The ecological validity of traditional memory evaluation in relation with controlled memory processes and routinization

Page 1

A

ohtp(pnr©

K

1

uppp1

adt

C

0

Archives of Clinical Neuropsychology 22 (2007) 979–989

The ecological validity of traditional memory evaluation inrelation with controlled memory processes and routinization

Philippe Dubreuil a, Stephane Adam b,c, Nathalie Bier d, Lise Gagnon d,∗a Universite de Sherbrooke, Canada

b Neuropsychology Unit, University of Liege, Belgiumc Memory Clinic, Centre Neurologique et de Revalidation Fonctionnelle (C.N.R.F.), Fraiture-en-Condroz, Belgium

d Research Center on Aging, Universite de Sherbrooke, Canada

Accepted 12 August 2007

bstract

The neuropsychological evaluation of memory by traditional tests raises questions about their ecological validity, as the resultsn these tests often have little relation to the memory complaints. In an attempt to explain this lack of relationship, the present studyad two objectives: (1) explore the ecological superiority of the Process Dissociation Procedure (PDP) over traditional memoryests and (2) explore the effects of routinization on the relationship between memory complaints and memory tests. Thirty-threearticipants aged 55–86 years were given the PDP (memory evaluation), two questionnaires evaluating daily memory complaintsQAM and CDS) and a questionnaire evaluating routinization (EPR). The results indicate that the PDP, with its measure of controlledrocesses, is more ecological than traditional memory tests for elderly people. As well, the participants’ lifestyle (routinized versuson-routinized) influenced their results on memory tests. The results are discussed in relation to neuropsychological evaluation andehabilitation.

2007 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.

eywords: Memory; Controlled processes; Ecological validity; Routinization

. Introduction

Traditionally, explicit recall tests are used for the neuropsychological evaluation of memory. These tests are madep of “classic” free recall, cued recall and recognition tasks. Given that these tests were designed to detect memoryroblems, one would expect that their results would be related to the degree of memory complaints manifested byeople in their daily lives. However, a number of studies that evaluated subjects’ complaints of everyday memoryroblems report only weak associations with the subjects’ results on classic memory tests (Jacoby, Jennings, & Hay,996; Pearman & Storandt, 2004; Reid & Maclullich, 2006). This lack of association raises numerous questions.

One hypothesis evoked to explain this situation proposes that the measurements made during classic memory tests

re not “pure” (Jacoby, 1991) and do not allow one to distinguish implicit and explicit memory, as they purport too (Jacoby et al., 1996). The presence of certain ambiguous data in the literature has led several authors to considerhat explicit memory tasks are contaminated by implicit memory processes (Jacoby, Toth, & Yonelinas, 1993; Voss &

∗ Corresponding author at: Research Center on Aging, Sherbrooke Geriatric University Institute, 1036 rue Belvedere Sud, Sherbrooke, Quebec,anada J1H 4C4. Tel.: +1 819 821 1170; fax: +1 819 829 7141.

E-mail address: [email protected] (L. Gagnon).

887-6177/$ – see front matter © 2007 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.acn.2007.08.002

Page 2

980 P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989

Paller, 2007). Conversely, implicit memory tasks are also said to be influenced, to a certain degree, by explicit memoryprocesses (Butler & Berry, 2001; Fay, Isingrini, & Pouthas, 2005; Holender, 1986; Jacoby, 1991; Mitchell & Bruss,2003; Reingold & Merikle, 1990; Richardson-Klavehn & Bjork, 1988). Thus, more and more researchers tend to adoptthe view that there is a memory system wherein all memory tasks depend, in varying proportions, on processes thatare both automatic and controlled (Adam, 2003; Jacoby, 1991; Jacoby et al., 1996; Toth, Reingold, & Jacoby, 1994).By automatic processes we mean processes involving aspects of memory that are unconscious and non-intentionaland require few or no attentional resources. These processes are generally associated with implicit memory tasks.Controlled processes are intentional retrieval processes, which are conscious and require more attentional resources.These processes are generally associated with explicit memory tasks. It was within this framework of a single memorysystem that Jacoby (1991) developed the Process Dissociation Procedure (PDP), which allows one to separate andquantify the respective contributions of automatic and controlled processes involved in the same memory task.

The PDP compares a person’s performance in a condition in which the automatic and controlled processes act inconcert (inclusion condition) to his or her performance in a condition in which the same processes act in opposition(exclusion condition). This test can be administered using a trigram completion task (Adam, Van der Linden, Collette,Lemauvais, & Salmon, 2005; Jacoby et al., 1993; Ste-Marie, Jennings, & Finlayson, 1996; Toth et al., 1994). In theinclusion condition, word stems are presented to participants, who must complete them with six-letter words theyhave previously memorized. If they are incapable of retrieving the target word, they must complete the stem withthe first six-letter word that comes to mind. The processes act in concert: The participant may complete the stem bythe target word either because it was retrieved consciously (C), or because the item came to mind automatically (A)when conscious retrieval failed (1–C). The latter condition is therefore equivalent to a contaminated measure, as inclassic memory tests. The probability of completing the stem with the target word is then Inclusion = C + A(1–C). Inthe exclusion condition, the procedure is identical, except that the retrieval instructions ask participants to use the stemas an indicator to find a six-letter word that is not the same as the one presented during the memorization phase. Theprocesses then work in opposition: the controlled processes work to avoid an error while the automatic processes tendto produce an erroneous response. Thus, the participant erroneously completes a stem by a word from the studied list ifit comes to mind automatically (A), without conscious retrieval from its previous presentation (1–C). The probabilityof producing an error can then be described by Exclusion = A(1–C). From these two equations, we can quantify therespective contributions of controlled and automatic processes to participants’ performance by a simple algebraictransformation: C = Inclusion − Exclusion, and A = Exclusion/(1 − C).

However, it should be noted that the PDP is not without its critics. The most controversial aspect is the assumptionthat controlled and automatic processes contribute independently to performance, that is, controlled processes canoccur with or without automatic processes, and vice versa (see for example, Curran & Hintzman, 1995, 1997, for arefutation of the independence assumption; and Jacoby, 1998; Jacoby, Begg, & Toth, 1997; Jacoby & Shrout, 1997, fora response to these criticisms). Alternative hypotheses include the Redundancy and Exclusivity hypotheses (however,see Jacoby, Yonelinas, & Jennings, 1997, for a refutation of these hypotheses). The major argument put forward byJacoby and colleagues in favor of the independence assumption is that several studies have identified variables thatproduce dissociated effects on the estimates of controlled and automatic processes (e.g., Debner & Jacoby, 1994; Hay& Jacoby, 1996; Jacoby, 1991, 1998, 1999; Jacoby et al., 1993; Jennings & Jacoby, 1993, 1997; Kelley & Jacoby, 2000;Yonelinas & Jacoby, 1995). Another criticism of the PDP is that A and C estimated in a memory task could reflectcontrolled or automatic processes not related to memory, for instance inhibitory processes (Nigg, 2001; see however,Hay & Jacoby, 1999, who demonstrated that A and C, as evaluated with the PDP, do reflect memory processes).

Considering that the PDP is a good procedure for obtaining a pure measure of controlled memory processes, someauthors (Jacoby et al., 1996; Jennings & Hay, 1994) have proposed that performances obtained with the PDP constitutea more ecological measure of memory capacities given that daily memory difficulties are related almost exclusivelyto controlled processes. In this context, Jennings and Hay (1994) demonstrated that the memory complaints of olderadults, evaluated with a self-evaluation questionnaire of daily memory and attention difficulties, was strongly correlated(.56) with the estimation of controlled processes calculated with the PDP model, but not with the estimation ofautomatic processes (.08). Moreover, they observed weak correlations (.33) between the self-evaluation questionnaire

and performance on a classic recognition task where the controlled and automatic processes were not dissociated.

The weak relationship between results on memory tests and the degree of memory complaints also raises thepossibility that people’s complaints are influenced by the type of functioning they favor in their daily lives, andmore specifically by routinization. This hypothesis has recently been explored in a study by Bergua et al. (2006),

Page 3

wap

icRbcdstwwoo

(paeo

2

2

fiSPwtS

2

t

TD

AEMMEQCPPP

P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989 981

hich revealed a positive association between general cognitive complaints and preferences for routines. However, thessociation was never analyzed with regard to memory complaints specifically, and accordingly, a PDP model was notart of the hypothesis in the Bergua et al. (2006) study.

Routinization refers to the level of repetition of the same behaviors and to routines that follow a particular order thats difficult to modify (Bouisson, 2002). For example, doing the same things everyday, at the same hours of the day, isonsidered to be routinized behavior. Routinization becomes more and more pronounced with age (Kastenbaum, 1984;eich & Zautra, 1991) and may constitute an adaptive means to protect the individual from the dangers or stress causedy new situations (Bouisson, 2002). Accordingly, preference for routines should increase with growing awareness ofognitive decline (Bergua et al., 2006). Thus, two individuals with the same results on a memory test could presentifferent degrees of daily memory complaints, because of their different lifestyles: the first living a routinized life in atable environment and needing few memory resources, the second living a non-routinized life involving new situationshat require more memory resources. In this way, if an individual has a strongly routinized lifestyle that would explainhy that person presents poor results on memory tests yet has few daily memory complaints. A non-routinized person,ho also performs poorly on memory tests, might complain more about everyday memory difficulties. Thus, whetherr not a subject leads a routinized life may influence the relationship between memory complaints and the resultsbtained on memory tests.

In this context, the current study had two objectives. The first was to replicate the results of Jennings and Hay1994), who demonstrated that the daily memory complaints of elderly people are strongly correlated with controlledrocesses, but not automatic processes, and also that an evaluation model with the PDP is ecologically superior toclassic measure of cued recall where the processes are not dissociated. The second objective consisted of further

xploring the role of routinization as a potential confounding factor that may modulate the relations between resultsn memory tests and the level of daily memory complaints.

. Methods

.1. Participants

A total of 33 participants, aged 55 and older, agreed to take part in the study. All participants spoke French as theirrst language and presented normal cognitive functioning, according to their results on the Mattis Dementia Ratingcale (Mattis, 1976) and on the Mini-Mental State Exam (MMSE; Folstein, Folstein, & McHugh, 1975) (see Table 1).articipants presenting cognitive or psychiatric problems, a history of depression or previous cranial traumas or whoere taking medications that might alter memory functions were excluded. All participants gave their written consent

o participate in the study. The project was accepted by the ethics committee of the Research Center on Aging of theherbrooke Geriatrics University Institute, Sherbrooke (Canada). Table 1 presents the participants’ characteristics.

.2. Materials and procedure

The different tests were administered in about 2.5 h. All participants were given a stem completion task applyinghe Process Dissociation Procedure (PDP: as adapted by Adam et al., 2005). In addition, they all had to answer the

able 1emographic characteristics, questionnaires and PDP results

Group results

ge 71.17 ± 8.5 (55–86)ducation 11.66 ± 4.7MSE 28.94 ± 1.0attis 138.40 ± 4.54

PR (/50) 27.2 ± 4.5AM (/320) 132.8 ± 26.9DS (/195) 69.5 ± 14.9DP—inclusion 0.62 ± 0.14DP—controlled processes 0.42 ± 0.18DP—automatic processes 0.28 ± 0.11

Page 4

982 P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989

Memory Self-evaluation Questionnaire (QAM: Van der Linden, Wyns, Coyette, von Frenckell, & Seron, 1989), theCognitive Difficulties Scale (CDS—French version: Derouesne et al., 1993), and the Routinization Preferences Scale(EPR: Bouisson, 2002).

The PDP, adapted in the form of a trigram completion task (Adam et al., 2005), applies the process dissociationprocedure formulated by Jacoby (1991) and allows for a quantitative assessment of automatic and controlled processes.The task, which has 14 equivalent versions, is administered on a computer using E-Prime software system Version 1.0(Schneider, Eschman, & Zuccolotto, 2002). It includes both an inclusion condition and an exclusion condition, with 56items each. In each condition, each six-letter word is presented in the center of the screen for 3 s. The participant mustread the words out loud and memorize them. At certain points (intervals of 0, 3 or 12 items), the stems, consisting ofthe first three letters of a word the participant has already memorized, are presented for 15 s. In the inclusion condition,the participant must complete the stem aloud from the previously memorized items. If the participant cannot recallthe target word, he or she is asked to recall the first six-letter word that comes to mind that starts with those three firstletters. Conversely, in the exclusion condition, participants must not complete the stem with a word that was memorized.Rather, they must find a new six-letter word that begins with the first three letters. For each condition, participantsare not allowed to complete the stem with a proper noun, a plural, or a conjugated verb (except for past participles).If participants generate a response that does not meet these criteria, they are informed of the error and encouragedto formulate a new response. As well, for 3 s before each presentation of a stem, the word “old” in blue (inclusioncondition) or “new” in red (exclusion condition) is presented on the screen. During this period, the experimenter brieflyrepeats the instructions, by saying, in the inclusion condition, “What is the six-letter word that you have previouslyseen that begins with these first three letters?” If the participant does not find the word, the experimenter says, “If youdon’t remember, try to give the first six-letter word that comes to mind.” In the exclusion condition, the experimentersays, “You must not give a word you have seen before, but find a new six-letter word that begins with these first threeletters.” To ensure the validity of the procedure, two methodological controls were applied. First, for each condition,16 control items, where the stem cannot be associated with a target word, are presented to evaluate the rate of randomcompletion. This rate of completion serves (a) to ensure that the participants use the same response criteria in both theinclusion and exclusion conditions, by comparing the base level for each condition and (b) to verify that the automaticprocesses correspond to the unconscious influence of memory by showing that the estimates of automatic processesare significantly over baseline levels (chance). The second methodological control consisted of presenting some stemswithout any recall of instructions (pre-presentation lag 0) to ensure that the participants have understood the instructions(while pre-presentation lags 3 and 12 measure memory processes; for more details, see Adam et al., 2005).

The QAM (Van der Linden et al., 1989) measures the memory problems participants encounter in their daily lives. It

comprises 62 questions categorized according to ten types of forgetting (or rubrics: conversations, people, news events,forgotten facts of one’s personal life, distractions, objects, books and films, use of objects, etc., see also Table 2), suchas “Do you forget the content of a conversation you just had?”, “Do you forget how to use certain objects?”, “Do

Table 2Results of QAM principal component analyses

QAM questions Varimax rotated component loadings

Component 1 Component 2 Component 3

Do you think you have memory problems 1 .317 .827 −.101Conversations .841 .288 .018Films and books .502 .521 .449Distractions .367 .565 .570Persons .538 .538 .397Instructions −.074 −.046 .929News events .746 .155 .319Places .585 .293 .599Things to do .379 .340 .613Forget personal facts .909 .052 .035Do you think you have memory problems 2 −.054 .864 .130General questions .274 .618 .320

Bold are results showing the inclusion of a question into one component.

Page 5

yNrhrac

odfoLgpmca

espsoa(tS

3

ma

3

mcmtrQttt

f

P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989 983

ou forget appointments?” The participant must respond to these statements on a 6-point scale, from never to always.orms and their analysis are available according to age (18–70), cultural and social level (1–3) and gender. Test–retest

eliability (in between 1 and 2 months), estimated for a group of 25 cognitively normal subjects (aged 18–64 years),as been demonstrated for each rubric (mean r = .80). The correspondence between responses from 20 subjects andesponses from their spouses was examined. Correlations were obtained for three rubrics: Conversations, Distractionsnd Forget personal facts (mean r = .55). Moreover, this measure is widely used in neuropsychology because of itslinical utility.

The CDS (McNair & Kahn, 1984, French version developed by Derouesne et al., 1993) is a self-report questionnairef perceived cognitive difficulties as manifested in activities of daily living. There are 39 statements relating to cognitiveisorders such as concentration, attention, memory, expression, praxis and others. For example, the scale includes theollowing statements: “I forget to call back if someone phoned me,” “I have difficulty staying focused on a task or anccupation,” and “I forget ingredients when I cook.” The participant must respond to these statements based on a 5-pointikert scale ranging from never to very often. The total scale score ranges from 0 to 156 (higher scores indicatingreater perceived cognitive difficulties). Derouesne et al. (1993) administered the CDS to 1628 cognitively normalarticipants aged 45–75 years. They found that scores on the scale correlated only weakly with scores on objectiveeasures of memory and attention. Moreover, the time frame for assessment of perceived cognitive difficulties was

onsistent with that for depression and anxiety. Cronbach’s internal consistency for the current sample was estimatedt α = .94 and .96, respectively for the first and second evaluations.

The EPR (Bouisson, 2002) measures the routinized aspect of one’s daily life with the aid of 10 statements thatvaluate the desire to change daily habits and the preference for doing tasks in a specific order. The EPR containstatements that refer to a routinized lifestyle such as “I can’t tolerate it when my things are moved around” and “Irefer to get up and go to bed at the same time everyday.” It also contains statements that refer to non-routinized lifestyleuch as “I like unexpected situations” and “I really like to move and change activities.” The participant must respondn a 5-point scale, from not at all true to very true. The total score is calculated in terms of a preference for routinend ranges from 10 to 50. Bouisson (2002) showed that the ERP has (1) acceptable internal consistency (α = .73) and2) high test–retest reliability (r = .84; reliability being tested over a 2-week period). The ERP is also short enougho use with elderly participants and is valid as a measure of actual behavioral routinization in daily life (Bouisson &wendsen, 2003).

. Results

Group results are presented in Table 1. Concerning the PDP task, the inclusion score (i.e., the memory-contaminatedeasure) and estimates of controlled and automatic memory processes were collapsed across lags 3 and 12 to providesingle, more sensitive, estimate of these processes.1

.1. Associations between memory complaints and measures of memory

The first objective was to verify whether a pure measure of controlled processes for retrieval of information fromemory would allow a better understanding of complaints made by older adults than a more classic measure (a

ontaminated measure). To do so, simple regression analyses were performed between the inclusion score (contaminatedeasure) and the total scores for the QAM and CDS, and thus with the complaints reported by the group. A significant

rend was observed between the inclusion score and the QAM (r = .33; t = −1.9, p = .062). The inclusion score was notelated to the CDS (r = .20; t = −1.2, p = .25). For controlled processes, a significant relationship was observed with the

AM (r = .43; t = −2.7, p = .012). Scores obtained for the controlled processes and the CDS were not related, although

hey showed a significant trend (r = .29; t = −1.7, p = .097). Finally, the relationship between automatic processes andhe level of complaints was also analyzed. The scores from automatic processes were not related to scores obtained onhe QAM or the CDS (rs < .05; ps > .80).

1 Performance at Lag 0 was not considered as the memory load is very weak and this lag served as a control to verify if subjects were correctlyollowing their instructions.

Page 6

984 P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989

To see whether certain parts of the QAM might be better related to memory measures than to the total score,a principal component factorial analysis was performed. Using a varimax rotation, three components emerged witheigenvalues greater than 1, explaining 74.6% of the variance. The values of the three components were 6.2, 1.4 and 1.3,respectively (see Table 2). The first component contained the questions for conversations, persons, news, and personalevents. These questions contain mainly autobiographical elements (episodic and semantic) related to memories of eventsand information, as well as a small component referring to the inhibition of automatic processes during conversations(i.e., repeating the same thing several times). Together, these questions seem to refer generally to controlled retrieval ofmemories. The second component contains the two identical questions “Do you think you have memory problems?”,questions about forgetting films and books, and the general questions. All these questions share common elementsrelated to working memory (following the line of a film, doing two things at the same time, learning while under stress,etc.). The last component contains the questions about distractions, instructions, places and things to be done. The firstgroup of questions (distractions) concern attentional elements in relation to routinized behavior, while the other threerequire an array of cognitive functions and so are related to complex activities.2

The three components are significantly correlated with scores for controlled processes, especially the third compo-nent referring to complex activities (r = −.34, p = .052; r = −4.0, p = .03; r = −.49, p = .004, respectively). In all cases,low scores for controlled processes indicate more complaints. No component was related to scores for inclusion, thoughsome significant trends were found for components 2 and 3 (component 1, p = .13; component 2, p = .08; component3, p = .07).

3.2. The influence of routine on the relation between memory complaints and measures of memory

The second objective of this study was to evaluate whether living a routinized lifestyle would influence memorycomplaints and thus the relationship between memory complaints and scores obtained on memory tests. Routinizationwas therefore considered here as a confounding variable. The hypothesis is that two individuals with the same scoreson a memory test could have different levels of complaints due to their different lifestyles (a routinized style, whichrepresents a low memory load, or a non-routinized style, which represents a greater memory load).

In the first stage, simple regression analyses were carried out between total scores for the EPR (routine) and the totalscores for the QAM (complaints). The total score on the routinization scale was not related to memory complaints, asmeasured by the QAM (r = .21; t = −1.2, p = .23). The total score on the EPR was not related to the three componentsof the QAM either (ps > .10).

Because the EPR contains questions about both routine behaviors and non-routine behaviors, and because thetotal score is calculated as a function of preference for routine, a principal component factorial analysis was doneon all of the questions to see whether the instrument could be split into two components: one routine component,composed of questions pertaining only to routine behaviors, and one non-routine component. The hypothesis wasthat the non-routine component should be more related to complaints than the routine component. Using a vari-max rotation, two components emerged with eigenvalues greater than 1, explaining 55.3% of the variance. Thevalues of the two components were 2.4 and 1.5, respectively (see Table 3). The items grouped under component1 related to routinized behaviors. Component 2 had all the questions about non-routinized behavior. Two questionscould not be classified and dealt with routine behaviors: “In general, I like to do the same things every day” and“I prefer to get up and go to bed at the same time every day.” These two questions were excluded from furtheranalyses.

Regression analyses between the EPR and memory complaints were therefore redone using the total scores for thesetwo components, routine and non-routine. The routine component was not related to memory complaints, as measuredby the QAM (r = .07; t = −.37, p = .72). On the other hand, the non-routine component was significantly related to

total scores on the QAM (r = .37; t = −2.2, p = .035); the less routinized a lifestyle a person leads, the more memorycomplaints he or she has. More specifically, the non-routine component was associated with component 2 of the QAM(working memory) (r = .37; t = −2.2, p = .037), but not with component 1 (controlled retrieval of memories) or withcomponent 3 (complex activities), although there was a significant trend (ps > .08).

2 It is worth mentioning that the association of certain questions with particular cognitive constructs is an a posteriori assumption, based on thegrouping of questions by component factorial analysis. It seems clinically valid but other associations may be plausible.

Page 7

P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989 985

Table 3Results of EPR principal component analyses

EPR questions* Varimax rotated component loadings

Component 1 (routinized behaviors) Component 2 (non-routinized behaviors)

I don’t like people who are late .611 −.326I can’t stand my things being moved around .631 −.185I like unexpected situations −.413 .717I don’t like waiting at mealtimes .686 .083I like to move and change activities .229 .696I don’t like it when someone takes my seat .781 .100I like to meet new people −.112 .792

Bold are results showing the inclusion of a question into one component.* Cited with permission of the author.

Table 4Results for the relation between complaints from the QAM and the controlled processes score of the PDP (controlled for the non-routinizedcomponent or not)

Variables Beta t p value

Total QAM—controlled processes −.430 −2.654 .012Total QAM—controlled processes/non-routinized component −.353 −2.133 .041QAM component 2 (working memory)—controlled processes −.395 −2.397 .023Q

L

ticwccfcr

4

satrceuCsctat

AM component 2 (working memory)—controlled processes/non-routinized component −.316 −1.878 .070

egend: / = controlled for.

In the second stage, the non-routine component was introduced into the relation between memory complaints andhe scores for controlled processes, to see if it plays a modulator role. The measure retained for memory complaintss the one evaluated by the total score on the QAM, which is related to the non-routine component and also toontrolled processes. Simple regression analyses, introducing the non-routine component as a confounding variable,ere performed (see Table 4). The results indicate that the non-routine component influences the relationship between

omplaints (total on QAM) and controlled processes by 8% (Beta changes from −.43 to −.35). Because the non-routineomponent was strongly related to the second component of the QAM (working memory), the same analysis was doneor these two variables. The results indicated that the non-routine component also influenced the relation betweenomplaints (QAM component 2) and controlled processes by 8% (Beta changed from −.395 to −.316); however, thiselation was not significant (p = .07).

. Discussion

The first objective of this study was to replicate the results obtained by Jennings and Hay (1994), who demon-trated that everyday memory complaints from older adults are correlated with controlled processes but not withutomatic processes. They also showed that a method of memory evaluation with the PDP allows one to iden-ify an isolated measure of controlled processes that is better correlated with memory complaints than is a classicecall measure. The results obtained allowed us to partially replicate Jennings and Hay’s (1994) results. First, theontrolled processes evaluated with the PDP were shown to be significantly correlated with memory complaintsvaluated with the QAM questionnaire, but not with the CDS questionnaire. Next, the automatic processes eval-ated with the PDP were not significantly correlated with memory complaints, either on the QAM or on theDS. Finally, the results of the inclusion task of the PDP, which replicated a classic cued recall test, were not

ignificantly correlated with memory complaints. These results suggest that, by obtaining an isolated measure of

ontrolled processes, the PDP allows an evaluation that better predicts daily memory difficulties than a “tradi-ional” memory task (that is, a task contaminated by automatic processes). The results obtained are therefore inccord with the position of several authors (Adam, 2003; Jacoby et al., 1996; Jennings & Hay, 1994) who proposehat a purer measure of controlled processes of information retrieval from memory constitutes a more ecological

Page 8

986 P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989

and valid approach, and is therefore better adapted to evaluations of memory difficulties encountered in everydaylife.

The second objective was to verify whether routinization could modulate the relationship between results on memorytests and the degree of daily memory complaints, as a routine lifestyle may represent an adjustment strategy to cope withmemory loss. Bergua et al. (2006) found a positive association between general cognitive complaints and preferences forroutines. The results of our study more specifically demonstrated that routine has an impact on the relationship betweenmemory test results and memory complaints. The relation between controlled processes and complaints (QAM) wasattenuated when the non-routine component was introduced into the model. The routinization factor, especially beingnon-routinized, acts as a confounding variable. Thus, it can be suggested that, of two people whose memory capacitiesare similar, the one who has a lifestyle that is more non-routinized risks manifesting more memory complaints thanthe one whose lifestyle is routinized.

Therefore, our results emphasize the importance of anamnesis in neuropsychology and suggest that lifestyle is a non-negligible factor that should be considered in future studies. However, the important and thorny problem of the “causalrelationship” has still to be considered. Indeed, the question is whether a person becomes more routinized because hisor her cognitive problems are increasing, or whether the fact that that the person has a routinized lifestyle influencesthe development of the cognitive problems. Probably, there is a reciprocal relation between cognitive functioning androutinization.

In the context of the effect of cognitive functioning on lifestyle, the results obtained in normal aging suggest thatindividuals presenting a heightened level of memory complaints, who consider their problems to be disabling, couldbe relieved in part by living in an environment that solicits less controlled processes, and in part by installing aspectsof routine in their lives. More specifically, it is possible that individuals who live in environments with more memorysupports (e.g., using written supports such as diaries, memos and written instructions) would experience less demandfor controlled processes. Studies that have explored using memory supports with environmental cues by comparingthe performance of younger individuals with that of older individuals have produced mixed results. Thus, Craik andByrd (1982) observed that older adults benefited more from a cued environment than the younger ones, while Light(1991) and Craik and Jennings (1992) obtained different results in similar studies. Jacoby et al. (1996) hypothesizedthat these studies did not take controlled processes or automatic processes into account in the form of environmentalsupport offered and that this lack was at the origin of the divergent results observed. Another alternative that wouldallow for the reduction of memory complaints would be to install aspects of routine into everyday life. Thus, developingfixed sequences of actions as daily routines (e.g., get up, get dressed, eat, brush teeth, etc.), or even putting in placedaily activities at fixed times (i.e., each leisure activity takes place on the same day each week), could possibly reducememory complaints.

The association between routinization and cognitive functions probably also involves executive functioning. Indeed,executive functions refer to various processes (inhibition, planning, shifting, updating, control, etc.) whose main purposeis to facilitate a subject’s adaptation to novel situations, especially when action routines are not sufficient (Miyake,Friedman, Emerson, Witzki, & Howerter, 2000; Shallice, 1982). In this context, it seems evident that reduced executivefunctioning (as observed in normal aging) might promote a routinized lifestyle. Further research will be necessary toaddress this issue.

However, even though the installation of routines and environmental memory supports offers the possibility ofreducing memory complaints, some data from the cognitive literature suggest that using these aids could also havenegative consequences on cognitive functioning. Indeed, several studies have demonstrated that active involvement inphysically, intellectually and socially rich and cognitively stimulating activities (and thus potentially non-routinizedactivities) is a key factor allowing some people to prevent the cognitive decline associated with normal aging orAlzheimer’s disease (Adam, Bay, Bonsang, Germain, and Perelman, submitted; Lindsay et al., 2002; Scarmeas &Stern, 2003; Wang, Karp, Winblad, & Fratiglioni, 2002; Wilson et al., 2002).

This view refers to the concept of “cognitive reserve” (proposed by Stern, 2002, 2003; Scarmeas & Stern, 2003),which suggests that innate intelligence or aspects of life experience such as educational or occupational attainmentsconstitute a reserve, represented by a set of skills or repertoires that allows some people to prevent the cognitive decline

associated with normal aging or Alzheimer’s disease. Recent studies have tried to identify parameters contributing tothe development of a cognitive reserve such as (1) education (e.g., Le Carret et al., 2003), (2) occupation (Schooler,Mulatu, & Oates, 1999), (3) professional or leisure activities (Capurso et al., 2000; Scarmeas, Levy, Tang, Manly, &Stern, 2001; Wilson et al., 2002), (4) rich social environment (Berkman, Glass, Brissette, & Seeman, 2000), and (5)

Page 9

am

ob2eocts

r&tbtcpf

A

b

R

A

A

A

B

B

BB

BC

C

C

C

C

D

D

P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989 987

ctive lifestyle (for a review, see Fillit et al., 2002; Fratiglioni, Paillard-Borg, & Winblad, 2004). All these variablesay have an opposite relationship with routinization.As well, other studies done with older adults have demonstrated that routinization can limit the stimulating potential

f new situations (Langer, 1981, cited in Jacoby et al., 1996) and even increase levels of anxiety and depression, causedy a reduced sense of control in the face of unexpected situations and lower self-esteem caused by routine (Bouisson,002). Thus, a non-routinized person who has memory complaints, but does not consider them problematic, should bencouraged to keep the non-routinized aspects of their daily life. And one avenue to reduce the memory complaintsf a person who considers them to be disabling, could take the form of a compromise between routinization andognitive stimulation. For example, implementing a fixed schedule that includes cognitively rich activities would allowhe introduction of a routine that diminishes the daily memory load, while at the same time providing beneficial newtimulation.

In conclusion, the results obtained in this study clarify the questions of authors who have highlighted the weakelations between the results on traditional memory tests and actual memory complaints (Jacoby et al., 1996; Reid

Maclullich, 2006). The results support the remarks of authors who claim that the PDP is ecologically superior toraditional memory tests (Adam, 2003; Jacoby et al., 1996). They also indicate that routinization, more specificallyeing non-routinized, is a variable that seems to play an important role in older adults’ memory complaints. Finally,he results obtained suggest that dissociating controlled and automatic processes and taking lifestyle into account,ould significantly contribute to the neuropsychological assessment of older adults with memory complaints. From theerspective of cognitive management for persons with memory difficulties, future studies should explore the potentialor a combination of routinization and a sufficient level of cognitive stimulation.

cknowledgments

We are grateful to all participants and to Annouk Berube for her help in testing them. The research was supportedy a fellowship from the Fonds de la Recherche en Sante du Quebec to the last author.

eferences

dam, S. (2003). Nouvelles techniques d’evaluation de la memoire: Procedure de dissociation des processus et paradigme R/K. In T. Meulemans,B. Desgranges, S. Adam, & F. Eustache (Eds.), Evaluation et prise en charge des troubles mnesiques (pp. 141–167). Marseille: Solal.

dam, S., Bay, C., Bonsang, E., Germain, S., & Perelman, S. (submitted). Occupational activities and cognitive reserve: A frontier approach appliedto the survey on health, ageing, and retirement in Europe (SHARE).

dam, S., Van der Linden, M., Collette, F., Lemauvais, L., & Salmon, E. (2005). Further exploration of controlled and automatic memory processesin early Alzheimer’s disease. Neuropsychology, 19, 420–427.

ergua, V., Fabrigoule, C., Barberger-Gateau, P., Dartigues, J.-F., Swendsen, J., & Bouisson, J. (2006). Preferences for routines in older people:Associations with cognitive and psychological vulnerability. International Journal of Geriatric Psychiatry, 21, 990–998.

erkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to health: Durkheim in the new millennium. Social Science& Medicine, 51, 843–857.

ouisson, J. (2002). Routinization preferences, anxiety, and depression in an elderly French sample. Journal of Aging Studies, 16, 295–302.ouisson, J., & Swendsen, J. (2003). Routinisation and emotional well-being: An experience sampling investigation in an elderly French sample.

Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 58, 280–282.utler, L. T., & Berry, D. C. (2001). Implicit memory: Intention and awareness revisited. Trends in Cognitive Sciences, 5, 192–197.apurso, A., Panza, F., Solfrizzi, V., Capurso, C., Mastroianni, F., & Del Parigi, A. (2000). Age-related cognitive decline: Evaluation and prevention

strategy. Recenti Progressi in Medicina, 91, 127–134.raik, F. I. M., & Byrd, M. (1982). Aging and cognitive deficits. The role of attentional resources. In F. I. M. Craik & S. Trehub (Eds.), Aging and

cognitive processes (pp. 191–211). New York: Plenum Press.raik, F. I. M., & Jennings, J. M. (1992). Human memory. In F. I. M. Craik & T. A. Salthouse (Eds.), The handbook of aging and cognition (pp.

51–110). Hillsdale, NJ: Lawrence Erlbaum Associates.urran, T., & Hintzman, D. G. (1995). Violations of the independence assumption in process dissociation. Journal of Experimental Psychology:

Learning, Memory and Cognition, 21, 531–547.urran, T., & Hintzman, D. G. (1997). Consequences and causes of correlations in process dissociation. Journal of Experimental Psychology:

Learning, Memory and Cognition, 23, 496–504.

ebner, J. A., & Jacoby, L. L. (1994). Unconscious perception: Attention, awareness, and control. Journal of Experimental Psychology: Learning,

Memory, and Cognition, 20, 304–317.erouesne, C., Dealberto, M. J., Boyer, P., Lubin, S., Sauron, B., Piette, F., et al. (1993). Empirical-evaluation of the cognitive difficulties scale for

assessment of memory complaints in general-practice: A study of 1628 cognitively normal subjects aged 45–75 years. International Journal ofGeriatric Psychiatry, 8, 599–607.

Page 10

988 P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989

Fay, S., Isingrini, M., & Pouthas, V. (2005). Does priming with awareness reflect explicit contamination? An approach with a response-time measurein word-stem completion. Consciousness and Cognition, 14, 459–473.

Fillit, H. M., Butler, R. N., O’Connell, A. W., Albert, M. S., Birren, J. E., Cotman, C. W., et al. (2002). Achieving and maintaining cognitive vitalitywith aging. Mayo Clinic Proceedings, 77, 681–696.

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-mental state: A practical method for grading the cognitive state of patients for theclinician. Journal of Psychiatric Research, 12, 189–198.

Fratiglioni, L., Paillard-Borg, S., & Winblad, B. (2004). An active and socially integrated lifestyle in late life might protect against dementia. TheLancet, 3, 343–353.

Hay, J. F., & Jacoby, L. L. (1996). Separating habit and recollection: Memory slips, process dissociations, and probability matching. Journal ofExperimental Psychology: Learning, Memory, and Cognition, 22, 1323–1335.

Hay, J. F., & Jacoby, L. L. (1999). Separating habit and recollection in young and older adults: Effects of elaborative processing and distinctiveness.Psychology and Aging, 14, 122–134.

Holender, D. (1986). Semantic activation without conscious identification in dichotic listening parafoveal vision, and visual masking: A survey andappraisal. Behavioral and Brain Sciences, 9, 1–23.

Jacoby, L. L. (1991). A process dissociation framework: Separating automatic from intentional uses of memory. Journal of Memory and Language,30, 39–55.

Jacoby, L. L. (1998). Invariance in automatic influences of memory: Toward a user’s guide for the process dissociation procedure. Journal ofExperimental Psychology: Learning, Memory and Cognition, 24, 3–26.

Jacoby, L. L. (1999). Ironic effects of repetition: Measuring age-related differences in memory. Journal of Experimental Psychology: Learning,Memory, and Cognition, 25, 3–22.

Jacoby, L. L., Begg, I. M., & Toth, J. P. (1997). In defence of functional independence: Violations of assumptions underlying the process dissociationprocedure? Journal of Experimental Psychology: Learning, Memory and Cognition, 23, 484–495.

Jacoby, L. L., Jennings, J. M., & Hay, L. F. (1996). Dissociating automatic and consciously controlled processes: Implications for diagnosis andrehabilitation of memory deficits. In D. J. Herrmann, C. L. McEvoy, C. Hertzog, P. Hertel, & M. K. Johnson (Eds.), Basic and applied memoryresearch: Theory in context (pp. 161–193). Mahwah, NJ: Lawrence Erlbaum Associates.

Jacoby, L. L., & Shrout, P. E. (1997). Toward a psychometric analysis of violations of the independence assumption in process dissociation. Journalof Experimental Psychology: Learning, Memory and Cognition, 23, 505–510.

Jacoby, L. L., Toth, J. P., & Yonelinas, A. P. (1993). Separating conscious and unconscious influences of memory: Measuring recollection. Journalof Experimental Psychology: General, 122, 139–154.

Jacoby, L. L., Yonelinas, A. P., & Jennings, J. M. (1997). The relation between conscious and unconscious (automatic) influence: A declarationof independence. In J. D. Cohen & J. W. Schooler (Eds.), Scientific approaches to the question of consciousness (pp. 13–47). Hillsdale, NJ:Lawrence Erlbaum Associates.

Jennings, J. M., & Hay, J. F. (1994). Predicting everyday memory failure in older adults. In Poster presented at the 1994 Canadian aging researchnetwork (CARNET) conference.

Jennings, J. M., & Jacoby, L. L. (1993). Automatic versus intentional uses of memory: Aging, attention, and control. Psychology and Aging, 8,283–293.

Jennings, J. M., & Jacoby, L. L. (1997). An opposition procedure for detecting age-related deficits in recollection: Telling effects of repetition.Psychology and Aging, 12, 352–361.

Kastenbaum, R. J. (1984). When aging begins. Research on Aging, 6, 105–117.Kelley, C. M., & Jacoby, L. L. (2000). Recollection and familiarity: Process dissociation. In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook

of memory (pp. 215–228). London: Oxford University Press.Le Carret, N., Lafont, S., Letenneur, L., Dartigues, F., Mayo, W., & Fabrigoule, C. (2003). The effect of education on cognitive performances and

its implication for the constitution of the cognitive reserve. Developmental Neuropsychology, 23, 317–337.Light, L. L. (1991). Memory and aging: Four hypotheses in search of data. Annual Review of Psychology, 43, 333–376.Lindsay, J., Laurin, D., Verreault, R., Hebert, R., Helliwell, B., Hill, G. B., et al. (2002). Risk factors for Alzheimer’s disease: A prospective analysis

from the Canadian Study of Health and Aging. American Journal of Epidemiology, 156, 445–453.Mattis, S. (1976). Mental status examination for organic mental syndrome in the elderly patient. In L. Bellak & T. B. Karasu (Eds.), Geriatric

psychiatry (pp. 77–121). New York: Grune & Stratton.McNair, D. M., & Kahn, R. J. (1984). Self-assessment of cognitive deficits. In T. Crook, S. Ferris, & R. Bartus (Eds.), Assessment in geriatric

psychopharmacology (pp. 137–143). New Canaan, CT: Mark Powley.Mitchell, D. B., & Bruss, P. J. (2003). Age differences in implicit memory: Conceptual, perceptual, or methodological? Psychology and Aging, 18,

807–822.Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., & Howerter, A. (2000). The unity and diversity of executive functions and their

contributions to complex “frontal lobe” tasks: A latent variable analysis. Cognitive Psychology, 41, 49–100.Nigg, J. T. (2001). Is ADHD a disinhibitory disorder? Psychological Bulletin, 127, 571–598.Pearman, A., & Storandt, M. (2004). Predictors of subjective memory in older adults. Journals of Gerontology Series B: Psychological Sciences

and Social Sciences, 59, 4–6.

Reich, J. W., & Zautra, A. J. (1991). Analysing the trait of routinization in older adults. International Journal of Aging and Human Development,

32, 161–180.Reid, L. M., & Maclullich, A. M. (2006). Subjective memory complaints and cognitive impairment in older people. Dementia and Geriatric Cognitive

Disorders, 22, 185–471.

Page 11

R

R

S

SSS

SS

S

ST

V

V

W

W

Y

P. Dubreuil et al. / Archives of Clinical Neuropsychology 22 (2007) 979–989 989

eingold, E. M., & Merikle, P. M. (1990). On the inter-relatedness of theory and measurement in the study of unconscious processes. Mind andLanguage, 5, 9–28.

ichardson-Klavehn, A., & Bjork, R. A. (1988). Primary versus secondary rehearsal in an imagined voice: Differential effects on recognitionmemory and perceptual identification. Bulletin of the Psychonomic Society, 26, 187–190.

carmeas, N., Levy, G., Tang, M. X., Manly, J., & Stern, Y. (2001). Influence of leisure activity on the incidence of Alzheimer’s disease. Neurology,57, 2236–2242.

carmeas, N., & Stern, Y. (2003). Cognitive reserve and lifestyle. Journal of Clinical and Experimental Neuropsychology, 25, 625–633.chneider, W., Eschman, A., & Zuccolotto, A. (2002). E-Prime user’s guide. Pittsburgh, PA: Psychology Software Tools Inc.chooler, C., Mulatu, M. S., & Oates, G. (1999). The continuing effects of substantively complex work on the intellectual functioning of older

workers. Psychology and Aging, 14, 483–506.hallice, T. (1982). Specific impairments of planning. Philosophical Transactions of the Royal Society of London B, 298, 199–209.te-Marie, D. M., Jennings, J. M., & Finlayson, A. J. (1996). Process dissociation procedure: Memory testing in populations with brain damage.

The Clinical Neuropsychologist, 10, 25–36.tern, Y. (2002). What is cognitive reserve? Theory and research application of the reserve concept. Journal of the International Neuropsychological

Society, 8, 448–460.tern, Y. (2003). The concept of cognitive reserve: A catalyst for research. Journal of Clinical and Experimental Neuropsychology, 25, 589–593.oth, J. P., Reingold, E. M., & Jacoby, L. L. (1994). Toward a redefinition of implicit memory: Process dissociations following elaborative processing

and self-generation. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20, 290–303.an der Linden, M., Wyns, C., Coyette, F., von Frenckell, R., & Seron, X. (1989). Le Q.A.M. Questionnaire d’auto-evaluation de la memoire.

Brussels: Editions EDITEST.oss, J. L., & Paller, K. A. (2007). Neural correlates of conceptual implicit memory and their contamination of putative neural correlates of explicit

memory. Learning and Memory, 14, 259–267.ang, H.-X., Karp, A., Winblad, B., & Fratiglioni, L. (2002). Late-life engagement in social and leisure activities is associated with a decreased

risk of dementia: A longitudinal study from the Kungsholmen project. American Journal of Epidemiology, 155, 1081–1087.ilson, R. S., Mendes de Leon, C. F., Barnes, L. L., Schneider, J. A., Bienias, J. L., Evans, D. A., et al. (2002). Participation in cognitively stimulating

activities and risk of incident Alzheimer disease. The Journal of the American Medical Association, 287, 742–748.onelinas, A. P., & Jacoby, L. L. (1995). Dissociating automatic and controlled processes in a memory-search task: Beyond implicit memory.

Psychological Research, 57, 156–165.