Differential Involvement of Knowledge Representation ...

Canadian Journal of Experimental Psychology / Revue canadienne de psychologie expérimentale 2013, Vol. 67, No. 2, 100 –107

© 2012 Canadian Psychological Association 1196-1961/13/$12.00 DOI: 10.1037/a0028517

Differential Involvement of Knowledge Representation and Executive Control in Episodic Memory Performance in Young and Older Adults Badiaˆa Bouazzaoui, Se´verine Fay, Laurence Taconnat, Lucie Angel, Sandrine Vanneste, and Michel Isingrini Universite´ Franc¸ois–Rabelais de Tours Craik and Bialystok (2006, 2008) postulated that examining the evolution of knowledge representation and control processes across the life span could help in understanding age-related cognitive changes. The present study explored the hypothesis that knowledge representation and control processes are differentially involved in the episodic memory performance of young and older adults. Young and older adults were administered a cued-recall task and tests of crystallized knowledge and executive functioning to measure representation and control processes, respectively. Results replicate the classic finding that executive and cued-recall performance decline with age, but crystallized-knowledge performance does not. Factor analysis confirmed the independence of representation and control. Correlation analyses showed that the memory performance of younger adults was correlated with representation but not with control measures, whereas the memory performance of older adults was correlated with both representation and control measures. Regression analyses indicated that the control factor was the main predictor of episodic-memory performance for older adults, with the representation factor adding an independent contribution, but the representation factor was the sole predictor for young adults. This finding supports the view that factors sustaining episodic memory vary from young adulthood to old age; representation was shown to be important throughout adulthood, and control was also important for older adults. The results also indicated that control and representation modulate age-group-related variance in episodic memory. Keywords: episodic memory, aging, knowledge representation, executive control

of crystallized schemas that are the basis for memory and knowledge of the world. Control is the set of fluid operations that enable executive adaptive processes. According to Craik and Bialystok (2006), representations are crystallized schemas and can therefore be measured with classic tests of knowledge ability, such as the vocabulary and information subtests of the Wechsler Adult Intelligence Scale (WAIS; Wechsler, 1981). By contrast, control involves executive processes and can thus be measured using standard executive-function tasks, whose main goal is to overcome the prepotent “default mode” of automatic behaviour. These two factors are also assumed to have different life-span trajectories. Representational knowledge increases during childhood, continues to accumulate at a slower pace throughout adulthood, and remains relatively stable in old age. This is the pattern classically depicted for crystallized abilities (Cattell, 1971; Horn, 1982). By contrast, cognitive control increases in power, speed, and complexity from infancy to young adulthood, and declines thereafter, as shown for fluid abilities (Cattell, 1971; Horn, 1982). Craik and Bialystok (2006) suggest that the relative growth and decline of these two factors may help explain the role played by these two components in sustaining cognitive abilities (e.g., intelligence, language, and memory) at different periods of life. This may also help identify factors explaining age-related changes in cognition. Based on Craik and Bialystok’s (2006, 2008) framework, the aim of the present study was to examine whether representation (crystallized knowledge) and control (executive components) play different roles in the episodic-memory functioning of young and older adults.

Research has shown that episodic memory involves both agerelated losses and compensation mechanisms. Episodic memory is defined as the kind of memory that enables conscious recollection of personal happenings and events from the past (Wheeler, Stuss, & Tulving, 1997). It is classically assessed by free-recall, cuedrecall, and recognition tasks. In a seminal article, Craik and Bialystok (2006) proposed a framework comprising two components, cognitive representation and control, to account for these cognitive changes across the life span, and particularly during adulthood. This framework builds on Cattell’s model (1971), distinguishing between crystallized and fluid processes by specifying key factors explaining not only cognitive performance at a given period of life, but also cognitive changes across the life span. Representations are defined as the set

This article was published Online First July 9, 2012. Badiaˆa Bouazzaoui, Se´verine Fay, Laurence Taconnat, Lucie Angel, Sandrine Vanneste, and Michel Isingrini, Centre de Recherches sur la Cognition et l’Apprentissage (CeRCA), Unite´ Mixte de Recherche, Centre National de la Recherche Scientifique (UMR-CNRS) 7295, Universite´ Franc¸ois–Rabelais de Tours, Tours, France. This research was supported by a grant from the French Agence National de la Recherche (ANR): ANR-2010-BLAN-1912– 01. The authors are grateful to F.I.M. Craik and the other anonymous reviewers for their help and comments, and to Elisabeth Yates for her English review. Correspondence concerning this article should be addressed to Badiaˆa Bouazzaoui, Universite´ Franc¸ois–Rabelais de Tours, UMR 7295 CeRCA, 3, rue des Tanneurs, 37000 Tours, France. E-mail: bouazzaoui@ univ-tours.fr 100

REPRESENTATION AND CONTROL IN MEMORY AGING

It is widely accepted that representations are helpful for the functioning of episodic memory, as they provide schematic support and allow learning material to be processed at a deep level (Craik & Bialystok, 2006, 2008; Bransford, Franks, Morris, & Stein, 1979; Craik & Lockart, 1972; Craik & Tulving, 1975). As observed by Newcombe, Lloyd, and Balcomb (2011), the semantic-memory system—which provides a general understanding of the world—may give structure and meaning to the episodicmemory system. Greater representation abilities would thus enhance memory through richer and more elaborate encoding and retrieval cues, thanks to an organisational structure and the ability to chunk information in terms of familiar groupings (Salthouse, 2002). Some studies have confirmed a positive link between knowledge and memory measures (Hedden, Lautenschlager, & Park, 2005; Meinz & Salthouse, 1998; see Salthouse, 2002). On the other hand, studies have shown that memory performance requires processes pertaining to executive functioning. Executive functions are classically defined as a set of cognitive processes encompassing a wide variety of control abilities, involved particularly when dealing with novel problems (Elliott, 2003). These mechanisms can be viewed as supporting the generation of memory-strategy processes (Bouazzaoui et al., 2010; Bryan, Luszcz, & Pointer, 1999; Taconnat et al., 2006; Taconnat, Clarys, Vanneste, Bouazzaoui, & Isingrini, 2007; Taconnat et al., 2009). This view is supported by studies on patients with frontal lobe lesions (Wheeler, Stuss, & Tulving, 1995) and neuroimaging studies which have shown that the prefrontal cortex (PFC), classically associated with executive functioning, sustains episodic memory functioning (Nyberg, Cabeza, & Tulving, 1996). The idea put forward by several authors (Moscovitch & Winocur, 1992; Shimamura, 1995) is that the PFC is not critically involved in the storage of new information but is used to manipulate and organise information to be learned or retrieved. It thus acts as an executive supervisor, viewed as a set of potential support processes aiding episodic-memory performance (Baddeley, 1996; Shallice & Burgess, 1993; Moscovitch & Winocur, 1992). Aging studies have consistently reported a decline in episodicmemory abilities with increasing age (for a review, see McDaniel, Einstein, & Jacoby, 2008). Some studies have attempted to show that such age-related differences can be modulated either by individual levels of representation and/or control, with smaller agerelated decline for high-knowledge and/or high-control individuals. The results regarding representation are not conclusive, as overall, they do not clearly demonstrate that the age-related decline in memory is linked to the individual-knowledge level (for a review see Salthouse, 2002; Meinz & Salthouse, 1998). However, although Hedden et al. (2005) found that correlations between knowledge and performance on a cued-recall, episodic-memory task were similar in young and older adults, they observed that knowledge appeared to be more important for older than for younger adults in mediating cued recall, suggesting that older adults are more likely to take advantage of knowledge in this type of task than younger adults. With regard to control, the individual level of executive functioning was found to modulate age-related differences in episodic memory, suggesting that the control process is an important mediator of episodic-memory aging (Angel, Fay, Bouazzaoui, & Isingrini, 2010; Bugaiska et al., 2007; Crawford, Bryan, Luszcz, Obonsawin, & Steward, 2000; Troyer, Graves, & Cullum, 1994).

101

In line with this observation, recent functional neuroimaging studies focusing on episodic memory (see Craik & Rose, in press; Dennis & Cabeza, 2008; Grady, 2008; Park & Reuter-Lorenz, 2009 for reviews) have shown a greater activation of the PFC in older than in young adults. This phenomenon has been interpreted as a compensation mechanism helping to reduce age-related memory deficits by recruiting alternative networks (Dennis & Cabeza, 2008; Park & Reuter-Lorenz, 2009). As suggested by some authors (Park & Reuter-Lorenz, 2009), if one of the main roles of the PFC is to sustain executive functioning, it is possible that this additional PFC recruitment reflects a greater need of executive functions to sustain episodic-memory functioning. Accordingly, Glisky and Kong (2008) observed that an executive composite measure was associated with source memory in older but not young adults, suggesting that source memory places greater demands on control processes in older than in young adults. Thus, both behavioural and imaging data suggest that control processes play a crucial role in episodic memory in old age. Turning to the putative influential roles of representation and control in young and older adults, we can infer that episodic memory relies on representation in both young adulthood and old age, and on representation and control with increasing age. The aim of the present study was to examine the respective contributions to episodic memory performance of representation and control in young and older adults. We therefore assessed the performance on a cued-recall task of two age groups (young adults and older adults). The groups were also compared on two measures assumed to reflect representation abilities (WAIS Vocabulary Test, WAIS Information Test) and two assumed to reflect control abilities (Wisconsin Card-Sorting Test; WCST; Nelson, 1976 and Initial-Letter-Fluency Test; ILFT; Stuss & Benson, 1986). We expected first to confirm the classic pattern of an age-related change in episodic memory and control measures, with no agerelated effect on representation measures. In keeping with Craik and Bialystok (2006, 2008), our second objective was to confirm whether the two assumed components (representation and control) could be described as independent psychological dimensions. A factor analysis was used to achieve this objective, expecting that two independent factors would be derived from the four measures. Third, we performed correlational and regression analyses to explore the hypothesis that young and older adults present different patterns of association between episodic memory and representation, and between episodic memory and control. To our knowledge, few studies have examined these relationships based on comparison of these age groups. Studies that have approached this issue have examined the influence of the two factors separately (Hedden et al., 2005; Glisky & Kong, 2008). In the present study, we hypothesised that correlations between episodic-memory performance and control would be greater, or even only appear in older adults, indicating that memory relies more heavily on control in old age. On the other hand, we expected to find correlations between episodic memory and representation in young and older adults, indicating that representation continues to play an important role throughout adulthood. With regard to the regression analyses, we expected that control would be the main predictor of episodic-memory variance in older adults, whereas representation would be the main predictor for younger adults. This is consistent with the idea that factors that are central to memory performance differ between young and older adults, with

BOUAZZAOUI ET AL.

102

control becoming preponderant in old age. Finally, regression analyses conducted on the entire sample were performed in order to examine directly whether representation and/or control mediate age-group-related differences in episodic-memory performance.

months are there in a year?”). The score is the sum of correct answers (one point for each correct response).

Executive-Functioning Tests (to Measure Control Abilities)

Method Participants The sample comprised 120 participants divided into two age groups: 60 young adults aged between 25 and 46 (M ⫽ 34.51, SD ⫽ 6.72) and 60 older adults aged between 60 and 80 (M ⫽ 71.45, SD ⫽ 6.11). The proportion of men to women was similar in the two groups (young adults: 29 men and 31 women; older adults: 27 men and 33 women, ␹2(1) ⫽ .13, p ⬍ .71. A significant and classic effect of age was found on educational level, younger adults having completed more years of education (M ⫽ 12.38, SD ⫽ 3.66 for the young group; M ⫽ 10.45, SD ⫽ 4.04 for the older group; (t(1,118) ⫽ 2.45, p ⬍ .05. Due to the age-related differences in education, this variable was partialed out in the statistical analyses. All participants reported themselves to be in good physical and mental health and free from medication known to affect their intellectual abilities. They scored over the cut-off of 128 points on the Dementia Rating Scale (DRS, Mattis, 1976), ensuring that none suffered from dementia. The study was approved by the local ethics committee of the University of Tours, Tours, France, and all participants signed consent forms.

Materials and Design Episodic-memory task. Stimulus items consisted of a list of 24 common nouns (e.g., soldier) presented for 5 s on a computer screen. Learning was incidental, as the participants were not informed about the imminent cued-recall test. They were asked to give a pleasant/unpleasant judgment of each word, with the aim of guiding them toward a semantic processing of to-be-remembered information (Craik & Lockart, 1972). They were told that their responses would be used to construct verbal materials for future experiments. The presentation was immediately followed by a 1-min interference task (back counting) to avoid any recency effect. This was followed by a cued-recall test in which only the first three letters of the words were provided (e.g., sol__). Participants were instructed to complete the stems only with a studied word, at their own pace. Each stem was unique. Memory score was the total number of correctly recalled items. After the cued-recall test, participants took the representation and control tests.

Crystallized-Knowledge Tests (to Measure Representation Abilities) Two subtests of the WAIS (Wechsler, 1981) were administered. The vocabulary test consists of 35 words (including three examples), which participants are asked to define (e.g., “What does the word ‘fiction’ mean?”).The score is the sum of correct answers (two points for a complete definition and one point when the definition is incomplete). The information test consists of 29 questions (including four examples) dealing with general knowledge (e.g., “How many

The executive measures were two experimental tasks widely used in the neuropsychological literature (Lezak, 1995). The Wisconsin Card Sorting Test (WCST; Nelson, 1976), modified, is a standardized test which measures set formation and attention shifts. Participants have to sort cards containing multidimensional drawings into different categories (colour, shape, and number of geometric patterns) by deducing the sorting criterion from feedback given by the experimenter (“right” or “wrong”). This test is a goal-oriented task that provides several scores believed to reflect executive functioning. The specific measure retained in our study was the number of perseverative errors, which are the most affected by age (Bryan et al., 1999). The Initial Letter Fluency Test (ILFT; Stuss & Benson, 1986) in which participants are asked to produce as many words as possible beginning with the letters F, A and S in 1 min for each letter. The score is the total number of correctly produced words in three minutes.

Results The data were analysed in three ways. The first examined the age-group-related effect on episodic-memory, executive, and crystallized-knowledge measures. The second aimed to differentiate representation (crystallized-knowledge measures) from control (executive measures) through factor analysis. Third, correlation and regression analyses were performed to determine which type of ability (representation and/or control) could best explain episodic-memory performance in each age group and mediate age-group-related differences.

Age-Group Comparisons To examine age differences in episodic memory, crystallized knowledge, and executive functioning, t tests were performed on the cued-recall score, the two crystallized-knowledge measures, and the two executive-functioning measures. The mean scores of each measure are summarised in Table 1, showing that age group had a significant effect on cued recall and executive functioning, which decreased with age.1 An age-group effect was found on the information score, which increased with age, but not on the vocabulary score. 1

Note that we used a younger adult sample aged between 25 and 46 to enhance variability in cognitive measures within this group, at the risk of minimizing age-related differences between young and older groups. Indeed, a recent longitudinal study (Singh-Manoux et al., 2012), indicated that age-related declines in executive skills and episodic memory may occur in the 30s and 40s. Nevertheless, when we split the young adults into two groups of 25–35 and 36 – 46 years, the age-related differences in cued-recall and executive measures were not significant. Likewise, no age-related differences were observed between the young, old group (60 –70 years) and the old, old group (71– 80 years).

REPRESENTATION AND CONTROL IN MEMORY AGING

103

Table 1 Means and Standard Deviations of Cued-Recall, Crystallized-Knowledge, and Executive Measures

Cued recall Vocabulary Information WCST (perseverative errors) ILFT ⴱ

p ⬍ .05.

ⴱⴱⴱ

Young (n ⫽ 60)

Old (n ⫽ 60)

M (SD)

M (SD)

t(df ⫽ 118)

7.98 (1.76) 55.88 (2.68) 20.95 (1.80) 4.65 (2.98) 34.15 (3.87)

5.75 (1.46) 55.20 (2.04) 25.18 (1.53) 6.26 (5.25) 29.70 (6.07)

7.54ⴱⴱⴱ 1.56 ⫺13.82ⴱⴱⴱ ⫺2.07ⴱ 4.78ⴱⴱⴱ

p ⬍ .001.

Factor Analysis Our aim was to differentiate between crystallized knowledge and executive functioning, reflecting representation and control, and we therefore carried out a principal component analysis on the four variables (vocabulary, information, WCST, ILFT). The extracted factors were rotated using a varimax rotation, allowing independence between factors. Two factors with an eigenvalue greater than one were extracted. The factor-loading matrix is presented in Table 2. The two measures of executive functioning contributed mainly to the first factor, whereas the two crystallizedknowledge measures were mainly loaded on the second factor. This pattern confirms that crystallized-knowledge and executive functioning components can be viewed as independent psychological dimensions, and thus enabled us to interpret the first factor (accounting for 37.44% of the variance) as reflecting control abilities, and the second factor (accounting for 30.20% of the variance) as reflecting representation abilities. Two composite z-score indexes were calculated in each age group and for each participant, a control index (CI), corresponding to the average of the standardized z scores on the two executive tests (with WCST z-score errors multiplied by ⫺1), and a representation index (RI), corresponding to the average of the standardized z scores on the two crystallized-knowledge tests, and thus representing the performance of each participant on each of the two factors. A high score indicates a good level of representation and/or control.

memory performance was positively correlated with the two crystallized-knowledge measures and with the RI, indicating that individuals with a high level of representation also exhibited high memory performance, whereas memory performance was not significantly related to the executive measures and the CI. In the older group, memory performance was negatively correlated with perseverative errors on the WCST and positively correlated with all other measures, and thus correlated with both control and representation measures and the two indexes. This result suggests that older individuals with a high level of representation and/or control exhibited high memory performance. To determine whether representation or control constitutes the best predictor of memory performance in young and older adults, stepwise regression analyses using the RI and CI indexes were performed in each age group. In the young group, only the RI proved to be a significant predictor of the variance related to memory performance, accounting for 19% (p ⬍ .001). By contrast, the CI was the main significant predictor in the older group, accounting for 19% (p ⬍ .001) of the variance, the RI adding a significant 14% (p ⬍ .01). Overall, these results confirm that Table 3 Correlation Between Cued Recall and Crystallized-Knowledge Measures, RI, Executive Measures, and CI Cued recall

Correlation and Regression Analyses

Younger (n ⫽ 60)

Pearson correlations were used to assess associations between the cued-recall score and crystallized-knowledge measures, executive-functioning measures, the RI, and the CI. Correlation analyses revealed different patterns of results for young and older participants (see Table 3). In the young group,

Table 2 Loading Extracted from the Factor Analysis

WCST ILFT Vocabulary Information % of variance accounted for

Factor 1

Factor 2

⫺.79 .83 .24 ⫺.34 37.44

⫺.04 ⫺.07 .81 .73 30.20

Vocabulary Information RI1 WCST (perseverative errors) ILFT CI2

ⴱⴱⴱ

.42 .38ⴱⴱ .44ⴱⴱⴱ .03 .02 ⫺.03

Older (n ⫽ 60) .34ⴱⴱ .26ⴱ .37ⴱⴱ ⫺.40ⴱⴱ .33ⴱⴱ .43ⴱⴱⴱ

Note. We used a younger adult sample aged between 25 and 46 to enhance variability in cognitive measures within this group, at the risk of minimizing age-related differences between young and older groups. Indeed, a recent longitudinal study (Singh-Manoux et al., 2012), indicated that age-related declines in executive skills and episodic memory may occur in the 30s and 40s. Nevertheless, when we split the young adults into two groups of 25–35 and 36 – 46 years, the age-related differences in cued-recall and executive measures were not significant. Likewise, no age-related differences were observed between the young, old group (60 –70 years) and the old, old group (71– 80 years). 1 Representation index. 2 control index. ⴱ p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.

104

BOUAZZAOUI ET AL.

representation is important for younger adults’ memory, whereas both representation and control are significant predictors of the memory performance of older adults. These results remained unchanged in both groups after age and educational level had been controlled for. Finally, regression analyses were computed to examine whether representation and/or control factors mediate age-related differences in cued recall. To this end, we calculated the age-related variance in cued recall before and after partialing out representation and control factors. To test the significance of the modified proportion of age-related variance, we used the significance test provided by Howell (2006). The statistical control of the representation factor increased the magnitude of the age-related variance by 32.68% (t(120) ⫽ 33.00; p ⬍ .001), thereby making the relationship between age and memory performance more negative. By contrast, partialing out the control factor reduced the magnitude of the age-related variance in cued recall by 31.10% (t(120) ⫽ 64.34; p ⬍ .001), making the relationship between age and memory performance less negative.

Discussion In this study we examined whether young and older adults present different patterns of relationship between episodic memory and representation on the one hand and between episodic memory and control on the other. According to Craik and Bialystok (2006, 2008), this could explain age-related changes in factors influencing episodic-memory performance. First of all, the results of the factor analysis clearly differentiated two factors, one involving the two crystallized-knowledge measures and the other involving the two executive measures. This finding can be interpreted as reflecting the two psychological components, representation and control. Our data also replicate the classic finding that memory and control are more affected by age than representation (Park & Reuter-Lorenz, 2009). More precisely, our study confirms earlier findings (see Salthouse, 1991) that the age difference is not significant for vocabulary, but is significant for information, showing an increase with advancing age. Second, with regard to our main objective, the results clearly confirm a different pattern of correlations between episodic memory and control measures in older and younger adults. The correlations were significant for older but not for young adults. However, correlations with representation were significant in both age groups. Regression analyses confirmed this agerelated differential pattern by adding that the control index is the primary and greatest predictor of episodic-memory performance for older adults, although the representation index adds a reliable independent contribution. By contrast, this analysis yielded a different pattern in young adults, showing that the sole predictor appeared to be the representation index. The data from correlation and regression analyses also show that none of the significant associations were affected by adjustment for age and educational level in either group. This provides overall converging evidence of independent links between episodicmemory functioning and both representation and control variables. The pattern of findings observed in older adults suggests that the qualitative nature of memory processes may differ from that of younger adults. Finally, regression analysis showed that both representation and control factors modulated the age-

group-related differences in episodic-memory variance. These results support the idea put forward by Craik and Bialystok (2006, 2008) that representation and control are key cognitive variables within episodic memory, sustaining cognitive abilities, and explaining cognitive changes across the life span. Overall, the data confirm the hypothesis of an age-related dissociation between the factors that sustain episodic-memory functioning, raising a number of interesting questions. One concerns the mechanisms whereby representation and control influence memory functioning. The second concerns why memory abilities are not completely sustained by the same processes during early and late adulthood, with episodic memory seeming to rely on the control factor in old age, but not in earlier adulthood. Although further research is required to answer these questions, these issues and their implications for understanding memory functioning in older and young adults are discussed below from the perspective of Craik and Bialystok’s (2006, 2008) framework. With regard to representation, what is notable in our results is that the role played by this factor in episodic memory appeared to remain roughly stable throughout the adult life span; the correlations between episodic memory and representation were very similar in young and older adults, and this factor explained a significant variance of memory performance in both age groups. These results highlight the interesting point that episodic-memory performance is influenced by the representational system. Mechanisms linking episodic memory and representation can be interpreted through the traditional view that new learning is facilitated when it is congruent with existing schematic knowledge (Bäckman, Small, Wahlin, & Larsson, 2000; Bransford et al., 1979; Craik & Bialystok, 2006, 2008; Newcombe et al., 2011; Salthouse, 2002). Thus, more information in semantic memory increases the opportunities to bind newly learned items to existing schematic knowledge at encoding. In other words, good episodic-memory performance reflects good encoding in terms of the person’s organized schematic knowledge, which is also the basis for the level-ofprocessing effect (Craik & Lockhart, 1972; Craik & Tulving, 1975). Because there is no age-related decrease in knowledge, correlations with memory could also remain stable throughout adulthood. Our results are not fully in line with those of Hedden et al. (2005), which showed that knowledge plays a greater role in sustaining cued recall in older than in younger adults. However, the two studies differ in the nature of the cued-recall task used, in that the cues in Hedden et al.’s (2005) study were weak associates, whereas in the present study they were the three first letters of the target word. This point suggests that the increasing involvement of representation in cued recall with age may vary, depending on the type of memory process enhanced by different cues; representation is likely to play a greater role when the association between cues and targets is semantically based, providing better environmental support in relation to knowledge, than when this association is based on phonological features. It would be interesting to carry out further research to investigate this hypothesis. On the other hand, our observation of an age-related increase in representation and that the statistical control of representation increased the magnitude of the age-related variance is consistent with the view that the negative relation between age

REPRESENTATION AND CONTROL IN MEMORY AGING

and cued recall is attenuated when older age is associated with greater amount of knowledge. As suggested by Salthouse (2002), one possible interpretation of this pattern of result is based on the “migration” hypothesis, which postulates that the age-related effect on memory is reduced among people with high levels of representation because people tend to “migrate” into higher knowledge groups with increasing age. Thus, representation might function as an attenuating factor in the age– memory relationship when it is positively correlated with older age. This is consistent with similar findings observed in other types of cognitive measures (see Salthouse, 2002). With regard to the role of control, our results are in line with those of Glisky and Kong (2008) on source memory, confirming that this factor is associated with episodic memory more in older than in younger adults. According to Craik and Bialystok (2006, 2008), this suggests that the reduced-memory ability of older adults is probably mainly associated with control processes reflecting difficulty accessing stored information about items and/or aspects of source. This view is supported by observations that the most common memory complaint of older adults is their difficulty in recalling names and words that are specific labels, although this information has not been lost in memory as it can be retrieved later, either spontaneously or with better cues (Cohen & Burke, 1993). To explain the influential role of control on memory, we can refer to the view put forward by numerous authors, that executive functions serve to carry out higher order memory strategies to sustain episodic memory (Moscovitch & Winocur, 1992; Shimamura, 1995). These processes can be viewed as supporting the generation of memory-strategy processes, such as focusing and maintaining attention on abstract representations, to generate sequences of complex memory goals, implement deep encoding processes, or implement appropriate retrieval processes related to encoding orientation, to facilitate access to stored information (Bouazzaoui et al., 2010; Bryan et al., 1999; Taconnat et al., 2006; Taconnat et al., 2007; Taconnat et al., 2009). Given that executive functioning is mainly sustained by the PFC, the stronger relationship between executive functions and episodic-memory measures found in older adults may also help explain recent functional imaging findings showing age-related differences in the use of the PFC, with greater activation in older than young adults (Cabeza, 2002; Dennis & Cabeza, 2008). As PFC sustains executive functioning, this additional PFC recruitment in older adults may reflect greater use of executive functions to compensate for age-related, reducedcognitive activity (Park & Reuter-Lorenz, 2009). The fact that executive functioning and episodic memory were only correlated in older adults suggests that this support is needed mainly in old age. The influence of control in old age was confirmed by regression analysis, which showed a moderate but significant decrease in the age-related variance in cued recall when the control factor was controlled for. This result supports a moderation interpretation, indicating that older adults with a high level of control exhibit smaller age-related declines in cued-recall performance than those with a low level of control. This suggests that the greater resort of older adults to control processes may serve as a compensatory process to limit age-related memory difficulties. Thus, despite the age-related decrease in control, older adults are more likely to rely on control processes than younger adults under similar memory-

105

task conditions. It should be noted that these results are not entirely consistent with findings of an age-related decline in recollection processes, which are classically associated with control (Jacoby, 1999; Jennings & Jacoby, 1997). This suggests that the correlation between memory and control would be greater in younger than in older adults. By contrast, our results show that, despite an agerelated decrease in control, this factor affects the memory performance of older more than younger adults, suggesting a specific interplay between decrease and use of control processes with advancing age. Although control declines with advancing age, it seems to be needed more to aid memory performance, particularly when encoding and retrieval are somewhat effortful. A final point to consider concerns the observation that episodic memory in young adults does not rely on executive functioning. This may seem surprising, and at first sight, suggests a structural difference between young and older adults. Alternatively, executive functions can be seen as memory-support processes enabling optimum performance when there are insufficient memory resources, as for example in the case of normal aging, or due to the level of task difficulty. In our study, the absence of relationship between control and memory in the younger group may reflect the fact that we used a memory task that limited the role of strategic and executive control because encoding was incidental and sustained by a semantically oriented task (judgment of agreement), and was followed by cued-recall retrieval. It could thus be supposed that if encoding and retrieval conditions were more open to strategy, then control would also become important for younger adults. A similar observation could be made about the effortful level of tests used to assess control. However, in line with our hypothesis of a differential pattern between younger and older adults, more stringent conditions could make control even more salient in older adults. Thus, although the young adults did not rely on executive functions to perform our memory task, they could call on them when carrying out particularly demanding memory tasks, such as for example a free-recall task. It would be interesting to carry out further research to investigate whether the pattern of the relationship between executive functioning and episodic-memory tasks varies with the level of task difficulty in young and older adults, for example using memory tasks differing in their dependence on strategic control at encoding and retrieval.

Re´sume´ Craik et Bialystok (2006, 2008) sugge`rent que l’e´tude des modifications des connaissances et des processus de controˆle tout au long de la vie pourraient permettre de comprendre les changements cognitifs lie´s a` l’aˆge. L’objectif de cette e´tude est d’explorer l’hypothe`se selon laquelle les connaissances cristallise´es et les processus de controˆle seraient diffe´remment implique´s dans la performance de me´moire e´pisodique chez les adultes jeunes et aˆge´s. Des adultes jeunes et aˆge´s ont e´te´ soumis a` une taˆche de rappel indice´ ainsi qu’a` des tests d’intelligence cristallise´e et de fonctions exe´cutives afin de mesurer respectivement leur niveau de connaissance et de controˆle. Les re´sultats reproduisent les donne´es classiques selon lesquelles le controˆle exe´cutif et le rappel indice´ diminuent avec l’aˆge alors que les connaissances cristallise´es se maintiennent. Une analyse factorielle a confirme´ l’inde´pendance de ces deux facteurs. Des analyses de corre´lation ont montre´ que la performance mne´sique des jeunes adultes est corre´le´e avec les mesures de connaissances

BOUAZZAOUI ET AL.

106

cristallise´es mais pas avec les mesures de controˆle, alors que la performance mne´sique des adultes aˆge´s s’ave`re corre´le´e avec les mesures de connaissances cristallise´es et de controˆle. Des analyses de re´gression ont indique´ que le facteur controˆle est le principal pre´dicteur de la performance mne´sique chez les adultes aˆge´s, le facteur connaissances apportant une contribution inde´pendante. Cependant, chez les adultes jeunes, le facteur connaissances apparaıˆt comme le seul pre´dicteur. Ce re´sultat indique que les facteurs qui sous-tendent les performances mne´siques e´voluent avec l’aˆge. Ainsi, les connaissances apparaissent importantes tout au long de l’aˆge adulte alors que le controˆle devient un facteur pre´ponde´rant chez les adultes aˆge´s pour le soutien de l’activite´ mne´sique. Mots-cle´s : me´moire e´pisodique, vieillissement, connaissances cristallise´es, controˆle exe´cutif.

References Angel, L., Fay, S., Bouazzaoui, B., & Isingrini, M. (2010). Individual differences in frontal functioning modulate age effects on the ERP correlates of retrieval success. Neuropsychologia, 48, 3540 –3553. doi: 10.1016/j.neuropsychologia.2010.08.003 Bäckman, L., Small, B. J., Wahlin, A., & Larsson, M. (2000). Cognitive functioning in late senescence. In F. I. M. Craik & T. A. Salthouse (Eds.), Handbook of cognitive aging: Vol. 2 (pp. 499 –558). Hillsdale, NJ: Erlbaum. Baddeley, A. (1996). Exploring the central executive. The Quartely Journal of Experimental Psychology, 49A, 5–28. Bouazzaoui, B., Isingrini, M., Fay, S., Angel, L., Vanneste, S., Clarys, D., & Taconnat, L. (2010). Aging and self-reported internal and external memory strategy uses: The role of executive functioning. Acta Psychologica, 135, 59 – 66. doi:10.1016/j.actpsy.2010.05.007 Bransford, J. D., Franks, J. J., Morris, C. D., & Stein, B. S. (1979). Some general constraints on learning and memory research. In L. S. Cermak & F. I. M. Craik (Eds.), Levels of processing in human memory (pp. 331–354). Hillsdale, NJ: Lawrence Erlbaum. Bryan, J., Luszcz, M. A., & Pointer, S. (1999). Executive function and processing resources as predictors of adult age differences in the implementation of encoding strategies. Aging, Neuropsychology, and Cognition, 6, 273–287. Bugaiska, A., Clarys, D., Jarry, C., Taconnat, L., Tapia, G., Vanneste, S., & Isingrini, M. (2007). The effect of aging in recollective experience: The processing speed and executive functioning hypothesis. Consciousness and Cognition, 16, 797– 808. doi:10.1016/j.concog.2006.11.007 Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: The HAROLD model. Psychology and Aging, 17, 85–100. doi:10.1037/ 0882-7974.17.1.85 Cattell, R. B. (1971). Abilities: Their structure, growth and action. Boston, MA: Houghton Mifflin. Cohen, G., & Burke, D. M. (1993). Memory for proper names: A review. Memory, 1, 249 –263. doi:10.1080/09658219308258237 Craik, F. I. M., & Bialystok, E. (2006). Cognition throughout the lifespan: Mechanisms of change. Trends in Cognitive Sciences, 10, 131–138. doi:10.1016/j.tics.2006.01.007 Craik, F. I. M., & Bialystok, E. (2008). Lifespan cognitive development: The role of representation and control. In F. I. M. Craik & T. A. Salthouse (Eds.), The handbook of aging and cognition (3rd ed., pp. 557– 601). New York, NY: Psychology Press. Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning & Verbal Behavior, 11, 671– 684. doi:10.1016/S0022-5371(72)80001-X

Craik, F. I. M., & Rose, N. S (in press). Memory encoding and aging: A neurocognitive perspective. Neuroscience and Biobehavioral Reviews. Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104, 268 –294. doi:10.1037/0096-3445.104.3.268 Crawford, J. R., Bryan, J., Luszcz, M. A., Obonsawin, M. C., & Stewart, L. (2000). The executive decline hypothesis of cognitive aging: Do executive deficits qualify as differential deficits and do they mediate age-related memory decline? Aging, Neuropsychology, and Cognition, 7, 9 –31. doi:10.1076/anec.7.1.9.806 Dennis, N. A., & Cabeza, R. (2008). Neuroimaging of healthy cognitive aging, in Craik, F. I. M., Salthouse, T. A. (Eds.), The handbook of aging and cognition (3rd ed., pp. 1–54). New York, NY: Psychology Press. Elliott, R. (2003). Executive functions and their disorders. British Medical Bulletin, 65, 49 –59. doi:10.1093/bmb/65.1.49 Glisky, E. L., & Kong, L. L. (2008). Do young and older adults rely on different processes in source memory tasks? A neuropsychological study. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 809 – 822. doi:10.1037/0278-7393.34.4.809 Grady, C. L. (2008). Cognitive neuroscience of aging. Annals of the New York Academy Science, 1124, 127–144. doi:10.1196/annals.1440.009 Hedden, T., Lautenschlager, G., & Park, D. C. (2005). Contributions of processing ability and knowledge to verbal memory tasks across the adult life-span. The Quarterly Journal of Experimental Psychology, 58A, 169 –190. Horn, J. L. (1982). The theory of fluid and crystallized intelligence in relation to concepts of cognitive psychology and aging in adulthood. In F. I. M. Carik & S. Trehub (Eds.), Aging and cognitive processes (pp. 237–278). New York, NY: Plenum Press. Howell, D. C. (2006). Statistical methods for psychology (6th ed.). Stamford, CT: Cengage Learning. 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. doi:10.1037/0278-7393.25.1.3 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. doi:10.1037/0882-7974 .12.2.352 Lezak, M. D. (1995). Neuropsychological assessment. New York, NY: Oxford University Press. Mattis, S. (1976). Mental status examination for organic mental syndrome in the elderly patient. In L. Bellack & T. B. Karasu (Eds.), Geriatric psychiatry (pp. 77–121). New York, NY: Grune & Stratton. McDaniel, M. A., Einstein, G. O., & Jacoby, L. L. (2008). New considerations in aging and memory: The glass may be half full. In F. I. M. Craik & T. A. Salthouse. (Eds), Handbook of aging and cognition (3rd ed., pp. 251–310). Mahwah, NJ: Erlbaum. Meinz, E., & Salthouse, T. A. (1998). The effects of age and experience on memory for visually presented music. Journal of Gerontology: Psychological science, 53B, 60 – 69. Moscovitch, M., & Winocur, G. (1992). The neuropsychology of memory and aging, in F. I. M. Craik & T. A. Salthouse. (Eds.), The handbook of aging and cognition (pp. 315–372). Hillsdale, NJ: Lawrence Erlbaum. Nelson, H. E. (1976). A modified card sorting test sensitive to frontal lobe deficits. Cortex: A Journal Devoted to the Study of the Nervous System and Behavior, 12, 313–324. Newcombe, N. S., Lloyd, M. E., & Balcomb, F. (2011). Contextualizing the development of recollection: Episodic memory and binding in young children. In S. Ghetti & P. J. Bauer (Eds.), Origins and development of recollection: Perspectives from psychology and neuroscience. (pp. 73– 100). New York, NY: Oxford University Press. Nyberg, L., Cabeza, R., & Tulving, E. (1996). PET studies of encoding and retrieval: The HERA model. Psychonomic Bulletin & Review, 3, 135– 148. doi:10.3758/BF03212412

REPRESENTATION AND CONTROL IN MEMORY AGING Park, D. C., & Reuter-Lorenz, P. (2009). The adaptive brain: Aging and neurocognitive scaffolding. Annual Review of Psychology, 60, 173–196. doi:10.1146/annurev.psych.59.103006.093656 Salthouse, T. A. (1991). Theoretical perspectives on cognitive aging. Hillsdale, NJ: Erlbaum. Salthouse, T. A. (2002). Interrelations of aging, knowledge, and cognitive performance. In: U. Staudinger & U. Lindenberger (Eds.), Understanding human development: Dialogues with lifespan psychology (pp. 265– 287). Dordrecht, the Netherlands: Kluwer. Shallice, T., & Burgess, P. (1993). Supervisory control of action and thought selection. In A. Baddeley & L. Weiskrantz (Eds.). Attention: Selection, awareness and control (pp. 171–187). Oxford, UK: Clarendon Press. Shimamura, A. P. 1995. Memory and frontal lobe function. In Gazzaniga, M. S. (Eds.), The cognitive neurosciences (pp. 803– 813). Cambridge, MA: MIT Press. Singh-Manoux, A., Kivimaki, M., Glymour, M. M., Elbaz, A., Berr, C., Ebmeier, K., . . . Dugravot, A. (2012). Timing of onset of cognitive decline: Results from Whitehall II prospective cohort study. British Medical Journal, 344, d7622– d7629. doi:10.1136/bmj.d7622 Stuss, D. T., & Benson, D. F. (1986). The frontal lobes. New York, NY: Raven Press. Taconnat, L., Baudouin, A., Fay, S., Clarys, D., Vanneste, S., Tournelle, L., & Isingrini, M. (2006). Aging and implementation of encoding strategies in the generation of rhymes: The role of executive functions. Neuropsychology, 20, 658 – 665. doi:10.1037/0894-4105.20.6.658

107

Taconnat, L., Clarys, D., Vanneste, S., Bouazzaoui, B., & Isingrini, M. (2007). Aging and strategic retrieval in memory: The role of executive functions. Brain and Cognition, 64, 1– 6. doi:10.1016/j.bandc .2006.09.011 Taconnat, L., Raz, N., Tocze´, C., Bouazzaoui, B., Sauze´on, H., Fay, S., & Isingrini, M. (2009). Aging and organization strategies in free recall: The role of cognitive flexibility. European Journal of Cognitive Psychology, 21, 347–365. doi:10.1080/09541440802296413 Troyer, A. K., Graves, R. E., & Cullum, C. M. (1994). Executive functioning as a mediator of the relationship between age and episodic memory in healthy aging. Aging, Neuropsychology, and Cognition, 1, 45–53. doi:10.1080/09289919408251449 Wechsler, D. (1981). WAIS-R manual. New York, NY: The Psychological Corporation. Wheeler, M. A., Stuss, D. T., & Tulving, E. (1995). Frontal lobe damage produces episodic memory impairment. Journal of the International Neuropsychological Society, 1, 525–536. doi:10.1017/ S1355617700000655 Wheeler, M. A., Stuss, D. T., & Tulving, E. (1997). Toward a theory of episodic memory: The frontal lobes and autonoetic consciousness. Psychological Bulletin, 121, 331–354. doi:10.1037/0033-2909.121.3.331

Received December 9, 2011 Accepted March 30, 2012 䡲

E-Mail Notification of Your Latest CPA Issue Online! Would you like to know when the next issue of your favorite Canadian Psychological Association journal will be available online? This service is now available. Sign up at http://notify.apa.org/ and you will be notified by e-mail when issues of interest to you become available!

Avis par courriel de la disponibilite´ des revues de la SCP en ligne! Vous voulez savoir quand sera accessible en ligne le prochain nume´ro de votre revue de la Socie´te´ canadienne de psychologie pre´fe´re´e? Il est de´sormais possible de le faire. Inscrivez-vous a` http://notify.apa.org/ et vous serez avise´ par courriel de la date de parution en ligne des nume´ros qui vous inte´ressent!