g., employing memory strategies); andcompensation denotes the employment of means that are necessary to main-tain a given level of functioning when confronted with a loss in goal-relevantmeans (e.g., enrolling in a memory program when previously used memorystrategies become ineffective).

There is ample research evidence from self-report studies across thelife-span suggesting that the selection of goals, striving to attain selectedgoals (optimization), and striving to maintain the goal despite a loss ingoal-relevant means (compensation) are related to various indicants of suc-cessful development. For example, Freund and Baltes (2002) providedevidence for associations between the SOC components and a range ofindicants of subjective well-being including positive emotions, autonomy,environmental mastery, personal growth, positive relations, purpose in life,and self-acceptance. In another study, Gestsdottir and Lerner (2007) foundassociations between SOC and positive youth development and inverse asso-ciations with negative youth development. Wiese, Freund, and Baltes (2000)examined SOC in relation to general well-being, as well as career- andpartnership-related successfulness among young adults. Results showed thatSOC endorsement was associated with all three domains in expected direc-tions. Thus, for a variety of life domains, employment of SOC strategies isassociated with successful development. To the authors’ best knowledge, theSOC model has not been utilized in the contexts of everyday memory andmemory controllability. The current study therefore set out to examine therelationships between individual’s memory control beliefs and experiences ofeveryday forgetfulness in the context of the Theory of Selective Optimizationwith Compensation.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

366 GUNNAR BENJAMIN SCHEIBNER AND JANET LEATHEM

HYPOTHESES OF THE CURRENT STUDY

SOC postulates that individuals who select, optimize, and compensate aremore likely to age successfully. A sense of control is a key marker of suc-cessful aging, while increasing forgetfulness is undoubtedly an undesirableage-related change. As the beneficial effects of memory control beliefs onperformance have been shown to be mediated by memory strategy use, thisresearch examines whether the SOC strategies may also have a mediatingeffect on the control belief/performance relationship.

The following hypotheses were tested:

H1: Stronger beliefs in memory ability and controllability will account for lower ratesof self-reported everyday forgetfulness.

H2: Higher levels of endorsement of the life management strategies of selection,optimization, and compensation will account for lower levels of self-reportedforgetfulness.

H3: Selection, optimization, and compensation strategies mediate the relationshipbetween memory control beliefs and levels of self-reported forgetfulness.

METHOD

Participants

The sample included 409 members of the general population of NewZealanders (241 [58.9%] females and 168 [41.1%] males) with a mean age of48.10 years of age (SD = 12.94; range 18–85). Table 1 provides a descriptionof the sample in terms of age, education, and self-reported health. Of theparticipants, 291 (71.1%) identified themselves as employed, 25 (6.1%) asunemployed, 44 (10.8%) as retired, and 45 (11%) as students (4 people didnot answer the question).

Measures

Background Measures

A questionnaire was used to assess age, gender, and self-ratings ofcurrent health status (i.e., 1 = excellent; 2 = good; 3 = fair; 4 = poor/bad).

Memory Controllability Inventory

The 12-item MCI (Lachman et al., 1995) assesses beliefs about mem-ory ability and controllability. The MCI is comprised of four sub-scales:(1) Present Ability (e.g., ‘I can remember the things I need to’); (2) PotentialImprovement (e.g., ‘I can think of strategies to help me keep up my mem-ory’); (3) Effort Utility (e.g., ‘If I work at it, I can improve my memory’);and (4) Inevitable Decrement (e.g., ‘There’s not much I can do to keep mymemory from going downhill’). A 4-point Likert scale was used as response

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 367

TA

BL

E1.

Mea

nsan

dst

anda

rdde

viat

ions

ofed

ucat

ion

and

self

-rat

edhe

ath

stat

usby

age

grou

p

Tota

lsam

ple

Age

grou

ps

Age

grou

p18

–81

yrs

(n=

409)

18–2

4yr

s(n

=26

)25

–31

yrs

(n=

25)

32–3

8yr

s(n

=30

)39

–45

yrs

(n=

78)

46–5

2yr

s(n

=86

)53

–59

yrs

(n=

80)

60–6

6yr

s(n

=62

)67

–74

yrs

(n=

18)

75–8

1yr

s(n

=4)

Edu

catio

nM

(SD

)3.

14(.

84)

2.96

(.77

)3.

60(.

65)

3.70

(3.7

0)3.

00(.

84)

3.12

(.86

)3.

02(.

86)

3.19

(.87

)2.

89(.

83)

3.00

(1.0

)H

ealth

M(S

D)

2.02

(.73

)2.

38(.

80)

2.00

(.82

)2.

13(.

78)

2.00

(.66

)1.

96(.

78)

1.99

(.67

)2.

02(.

70)

2.00

(.77

)1.

33(.

58)

Not

e:E

duca

tion:

1=

prim

ary

scho

ol;2

=se

cond

ary

scho

ol;3

=un

derg

radu

ate

univ

ersi

tyco

urse

;4=

post

grad

uate

univ

ersi

tyco

urse

.Hea

lth:1

=ex

celle

nt;2

=go

od;

3=

fair

;4=

poor

/ba

d.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

368 GUNNAR BENJAMIN SCHEIBNER AND JANET LEATHEM

format (strongly disagree, disagree, agree, strongly agree). Total scores canrange from 3 to 12 for each subscale, and higher scores denote a more positiveconception of one’s memory capabilities (items were reverse scored wherenecessary). Previous research suggests that the MCI scales possess acceptablelevels of internal reliability and construct validity (Lachman et al., 1995).

Selection, Optimization, and Compensation Questionnaire

The 12-item, domain general, short-form of the Selection, Optimization,and Compensation (SOC; Freund & Baltes, 2002) questionnaire was usedto examine the two aspects of selection (elective, ES and loss-based, LBS),optimization, and compensation. The measure has equivalent psychometricproperties to the 48-item version (Freund & Baltes, 2002). Each of the12 items consisted of two statements, characterizing fictitious individualswho either adhere (Person A) or do not adhere (Person B) to SOC-relatedlife management strategies. For example:

Elective Selection. Person A: ‘I concentrate all my energy on a few things’ orPerson B: ‘I divide my energy among many things’.

Loss-Based Selection. Person A: ‘When things don’t go as well as before, Ichoose one or two important goals’ or Person B: ‘When things don’t go as wellas before, I still try to keep all my goals’.

Optimization. Person A: ‘I keep working on what I have planned until I succeed’or Person B: ‘When I do not succeed right away at what I want to do, I don’t tryother possibilities for long’.

Compensation. Person A: ‘When something in my life isn’t working as well asit used to, I ask others for advice or help’ or Person B: ‘When something in mylife isn’t working as well as it used to, I decide what to do about it myself, withoutinvolving other people’.

Participants were asked to choose which of the two individuals is mostsimilar to them and after selecting the item that best describes their behav-ior, they were asked to rate the degree of similarity between their behaviorand the behavior they selected on a 4-point scale ranging from 1 (a little) to4 (exactly). Thus, scores on each item can range from a 0 (non-SOC strat-egy selected) to 4 (SOC strategy selected and an exact degree of similarityindicated), and total scale scores can range from 0 to 48, with higher scoresdenoting higher levels of SOC endorsement. The SOC questionnaire has ade-quate construct validity (Freund & Baltes, 2002) and a number of studies haveprovided evidence that selection, optimization, and compensation are relatedto various subjective indicators of successful development (e.g., Freund &Baltes, 2002; Gestsdottir & Lerner, 2007; Wiese et al., 2000).

Cognitive Failures Questionnaire (CFQ)-Memory Scales

The full CFQ (Broadbent, Cooper, Fitzgerald, & Parkes, 1982) is a25-item self-report inventory that inquires about a person’s proneness to

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 369

TABLE 2. Items of the CFQ-Memory scale

Item Question

1 Do you forget whether you’ve turned off a light or a fire or locked the door?2 Do you find you forget which way to turn on a road you know well but rarely use?3 Do you fail to see what you want in a supermarket?4 Do you find you forget appointments?5 Do you forget where you put something?6 Do you throw away the thing you want and keep what you meant to throw away?7 Do you forget items to buy at the shop?8 Do you fail to listen to people’s names?9 Do you forget people’s names?

committing cognitive slips and errors in the completion of everyday tasksover the past 6 months. The scale contains four distinct factors including:(1) Distractibility, (2) Memory, (3) Blunders, and (4) Memory for Names.Only the Memory and Memory for Names Scales were used in the currentstudy (see Table 2), and the scores of both scales were combined to serveas a measure of everyday forgetfulness. The response format for the CFQ-Memory scale was a 4-point Likert scale (never, sometimes, often, always)and scores can range from 0 to 36 (higher scores denote higher levels of for-getfulness). The full CFQ has been shown to be a valid and reliable measurein a cohort of healthy older New Zealanders (Knight, McMahon, Green, &Skeaff, 2004). Internal consistency reliability estimates for the Memory andMemory for Names scales of the CFQ have been reported to range from .75 to.86 and .75 to .76, respectively (Wallace, 2004; Wallace, Kass, & Stanny,2002).

Procedure

All participants received an electronic version of an informed consentdocument which stated clearly that: (a) involvement in the study was vol-untary and could be terminated at any time without consequences; (b) noself-identifying information would be collected, thereby safeguarding respon-dents’ anonymity; (c) the data would be kept confidential (i.e., only theresearcher and the computer programmer associated with the psychologydepartment had access to the data-file); and (d) upon completion of thestudy, participants would be given access to a document outlining the study’spurpose and main findings. The survey was located on the website of the uni-versity associated with the researchers. Only the computer programmer wasable to access the data file via a secure password protected entry point.

A number of procedures were used to recruit participants. Firstly, thestudy was advertised online – a Facebook advertisement appeared on personalFacebook interfaces. Upon clicking on the advertisement, participants were

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

370 GUNNAR BENJAMIN SCHEIBNER AND JANET LEATHEM

directed to the information sheet and survey. Secondly, advertisements wereplaced into two local newspapers. These adverts contained brief descriptionsof the study and the details needed to access the online survey. Thirdly,an e-mail message was sent to a graduate student list associated with theresearchers’ institution. The message contained a brief outline of the studyand the link to the online survey. Recipients were asked to forward themessage to friends and acquaintances.

RESULTS

Analysis

Alpha coefficients, means, standard deviations, and correlation coeffi-cients for all psychometric measures are presented in Table 3. Most cor-relations were significant and magnitudes ranged from small to large. Theindicants of SOC were all significantly intercorrelated ranging from .20 (elec-tive selection and compensation) to .52 (optimization and compensation) andaveraged SOC component scores were used to calculate an overall SOC score.The sub-scales of the MCI were also intercorrelated in expected directions,ranging from .29 (present ability and effort utility) to .55 (present ability andpotential for improvement). Associations between the measures of interestand control variables were generally low: r values ranged from .00 (loss-basedselection and age) to .23 (gender and potential for memory improvementbeliefs).

Hypothesis 1

To examine the impact of memory control beliefs on the experienceof forgetfulness, hierarchical regression analysis was conducted. Age, gen-der, education, and self-rated health were entered at step 1 and the four MCIsub-scales were entered at step 2. CFQ-Memory scores served as the depen-dent variable. Step 1 was statistically significant, F(4, 403) = 6.77, p < .001,adjusted R2 = .05. Age (β = −0.12, t = −0.24, p < .05) and self-rated health(β = 0.21, t = 4.20, p < .001) accounted for variation in CFQ-Memory scoreswhile education (β = −0.04, t = −0.73, p = ns) and gender (β = −.03, t =−.63, p = ns) did not. Step 2 accounted for a significant increase in variance(i.e., F change = 32, p < .001; overall F(7, 396) = 20, 43; p < .001, � R2 =.28), present ability (β = −0.41, t = −0.78, p < .001), potential for improve-ment (β = −0.13, t = −2.14, p < .05), and effort utility (β = 0.11, t = 2.24,p < .05) emerged as significant predictors while inevitable decrement (β =0.05, t = 1.05, p = ns) did not. Hypothesis 1 is thus supported: After control-ling for age, education, and self-rated health, three memory control beliefsfacets jointly accounted for a 22.7% reduction in self-reported forgetfulness.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 371

TA

BL

E3.

Mea

ns,s

tand

ard

devi

atio

ns,a

lpha

coef

ficie

nts

and

inte

rcor

rela

tions

ofal

lpsy

chom

etri

cm

easu

res

MSD

12

34

56

78

910

1112

1SO

CE

lect

ive

Sele

ctio

n2.

291.

02(.

58)

2SO

CL

oss-

base

dSe

lect

ion

3.59

1.06

.38∗∗

(.44

)

3SO

CO

ptim

izat

ion

6.22

1.01

.26∗∗

.36∗∗

(.66

)4

SOC

Com

pens

atio

n4.

001.

06.2

0∗∗.2

5∗∗.5

2∗∗(.

46)

5SO

C(C

ompo

site

)22

.66

2.83

.61∗∗

.68∗∗

.80∗∗

.72∗∗

(.71

)6

MC

IPr

esen

tAbi

lity

8.62

1.69

.08

.13∗

.29∗∗

.18∗∗

.25∗∗

(.73

)7

MC

IPo

tent

ial

Impr

ovem

ent

8.60

1.45

.09

.14∗∗

.24∗∗

.18∗∗

.24∗∗

.55∗∗

(.56

)

8M

CI

Eff

ortU

tility

9.00

1.39

.05

.15∗∗

.10∗

.14∗∗

.15∗∗

.29∗∗

.43∗∗

(.69

)9

MC

IIn

evita

ble

Dec

rem

ent

7.00

1.65

.05

−.13

∗∗−.

17∗∗

−.18

∗∗−.

16∗∗

−.41

∗∗−.

52∗∗

−.37

∗∗(.

73)

10C

FQM

emor

y12

.33

2.76

−.00

1−.

05−.

25∗∗

−.10

∗−.

14∗∗

−.49

∗∗−.

33∗∗

−.09

∗.2

4∗∗(.

77)

11C

FQN

ames

4.8

1.18

−.00

5−.

09−.

19∗∗

−.16

∗∗−.

16∗∗

−.34

∗∗−.

30∗∗

−.16

∗∗.2

4∗∗.3

7∗∗(.

65)

12C

FQFo

rget

fuln

ess

17.2

13.

39.0

2−.

08−.

24∗∗

−.14

∗∗−.

16∗∗

−.50

∗∗−.

36∗∗

−.10

∗.2

6∗∗.9

5∗∗.6

5∗∗(.

78)

Not

e:N

=40

9.A

lpha

coef

ficie

nts

are

inpa

rent

hese

s.∗ p

<.0

5;∗∗

p<

.01.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

372 GUNNAR BENJAMIN SCHEIBNER AND JANET LEATHEM

Hypothesis 2

To examine whether SOC strategy endorsement accounts for a reductionin self-reported forgetfulness, hierarchical regression analysis was conducted.Age, gender, education, and self-rated health were entered at step 1, the SOCvariables of elective and loss-based selection, optimization and compensationwere entered at step 2, and CFQ-Memory scores served as the dependentvariable. Step 1 was statistically significant, F(4, 403) = 6.77, p < .001,adjusted R2 = .05. Age (β = −0.12, t = −2.39, p < .05) and self-rated health(β = 0.21, t = 4.20, p < .001) accounted for variation in CFQ-Memory scoreswhile gender (β = −0.04, t = −0.72, p = ns) and education (β = −0.04,t = −0.73, p = ns) did not. Step 2 accounted for a significant increase invariance (i.e., F change = 6.93, p < .001, �R2 = .11; overall F(8, 395) =7.05; p < .001, adjusted R2 = .11). Optimization (β = −0.22, t = −3.56, p <

.001) emerged as a significant predictor while the two measures of selection(ES: β = 0.09, t = 1.70; LBS: β = −0.03, t = −0.48) and compensation(β = −0.03, t = −0.44) did not. Hypothesis 2 is thus partially supported asonly optimization emerged as a significant predictor of CFQ-Memory scores.

Hypothesis 3

Baron and Kenny (1986) explain that a variable may function as amediator if three conditions are met: (1) the independent variable is a sig-nificant predictor of the dependent variable, (2) the mediator is related to thedependent variable, and (3) the independent variable is a significant predictorof the mediator. The above analysis indicated that the first two require-ments were met for the current data. To test whether the third requirementwas met (i.e., whether memory control beliefs predict SOC endorsementscores), overall SOC and MCI scores were computed. Regression analy-sis showed that MCI scores account for a significant amount of variationsin SOC composite scores, F(1, 406) = 23.02, p < .001, R2 = .054, con-firming that the third condition was also met. Mediation occurs when therelationship between the independent variable and the dependent variable isstatistically lower than before the introduction of the mediator variable. TheBaron and Kenny causal step approach for detection of mediation effectshas been criticized for having low statistical power (e.g., Hayes, 2009).By convention, the social sciences usually aspire to obtain a statistical powerof .8 (Cohen, 1988). Fitz and MacKinnon (2007) have provided empiri-cal estimates of sample sizes and effect sizes needed to obtain .8 powerfor several different approaches that may be used to determine media-tion effects. With the current data (e.g., present ability/forgetfulness: r =−.42; potential for improvement/forgetfulness: r = −.35; and inevitabledecrement/forgetfulness: r = .24; N = 409), the conventional power criteriawas met for each of the three mediation models reported below.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 373

Results of the analyses above indicated that of the SOC life managementstrategies, only optimization and compensation qualify as potential mediatorsfor the memory control beliefs/forgetfulness relationship as the measures ofselection were unrelated to forgetfulness (see Table 3). To test for mediation,four multiple regression equations – one for each belief facet – were run.Age, gender, education, and self-rated health were controlled for in step 1.The individual MCI sub-scale scores were entered at step 2, and optimizationand compensation scores were entered at step 3. CFQ-Memory scores servedas the dependent variable. For the effort utility model, step 2 did significantlycontribute to the equation and thus the model is not reported here. Table 4presents the results of the regressions. Step 1 was statistically significant, F(4,403) = 6.77, p < .001, adjusted R2 = .064, but only age (β = −.012, t =−2.39, p < .05) and self-rated health (β = 0.21, t =4.20, p < .001) emergedas significant predictors, while gender (β =–0.03, t = −0.63, p = ns) andeducation (β = −0.04, t = −0.73, p = ns) did not.

Model for Present Ability Beliefs

After entering present ability scores in step 2, the equation significantlyimproved (�R2 = .21, p = .001). The standardized beta coefficient for presentability declined from −.472 to −.442 in step 3, and the addition of optimiza-tion (but not compensation) did significantly improve the model in step 3(�R2 = .019, p <.01), indicating that optimization partially mediated therelationship between present ability beliefs and forgetfulness.

Model for Potential for Improvement Beliefs

After entering potential improvement scores in step 2 the equationsignificantly improved (�R2 = .104, p < .001). The standardized beta coeffi-cient for potential improvement did decrease from −.347 to −.319 in step 3.Compensation did not emerge as significant predictor while optimizationwas significant (�R2 = .036, p < .001), indicating that optimization par-tially mediates the relationship between potential for improvement beliefs andforgetfulness.

Model for Inevitable Decrement

After entering inevitable decrement scores in step 2, the equationsignificantly improved (�R2 = .055, p < .001). The standardized beta coef-ficient for inevitable decrement declined from .240 to .215 in step 3 (�R2

= .145, p < .001). Compensation did not emerge as significant predictorwhile optimization accounted for a small but significant amount for varia-tion, suggesting that optimization partially mediates the relationship betweeninevitable decrement beliefs and forgetfulness.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

374 GUNNAR BENJAMIN SCHEIBNER AND JANET LEATHEM

TABLE 4. Summary of regression analyses for memory beliefs facets, optimization,and compensation as predictors of forgetfulness

�R2 B SE β

Model for Present Ability Beliefs (PA)Step 2 .264∗∗

Age −0.023 0.011 −0.086∗

Gender −0.260 0.298 −0.038Education 0.146 0.199 0.036Health 0.575 0.203 0.124∗∗

PA −0.945 0.088 −0.472∗∗

Step 3 .280∗∗

Age −0.025 0.011 −0.094∗

Gender −0.239 0.295 −0.035Education 0.187 0.177 0.046Heath 0.545 0.201 0.118∗∗

PA −0.885 0.089 −0.442∗∗

Optimization −0.450 0.160 −0.134∗∗

Compensation −0.069 0.153 −0.022Overall Model: F(7, 396) = 23.37, p < .001

Model for Potential for Improvement Beliefs (PI)Step 2 .107∗∗∗

Age −0.025 0.012 −0.096∗

Gender 0.251 0.325 0.037Education 0.195 0.194 0.048Heath 0.748 0.216 0.161∗∗

PI −0.810 0.115 −0.347∗∗∗

Step 3 .189∗∗∗

Age −0.028 0.012 −0.107∗

Gender 0.238 0.319 0.035Education 0.272 0.191 0.067Heath 0.712 0.213 0.149∗∗

PI −0.747 0.114 −0.319∗∗

Optimization −0.591 0.169 −0.176∗∗

Compensation −0.115 0.162 −0.036Overall Model: F(7, 396) = 14, 42, p < .001

Model for Inevitable Decrement Beliefs (ID)Step 2 .107∗∗∗

Age −0.025 0.012 −0.097∗

Gender −0.044 0.330 −0.006Education 0.010 0.196 0.002Heath 0.897 0.219 0.194∗∗

ID 0.493 0.099 0.240∗∗

Step 3 .145∗∗

Age −0.028 0.012 −0.108∗

Gender −0.035 0.323 −0.005Education 0.104 0.193 0.026Heath 0.825 0.217 0.178∗∗

ID 0.441 0.098 0.215∗∗∗

Optimization −0.644 0.173 −0.192∗∗

Compensation −0.103 0.167 −0.032Overall Model: F(7, 396) = 10, 74, p < .001

Note: ∗p < .05; ∗∗p < .01; ∗∗∗p < .001. PA, Present Ability Beliefs; PI, Potential for ImprovementBeliefs; ID, Inevitable Decrement Beliefs.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 375

DISCUSSION

Previous research has shown that the beneficial effects of memory controlbeliefs on memory performance are partially mediated by memory strategyuse. The aim of the present study was to examine the effects of memorycontrol beliefs in the context of everyday memory and to determine whetherthe strategies of selection (elective and loss-based), optimization, and com-pensation are additional mediators for the control/performance relationship.Findings indicate that memory control beliefs are strong predictors of every-day forgetfulness. Specifically, even after controlling for the effects of age,gender, education, and self-rated health, memory control beliefs accountedfor a 22.7% reduction in self-reported forgetfulness. The data indicated thatbeliefs about one’s current memory ability account for most of this effect,followed by beliefs in the potential for memory improvement and beliefsabout the controllability of memory by means of effort expenditure. Despitea significant correlation between inevitable decrement beliefs and forgetful-ness, beliefs about inevitable memory decline had no significant effect onself-reported forgetfulness when all four facts of the MCI were included inthe regression model simultaneously. Nonetheless, the direct path betweeninevitable decrement and forgetfulness was significant in the mediation model– suggesting that beliefs in inevitable memory decline account for variationin self-reported forgetfulness.

The hypothesized relationship between SOC endorsement and self-reported forgetfulness was partially supported. Correlation coefficients sug-gested that higher scores on the measures of optimization and compensa-tion, but not selection, were associated with lower levels of forgetfulness.Subsequent regression analysis suggested that only optimization accountedfor variation in forgetfulness. Using Baron and Kenny’s (1986) guide-lines, the test for mediation for the control belief/forgetfulness relationshipincluded optimization and compensation but not selection as potential medi-ators. Regression analyses showed that optimization, if only to a smallextent, mediates the relationship between control beliefs and forgetting, whilecompensation had no mediating effect.

A key question that arises from these findings is why only optimiza-tion and to some degree compensation, but not selection, related to memorycontrol beliefs and self-reported forgetfulness? One possible answer residesin methodological limitations. A likely explanation for the lack of relation-ships between selection and other key variables may be that SOC was notassessed specifically for the domain of memory. The instructions and itemsof the domain general SOC questionnaire are frequently modified to exam-ine specific domains. For example, recent research has used work-, family-,partnership-, and exercise-specific SOC questionnaires (i.e., Wiese et al.,2000; Young, Baltes, & Pratt, 2007; Ziegelmann & Lippke, 2007). Wiese

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

376 GUNNAR BENJAMIN SCHEIBNER AND JANET LEATHEM

et al. (2000) indicated that domain-specific SOC scales may provide betterpredictive power, and thus, future research should use domain-specific SOCquestionnaires. In the planning stages of the current study, consideration wasgiven to the possibility of modifying the instructions of the domain-generalSOC questionnaire to create a memory-specific version. However, we decidedagainst this as it appeared to be a rather unusual request to think about individ-ual SOC items (e.g., ‘I concentrate all my energy on a few things’ [person A]versus ‘I divide my energy among many things’ [person B]) in terms of one’smemory. We believe that a memory-specific SOC questionnaire could be con-structed, but would involve modifications of the SOC items rather than theinstructions – a considerable research project in itself. While this was out-side the scope of the current research, a task for future researchers may be todevelop a SOC questionnaire that has adequate face and construct validity forthe domain of memory.

Another possibility that may account for the low predictive value ofSOC is the use of self-report measurement. Although SOC endorsement isnot related to social desirability (Freund & Baltes, 2002), it is impossible todetermine whether participants actually use SOC strategies or simply stateda preference for these strategies. Similarly, it is also unclear how well thestrategies are executed if they are used. It is conceivable that individuals whoreport high levels of selection and compensation, despite infrequent or poorexecution of these strategies, may have attenuated the expected relationshipswith other key variables. Because of these uncertainties, future researchersmay be well advised to employ behavioral rather than self-report measuresof SOC as has been done in previous research (e.g., Gignac, Cott, & Badley,2002; Li, Lindenberger, Freund, & Baltes, 2001).

Most alpha coefficients (see Table 2) for the SOC measures areextremely low in the current study. While this may be indicative of lowinternal consistency, it must be stressed that high homogeneity is not anexpectation for the elements of SOC as they in themselves denote ratherheterogeneous phenomena (e.g., optimization may refer to increased effortexpenditure, acquiring new skills, modeling successful others, etc.; seeBaltes, 1997 for a discussion of this issue).

Finally, it must be stressed that the current data are based on a samplewith above average levels of educational achievement. Due to the samplingprocedure (that included a message to a postgraduate university student emaillist), the educational achievement of the sample was extremely high, with71% of the sample reporting either undergraduate or postgraduate universityeducation. The generalizability of the findings is therefore limited as the datamay not be reflective of the general population of New Zealanders in termsof education.

In conclusion, while memory control beliefs were found to be impor-tant predictors of self-reported everyday forgetfulness, only limited support

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 377

was found for the hypothesis that SOC endorsement acts as a mediator forthis relationship. These findings complement previous research that indi-cated that a higher sense of memory control improves memory performancein laboratory-based memory tests. The current data therefore strengthen theargument that memory interventions that aim to increase objective/subjectivememory performance may effectively improve memory by modifying mem-ory control beliefs (i.e., Lachman & Andreoletti, 2006). The possibilitythat the limited support for the Theory of Selective Optimization withCompensation was due to methodological issues was explored and a numberof limitations were discussed. More research is needed to examine the SOCmodel in specific cognitive domains such as memory. As previous researchhas suggested, we close by emphasizing that behavioural, domain-specific(i.e., memory) SOC assessment may lead to increased predictive power of theSOC model.

Original manuscript received 12 January 2011Revised manuscript accepted 13 August 2011

First published online 2 November 2011

REFERENCES

Baltes, P. B. (1997). On the incomplete architecture of human ontogeny: Selection, optimiza-tion, and compensation as foundation of developmental theory. American Psychologist, 52,366–380.

Baltes, P. B., & Baltes, M. M. (1990). Psychological perspectives on successful aging: Themodel of selective optimization with compensation. In P. B. Baltes & M. M. Baltes (Eds.),Successful aging: Perspectives from the behavioural sciences (pp. 1–34). New York, NY:Cambridge University Press.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in socialpsychological research: Conceptual, strategic, and statistical considerations. Journal ofPersonality and Social Psychology, 51, 1173–1182.

Broadbent, D. E., Cooper, P. F., Fitzgerald, P., & Parkes, K. R. (1982). The Cognitive FailuresQuestionnaire (CFQ) and its correlates. British Journal of Clinical Psychology, 21, 1–16.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale,NJ: Erlbaum.

Commissaris, C. J. A. M., Ponds, R. W. H. M., & Jolles, J. (1998). Subjective forgetfulness in anormal Dutch population: Possibilities for health education and other interventions. PatientEducation and Counselling, 34, 25–32.

Evans, D. A., Becket, L. A., Albert, M. S., Hebert, L. E., Scherr, P. A., Funkenstein, H. H., &Taylor, J. O. (1993). Level of education and change in cognitive function in a communitypopulation of older persons. Annals of Epidemiology, 5, 71–77.

Fitz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediated effect.Psychological Science, 18(3), 233–239.

Freund, A. M. (2008). Successful aging as management of resources: The role of selection,optimization, and compensation. Research in Human Development, 5(2), 94–106.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

378 GUNNAR BENJAMIN SCHEIBNER AND JANET LEATHEM

Freund, A. M., & Baltes, P. B. (1998). Selection, optimization, and compensation as strategiesof life management: Correlations with subjective indicators of successful aging. Psychologyand Aging, 13(4), 531–543.

Freund, A. M., & Baltes, P. B. (2002). Life-management strategies of selection, optimiza-tion, and compensation: Measurement by self-report and construct validity. Journal ofPersonality and Social Psychology, 82(4), 642–662.

Gestsdottir, S., & Lerner, R. M. (2007). Intentional self-regulation and positive youth devel-opment in early adolescence: Findings from the 4-H Study of Positive Youth Development.Developmental Psychology, 43, 508–521.

Gignac, M. A. M., Cott, C., & Badley, E. M. (2002). Adaptation to disability: Applying selec-tive optimisation with compensation to the behaviors of older adults with osteoarthritis.Psychology and Aging, 17, 520–524.

Glodzik-Sobanska, L., Reisberg, B., De Santi, S., Babb, J. S., Pirraglia, E., & Rich, K. E., et al.(2007). Subjective memory complaints: Presence, severity, and future outcome in normalolder subjects. Dementia and Geriatric Cognitive Disorders, 24(3), 177–184.

Hayes, A. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the newmillennium. Communication Monographs, 76(4), 408–420.

Hertzog, C., & Hultsch, D. F. (2000). Metacognition in adulthood. In F. M. I. Craik & T. A.Salthouse (Eds.), The handbook of aging and cognition (2nd ed., pp. 417–466). Mahwah,NJ: Erlbaum.

Jessen, F., Wiese, B., Cvetanovska, G., Fuchs, A., Kaduszkiewicz, H., Kölsch, H., et al. (2007).Patterns of memory impairment in the elderly: Associations with memory performance.Psychological Medicine, 37, 1753–1762.

Knight, R. G., McMahon, J., Green, T. J., & Skeaff, C. M. (2004). Some normative andpsychometric data for the geriatric depression scale and the cognitive failures questionnairefrom a sample of healthy older persons. New Zealand Journal of Psychology, 33, 163–170.

Lachman, M. E. (1990). When bad things happen to older people: Age differences inattributional style. Psychology and Aging, 5, 607–609.

Lachman, M. E. (1991). Perceived control over memory aging: Developmental and interven-tion perspectives. Journal of Social Issues, 47, 159–175.

Lachman, M. E. (Ed.). (2001). Handbook of midlife development. New York, NY: Wiley.Lachman, M. E., & Andreoletti, C. (2006). Strategy use mediates the relationship between

control beliefs and memory performance for middle-aged and older adults. Journal ofGerontology: Psychological Sciences, 61B(2), 88–94.

Lachman, M. E., Bandura, M., Weaver, S. L. & Elliot, E. (1995). Assessing memory controlbeliefs: The Memory Controllability Inventory. Aging, Neuropsychology, and Cognition,2 (1), 67–84.

Larrabee, G. J., Trahan, D. E., Curtiss, G., & Levin, H. S. (1988). Normative data for the verbalselective reminding test. Neuropsychology, 2, 173–182.

Li, K. Z. H., Lindenberger, U., Freund, A. M., & Baltes, P. B. (2001). Walking while mem-orizing: Age-related differences in compensatory behavior. Psychological Science, 12,230–237.

Mol, M. E. M., van Boxtel, M. P. J., Willems, D., Verhey, F. R. J., & Jolles, J. (2009). Subjectiveforgetfulness is associated with lower quality of life in middle-aged and young-old individ-uals: A 9-year follow-up in older participants from the Maastricht Aging Study. Aging &Mental Health, 13(5), 699–705.

Mordechovich, Y. M., Merims, H. D., & Giladi, N. (2010). Subjective memory decline inhealthy community-dwelling elders. What does this complain mean? Acta NeurologicaScandinavica, 121, 194–197.

Nelson, T. O., & Narens, L. (1990). Metamemory: A theoretical framework and new findings.Psychology of Learning and Motivation, 26, 125–141.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014

MEMORY CONTROL & EVERYDAY FORGETFULNESS 379

Ponds, R. W., Commissaris, K. J., & Jolles, J. (1997). Prevalence and covariates of subjectiveforgetfulness in a normal population in The Netherlands. International Journal of Aging &Human Development, 45(3), 207–221.

Poon, L. W. (1985). Difference in human memory with aging: Nature, causes, and clini-cal implications. In J. Birren & K. Schaie (Eds.), Handbook of the psychology of aging(2nd ed., pp. 427–462). New York, NY: Van Nostrandt Reinhold.

Schaie, K. W. (1996). Intellectual development in adulthood: The Seattle longitudinal study.Cambridge, UK: University of Cambridge Press.

Statistics New Zealand, Tatauranga Aotearoa (2009). Older New Zealanders – 65 and beyond.Retrieved from http:/www.stats.govt.nz/analytical-reports/oldernzers/intro.htm

Treves, T. A., Verchovsky, R., Klimovitzky, S., & Korczyn, A. D. (2005). Incidence ordementia in patients with subjective memory complaints. International Psychogeriatrics,17(2), 265–273.

Wallace, J. C. (2004). Confirmatory factor analysis of the cognitive failures questionnaire:Evidence for dimensionality and construct validity. Personality and Individual Differences,37, 307–324.

Wallace, J. C., Kass, S. J., & Stanny, C. (2002). Cognitive failures questionnaire revisited:Correlates and dimensions. Journal of General Psychology, 129(3), 238–256.

West, R. L., & Crook, T. H. (1990). Age differences in everyday memory: Laboratoryanalogues of telephone number recall. Psychology and Aging, 5, 520–529.

Wiese, B. S., Freund, A. M., & Baltes, P. B. (2000). Selection, optimization, and compensation:An action-related approach to work and partnership. Journal of Vocational Behavior, 57,273–300.

Wilson, R. S., Becket, L. A., Bennet, D. A., Albert, M. S., & Evans, D. A. (1999). Chancein cognitive function in older persons from a community population: Relation to age andAlzheimer’s disease. Archives of Neurology, 56, 1274–1279.

Young, L. M., Baltes, B. B., & Pratt, A. (2007). Using selection, optimization, and compen-sation to reduce job/family stressors: Effective when it matters. Journal of Business andPsychology, 18, 1–29.

Ziegelmann, J. P., & Lippke, S. (2007). Use of selection, optimization, and compensationstrategies in health self-regulation: Interplay with resources and successful development.Journal of Aging and Health, 19(3), 500–518.

Dow

nloa

ded

by [

Gun

nar

Ben

jam

in S

chei

bner

] at

15:

28 0

6 Fe

brua

ry 2

014