Executive functioning and processing speed in age-related differences in memory: Contribution of a coding task

Executive functioning and processing speed in age-related differences in memory: Contribution of a coding task

Brain and Cognition 71 (2009) 240–245 Contents lists available at ScienceDirect Brain and Cognition journal homepage: www.elsevier.com/locate/b&c E...

170KB Sizes 3 Downloads 87 Views

Brain and Cognition 71 (2009) 240–245

Contents lists available at ScienceDirect

Brain and Cognition journal homepage: www.elsevier.com/locate/b&c

Executive functioning and processing speed in age-related differences in memory: Contribution of a coding task Alexia Baudouin *, David Clarys, Sandrine Vanneste, Michel Isingrini UMR-CNRS-6234, Centre de Recherches sur la Cognition et l’Apprentissage, Université de François-Rabelais de Tours, France

a r t i c l e

i n f o

Article history: Accepted 16 August 2009 Available online 30 September 2009 Keywords: Aging Memory Executive function Processing speed

a b s t r a c t The aim of the present study was to examine executive dysfunctioning and decreased processing speed as potential mediators of age-related differences in episodic memory. We compared the performances of young and elderly adults in a free-recall task. Participants were also given tests to measure executive functions and perceptual processing speed and a coding task (the Digit Symbol Substitution Test, DSST). More precisely, we tested the hypothesis that executive functions would mediate the age-related differences observed in the free-recall task better than perceptual speed. We also tested the assumption that a coding task, assumed to involve both executive processes and perceptual speed, would be the best mediator of age-related differences in memory. Findings first confirmed that the DSST combines executive processes and perceptual speed. Secondly, they showed that executive functions are a significant mediator of age-related differences in memory, and that DSST performance is the best predictor. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction One explanatory hypothesis of cognitive aging is that the agerelated decline in cognition is due to decreased executive functioning (Braver & West, 2008). This hypothesis is based on the observation that modifications in cognition in normal aging are similar to those in patients with frontal lobe damage (Daigneault, Braun, & Whitaker, 1992; Isingrini & Vazou, 1997). Memory studies have shown that older adults and frontal patients have impaired performance on the same tests, for example free recall (Craik & McDowd, 1987), source memory (Glisky, Rubin, & Davidson, 2001), memory of temporal order (Parkin, Walter, & Hunkin, 1995), conscious awareness in memory (Bugaiska et al., 2007), and metamemory (Perrotin, Isingrini, Souchay, Clarys, & Taconnat, 2006). The study of the relationship between age-related executive dysfunctioning and memory deficit suggests that executive functions play an important mediating role in the age-related variance observed in episodic memory (Parkin, 1997; Perfect, 1997). Moreover, neuropsychology and neuroimagery research has shown that frontal lobe structures are particularly vulnerable to advancing age (see Raz & Rodrigue, 2006; West, 1996 for a review). This brain region is considered to play an important role in memory-related operations such as information encoding, storage and retrieval, and it has been

* Corresponding author. Address: UMR-CNRS-6234, Centre de Recherches sur la Cognition et l’Apprentissage, Université de François-Rabelais de Tours, 3 rue des tanneurs, 37041 Tours Cedex 1, France. E-mail address: [email protected] (A. Baudouin). 0278-2626/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bandc.2009.08.007

suggested that its deterioration contributes to age-related memory decline (Head, Rodrigue, Kennedy, & Raz, 2008). There is also a wealth of literature indicating that processing speed is a limiting factor for most aspects of cognition in aging (Salthouse, 1996, 2000), notably memory (Bryan & Luszcz, 1996; Salthouse & Babcock, 1991; Slivinski & Buschke, 1997). It has been shown that statistically controlling for processing speed substantially reduces or eliminates age-related variance in a variety of episodic memory measures: free recall (Bryan & Luszcz, 1996), paired association (Salthouse, 1994), text memory (Lindenberger, Mayr, & Kliegl, 1993), and cued recall (Park et al., 1996). Studies comparing different predictors of the age-related deficit observed in memory, using various episodic memory tasks, indicate that processing slowing is the main mediator (Bryan & Luszcz, 1996; Clarys et al., 2007; Crawford, Bryan, Luszcz, Obonsawin, & Stewart, 2000; Parkin & Java, 2000). Crawford et al. (2000) showed that speed performance accounted for more of the age-related variance than either executive functions or general cognitive ability in all the measures of episodic memory studied. Comparing executive functioning, processing speed and fluid intelligence as potential mediators of the age-related deficit in episodic memory performance, Parkin and Java (2000) also found that processing speed was the best predictor of memory performance and that no other independent variables explained the memory deficit. Similar findings have recently been obtained by Clarys et al. (2007) when examining the influence of executive functions and processing speed on three episodic memory tasks: recall, recognition and story free recall.

241

A. Baudouin et al. / Brain and Cognition 71 (2009) 240–245

In all these studies, processing speed emerges as a more fundamental mediator of age-related differences than executive functioning in a number of episodic memory measures. However, they all used a substitution coding task (the Digit Symbol Substitution Test – DSST, Wechsler, 1981) to measure speed. When this task was contrasted with other speed tests, it was found to account for the age effects on memory performance better than the others (Bryan & Luszcz, 1996; Salthouse, 1993). A number of authors have suggested that a substitution coding task is not just a measure of speed and have proposed different processes underlying variation in performance on the DSST (Parkin & Java, 2000; Piccinin & Rabbitt, 1999; Salthouse, 1992). Studies of processing speed theory have described a number of different speed measures, for example, motor speed, perceptual speed, and decision speed (Salthouse, 1993, 2000). The latter involves completing cognitive tasks which have moderately complex content. The DSST may be more closely connected to decision/cognitive speed measures (Salthouse, 2000). Moreover, the coding task is dependent on different cognitive abilities, such as perception, intelligence and memory (Bell, Franzen, & Riley, 1999; Piccinin & Rabbitt, 1999; Salthouse, 1992; Wechsler, 1987). Parkin and Java (2000) suggested that the substitution coding task is linked to concepts developed in the framework of working memory theory (Baddeley, 2000). The assumption that executive control is involved in the coding task is plausible, as it requires higher order cognitive processes, such as planning, implementing strategies, monitoring performance, inhibiting irrelevant information, and directing attention towards the processing and temporary storage of information (Baddeley, 2000; Rabbitt, 1997), i.e., executive functioning. The coding task thus involves a combination of factors, notably processing speed and executive control. The first aim of the present study was to extend investigations into the mediation of age-related differences in episodic memory by testing the executive-aging and speed-mediation hypotheses, notably when perceptual speed tasks are used to assess processing speed. We hypothesized that executive functions would account for more of the age-related variance in episodic memory than perceptual speed. As described above, the DSST is thought to be a reliable tool for predicting the age-related differences in memory (Bryan & Luszcz, 1996; Parkin & Java, 2000) and to be a multi-process task (Parkin & Java, 2000; Salthouse, 1992). Our second aim was thus to investigate whether it does indeed explain age-related differences in episodic memory better than executive functions and perceptual speed. The third aim was to test the assumption that it combines several processes, notably perceptual speed and executive functions.

2. Methods 2.1. Participants A total of 100 adults took part in this study, 46 young adults (age range 20–34 years) and 54 older adults (age range 60– 96 years). Their demographic characteristics are summarized in Table 1. All subjects reported themselves to be in good physical and mental health and not to be taking medication known to affect the central nervous system. The older adults in the sample were screened for dementia on the Mini-Mental Status Examination, and scored at or above the cut-off of 27 points (MMSE, Folstein, Folstein, & McHugh, 1975). There was no difference between the two groups in knowledge level as assessed by their score on the Mill Hill vocabulary test (Raven, 1982), although younger adults had significantly more years of education. The number of years of education was therefore a covariate in the analyses of variance and entered (forced entry) in the regression analyses. There was

Table 1 Characteristics of participants in the two study groups (means and standard deviation) (n = 100). Characteristics

Age (years) Sexe (F/M) Education (years) Mill Hill vocabulary Depression score (HADS) Anxiety score (HADS) ***

Younger (n = 46)

Older (n = 54)

M

(SD)

M

(SD)

29.1 28/18 15 26.8 5.1 6.3

(3.8)

74.7 38/16 11.2 26.5 6.1 5.9

(8.8)

(1.7) (2.9) (2.4) (3.3)

(2.6) (5.4) (3.1) (3.4)

t-test

– 8.6*** 0.37 1.82 0.5

p < .001.

no difference in the depression and anxiety sub-scale scores of the Hospital Anxiety and Depression Scale, with cut-offs of 11 points (HADS, Zigmond & Snaith, 1983). 3. Material and procedure 3.1. Memory test (free recall of words) For the study phase, a list of 36 taxonomically unrelated concrete words was used. Words were type-written on cards and presented at a rate of 4 s/word. Participants were instructed to read the words aloud and to remember them for a later test. In the test phase lasting 5 min, the participants were asked to recall and write down as many words as they could. The recall test score was the number of correctly recalled words. 3.2. Executive tests The following executive tests were chosen because they are commonly used as measures of frontal functioning (Glisky, Polster, & Routhieaux, 1995; Glisky et al., 2001). Wisconsin Card Sorting Test (WCST – Modified; Nelson, 1976): the WCST is a standardized test for measuring the ability to shift cognitive strategies or sets in response to changing environmental contingencies. Participants have to sort cards containing multidimensional drawings into three different categories (color, form, and number of geometric patterns). The score is the number of correctly completed categories (Glisky et al., 1995, 2001; Rhodes, 2004). Running span: This task is a standard measure of updating function, and is associated with working memory and central executive functions (Baddeley, 1996; Doiseau & Isingrini, 2005; Miyake et al., 2000). Participants were read seven sets of letters. All the letters were consonants with six to twelve letters per set. They were read aloud at a rate of one letter per second. The sets were randomly presented. Participants were instructed to store in memory the last six letters only, for recall in the original order after the presentation. The score was the number of correct responses. Computation span: This task involves solving a series of arithmetic problems while also remembering the last digit from each problem (Salthouse & Babcock, 1991). The arithmetic problems were presented orally at a normal speaking rate. After presentation of each problem, participants had to select the correct answer from three alternatives on an answer sheet. The number of arithmetic problems presented in each trial increased successively from one to seven, with three trials for each length. The score was the total number of problems for which both the answer to the arithmetic problem and the recalled digit were correct. 3.3. Processing speed tests The two standard measures of speed used in this study are assumed to measure perceptual speed, in that they involve simple

242

A. Baudouin et al. / Brain and Cognition 71 (2009) 240–245

same/different decisions (Salthouse, 2000). Two versions of the tests were administered, one with numbers and one with letters. Number comparison test (Salthouse & Babcock, 1991): Participants had to compare two rows of three, six or nine numbers and decide whether the pairs were the same or different. They were given 30 s to make as many comparisons as possible. The score was the total number of correct comparisons. Letter comparison test (Salthouse & Babcock, 1991): In this test, participants had to compare two rows of three, six or nine letters and decide whether the pairs were the same or different. They were given 30 s to make as many comparisons as possible. The score was the total number of correct comparisons. 3.4. Coding task Digit Symbol Substitution Test (DSST, Wechsler, 1981): Participants were shown a code table with pairs of digits and symbols, and rows of double boxes with a digit in the top box and nothing in the bottom box. The task was to use the code table to find the symbol associated with each digit in the box, and write as many symbols as possible in the empty boxes below each digit in 30 s. 4. Results

Table 2 Means and standard deviations of scores for episodic memory task, executive functioning tests, and processing speed tests (n = 100). Variables

Episodic memory Free recall

Younger (n = 46)

Older (n = 54)

M

M

4.1. Age-related differences in episodic memory, executive functions and processing speed measures As recommended by Bryan and Luszcz (1996), the ‘‘age group” variable was used rather than chronological age in all the analyses. Participants were divided into two distinct age groups, coded as follows: young adults = 1, older adults = 2. To examine potential age-related differences, a series of covariance analyses was conducted on executive function measures, speed-processing scores, and episodic memory performance, with the number of years of education as covariate. The scores are summarized in Table 2. The results revealed that age had a significant impact on every measure: free recall (p < .001), WCST (p < .05), running span (p < .001), computation span (p < .001), number comparison (p < .001), letter comparison (p < .001), and DSST (p < .001). Simple correlations between all the measures were calculated and are shown in Table 3. All the correlations were significant (p < .001), making it possible to investigate relationships between the variables. In the following analyses, composite scores for each construct were created by averaging the z-scores for the WCST, running span, and computation span for executive functions, and averaging the zscores for number comparison and letter comparison for perceptual processing speed. The DSST was examined separately. 4.2. Relationships between the DSST, executive functions and processing speed As mentioned in the introduction, it was hypothesized that performance on the coding task would be closely related to both

F

(SD)

9.2

(3.2)

4.2

(2.2)

27.34***

Executives measures WCST Running span Computation span

5.9 25.3 15.1

(0.1) (4.9) (7.2)

5.4 18.3 5.5

(0.9) (5.1) (2.7)

6.34* 19.19*** 32.25***

Speed measures Number comparison Letters comparison DSST

10.8 14.63 27.7

(1.6) (2.5) (3.1)

7.1 8.4 15.2

(1.9) (2.3) (4.1)

35.50*** 66.75*** 135.05***

Note: WCST = Wisconsin Card Sorting Test; DSST = Digit Symbol Substitution Test. * p < .05. *** p < .001.

Table 3 Correlations between age, WCST, running span, computation span, executive index, number comparison, letter comparison, speed index, DSST and free-recall task. Age

The data were analyzed in three ways. The first analysis examined age-related differences in the cognitive measures (analyses of variance). The second examined the relationships between executive functions and processing speed measures and DSST performance (correlation analyses). Our main purpose here was to investigate the structure of DSST. The third analysis assessed the relationship between episodic memory and executive functions, processing speed and DSST scores, in order to determine the extent to which these variables account for age-related differences in episodic memory performance (regression analyses).

(SD)

1. 2. 3. 4. 5. 6. 7. 8. 9.

WCST Runspan span Computation span Executive index Number comparison Letter comparison Speed index DSST Free recall

1

2

3

4

5

.47 .59 .69 .74 .75

– .34 .40 .75 .48

– .54 .79 .61

– .82 .64

– .74



.82 .81 .90 .72

.50 .51 .59 .46

.63 .65 .64 .46

.66 .68 .70 .58

.76 .78 .82 .64

.84 .96 .80 .53

6

7

8

– .96 .87 .58

– .87 .58

– .74

Note: All correlations are significant at p < .001.

executive functioning and processing speed. The structure of the DSST was first investigated by examining the relationship between DSST performance and both perceptual speed and executive function scores using correlation analyses. Partial correlations were calculated to isolate any independent relationships between the executive and speed indexes and the DSST score. To this end, the partial correlation between the executive score and the DSST score was computed after partialling out the processing speed score, the age group variable and the number of years of education. Conversely, the partial correlation between the processing speed score and the DSST score was computed after partialling out the executive score, the age group variable and the number of years of education. For the sample as a whole, the results revealed two similar and significant correlations: the partial correlation between the executive score and the DSST score was .41 (p < .001), and the partial correlation between the perceptual speed score and the DSST score was .41 (p < .001). This pattern of results indicates that executive functions and processing speed were equally involved in the coding task. The same partial correlations were calculated in the two age groups separately. In the younger adult group, no significant correlation was observed between the executive score and the DSST score (.18), but the correlation between the perceptual speed score and the DSST score was significant (.35, p < .05). In the older group, the same correlation pattern as in the whole sample was obtained with two similar and significant correlations: a partial correlation between the executive score and the DSST score of .56 (p < .001), and a partial correlation between the perceptual speed score and the DSST score of .48 (p < .001).

A. Baudouin et al. / Brain and Cognition 71 (2009) 240–245

4.3. Relationships between episodic memory, executive functions, processing speed and the DSST The relationships between episodic memory, executive functions and perceptual speed were examined in two ways. The first aim was to determine whether executive functioning accounts for age-related differences in memory performance as much as processing speed, when processing speed is assessed only by a simple perceptual measure. Secondly, we tested DSST performance as a mediator of age-related variance in episodic memory in comparison with the other two variables. In order to address these aims, we used a series of forced entry regression analyses, with the number of years of education forced. The results of the analyses are shown in Table 4. For the first aim, five regression models were evaluated (Table 4). In model 1, age was entered as a predictor to determine the amount of age-related variance in memory performance. Model 2 looked at whether executive functioning reduced the contribution of age to a non-significant amount. The same procedure was repeated for the third model with perceptual speed. Model 4 examined whether perceptual speed continued to account for a significant amount of episodic memory variance once executive functions had been controlled for. Model 5 examined whether executive functions continued to account for a significant amount of episodic memory variance once perceptual speed had been controlled for. The percentage of the age-related variance (% ARV) accounted for by the mediating variables is also shown in Table 4 [i.e., ((R2 with age alone R2 change)/R2 with age alone)  100].

Table 4 Prediction of age-related variance in episodic memory (free recall) when number of education years, executive functioning, perceptual processing speed and DSST are partialled out (n = 100). Equation

Step

Variable

R2

1

1

.37

2

2 1

3

2 3 1

4

2 3 1

5

2 3 4 1

6

2 3 4 1

7

2 3 1

Number of education years Age group Number of education years Executive functions Age group Number of education years Processing speed Age group Number of education years Executive functions Processing speed Age group Number of education years Processing speed Executive functions Age group Number of education years DSST Age group Number of education years Processing speed Executive functions Age group DSST

2 3 4 5

DR 2

% ARV explained

.51 .37

.138***

.50 .55 .37

.127*** .047**

65.94

.42 .51 .37

.050** .088***

36.23

.50 .50 .56 .37

.126*** .0001 .061**

55.80

.42 .50 .56 .37

.050** .077*** .061**

55.80

.57 .57 .37

.198*** .002

98.55

.42 .50 .56 .62

.050** .076*** .060** .059***

243

Age predicted 13.8% of the variance in free recall when entered after controlling for the number of years of education (model 1); this represented 100% of the age-related variance obtained in the free-recall task. Model 2 showed that executive functions accounted for 65.94% of the age-related variance. In model 3, perceptual speed accounted for 36.23% of the age-related variance. Furthermore, model 4 showed that perceptual speed added a non-significant percentage of episodic memory variance once executive functioning had been controlled for. In contrast, model 5 showed that executive functioning continued to predict a significant 7.7% (p < .001) of episodic memory variance after perceptual speed had been entered. In models 4 and 5, age continued to account for a significant 6.1% of episodic memory variance once perceptual speed and executive functioning had been controlled for. It can be noted that the last two models also indicate a potential dependency between processing speed and executive function, as they accounted together for less age-related variance in episodic memory than the executive function measure alone (55.8% vs. 65.9% ARV). In summary, models 1–5 indicated that the executive functioning index was a better mediator of age-related variance in episodic memory than the perceptual speed index. When each measure was entered alone, mediation of the age-related variance was only 27– 51%. More precisely, for executive measures, WCST, running span and computation span accounted respectively for 29%, 28%, and 51% of the age-related variance of the free-recall task. For perceptual speed measures, number comparison and letter comparison accounted respectively for 27% and 36% of the variance. Nevertheless, taking each index or measure alone, age continued to be a significant contributor to the variance observed in memory after partialling out executive and speed measures. Model 6 examined the DSST as a predictor of age-related effects on free recall and looked at whether age continued to account for a significant amount of the episodic memory variance once the DSST had been controlled for (Table 4). It showed that the DSST accounted for 98.5% of the age-related variance reduced to a non-significant amount. Model 7 indicated that when perceptual speed, executive measures and age group were entered first, DSST continued to predict a significant 5.9% (p < .001) of episodic memory variance. Overall, these results indicate first that executive functions are a better mediator of age-related variance in episodic memory than perceptual speed, and secondly that DSST performance is the best mediator of the age-related differences observed in episodic memory.

5. Discussion

Note: DSST = Digit Symbol Substitution Test; % ARV explained = the percentage of the age-related variance explained by the mediating variables. ** p < .01. *** p < .001.

The aim of the study was to assess the age-related differences in episodic memory in the light of two important explanatory hypotheses of cognitive aging: the executive hypothesis (West, 1996), and the processing-slowing hypothesis (Salthouse, 1996). In this study, these two hypotheses were tested with the particularity of separating the DSST from the other processing speed measures. Most studies have in fact indicated that processing speed is the main mediator of age-related decline in episodic memory when evaluated using a coding task (Bryan & Luszcz, 1996; Clarys et al., 2007; Crawford et al., 2000; Parkin & Java, 2000). A number of authors (Parkin & Java, 2000; Piccinin & Rabbitt, 1999; Salthouse, 1992) have suggested that the coding task measures a combination of different factors, including perceptual speed and executive functions, which would explain its importance as a mediator of the age-related deficit in memory. The aim of this study was to examine the assumption that executive functions are a better mediator of age-related memory de-

244

A. Baudouin et al. / Brain and Cognition 71 (2009) 240–245

cline than processing speed, especially when processing speed is assessed by perceptual speed tasks. Our findings show that the age-related variance observed in free-recall performance was attenuated more after controlling for executive functions than perceptual processing speed, although the latter did emerge as a moderate mediator of the observed age-related variance. With regard to processing speed, our findings that perceptual speed mediates age-related memory decline support the hypothesis of cognitive slowing with aging (Salthouse, 1996, 2000). Salthouse (1996) suggested that the decrease in processing speed could influence both the quantity and quality of memory performance, because cognitive operations would be executed too slowly to be successfully carried out in the allotted time. However, our findings indicate that executive functioning is a more fundamental mediator of age-related differences in memory than perceptual speed. This outcome is consistent with the view that the age-related decline in memory is dependent on the integrity of the prefrontal brain regions, or at least with the hypothesis of an age-related executive dysfunctioning (Braver & West, 2008). In order to perform episodic memory tasks efficiently, such as the free-recall task used in our study, the individual has to self-initiate strategies, which requires executive control, including planning, implementing strategies, and monitoring performance, in both the encoding and retrieval phases. A large number of studies have shown that age-related differences in strategic encoding and retrieval depend on the executive function level (Bunce, 2003; Taconnat, Clarys, Vanneste, Bouazzaoui, & Isingrini, 2007; Taconnat et al., 2006). An age-related executive deficit would thus reduce the efficiency of these memory processes, affecting episodic memory performance. Overall, our findings can be interpreted as showing that both perceptual speed and executive functioning are required in memory functioning. A measure combining these two aspects of cognitive functioning would thus be a better mediator of the age-related decline in episodic memory and could account for the age-related variance in memory mediated by the DSST (Parkin & Java, 2000) which is assumed to combine these two processes. To investigate this hypothesis, we examined the structure of the DSST and its mediating role in age-related memory deficit. Partial correlations confirmed that it does indeed involve both executive functioning and perceptual speed, and that these processes are related to it equally. More precisely, this pattern was observed in the older adult group, whereas only the processing speed measure was significantly associated with performance on the DSST in the young group. This different pattern of results suggests that the younger adults’ performance varied essentially due to differences in performance speed rather than in the strategic aspect of the task. In contrast, the performance of the older adults involved both executive function and speed. Overall, these findings are therefore consistent with assumptions and investigations suggesting that the DSST involves not only processing speed but also processing resources linked to executive control, such as working memory capacity, and general cognitive abilities (Parkin & Java, 2000; Piccinin & Rabbitt, 1999; Salthouse, 1992). As expected, the results also indicated that the relationship between memory and DSST performance was stronger than the relationship between memory and perceptual processing speed or between memory and executive functions. In line with previous studies, the age-related variance in memory clearly attenuated after statistically controlling for the DSST (Clarys et al., 2007; Crawford et al., 2000; Parkin & Java, 2000). This suggests that the aspect of the DSST which mediates memory decline is not only the speed of carrying out mental operations, but also the executive processes involved. These results call into question the interpretation of certain findings which have suggested that processing speed is the main mediator of the age-related decline in cognition, as evaluated by

a coding task. For example, recent studies investigating age-related modifications in recollective experience (Bugaiska et al., 2007; Clarys, Isingrini, & Gana, 2002) have examined the relationship between explanatory hypotheses of cognitive aging and the Remember/Know paradigm (Gardiner, 2001; Gardiner & RicharsonKlavehn, 2000; Tulving, 1985). Clarys et al.’s (2002) findings confirmed an age-related decrease in R responses and showed that processing speed accounted for most of this decrease when evaluated using the DSST. When executive functions and simple speed measures were tested concurrently, executive functions emerged as the best predictor of age-related changes in remembering (Bugaiska et al., 2007). Bugaiska et al. (2007) suggested that remembering (R responses) required more control processes than ‘‘knowing” (K responses) and involved processing resources such as executive functions. The idea that processing speed and executive functions are two important mediators of the age-related decline in memory is consistent with the model proposed by Anderson and Craik (2000), which combines several explanatory factors. In this model, age-related memory deficits are mediated by a cascade of neurological and cognitive modifications. Two cognitive changes are proposed: a reduction in attentional/executive resources, and a reduction in processing speed. Following this approach, a recent study has explored the role of brain regions and cognitive measures, executive functions and processing speed in age-related differences in episodic memory (Head et al., 2008), confirming the relevance of this type of multivariate approach in the study of the age-related decline in episodic memory. The assumption that a measure combining speed and executive processes is a reliable predictor of memory, as seems to be the case for the DSST, thus appears relevant for investigating age-related memory decline. However, our results indicate that the DSST accounts for more of the age-related differences than speed and executive measures together. This unexplained variance raises the question of whether the coding task involves only executive and speed processes. In our study, after controlling for the number of years of education, executive functioning and perceptual speed together accounted for 83% of the age-related variance in DSST performance, age continuing to be a significant contributor to the variance. Several studies have proposed age-related memory limitations to explain age-related variance in the coding task. They suggest that older adults are less efficient at learning or remembering and need more time to find the symbol in the code table. However, this hypothesis does not always provide a complete explanation of DSST variance (Beres & Baron, 1981; Erber, Botwinick, & Storandt, 1981; Salthouse, 1988), and it would be interesting to add a memory test to the executive and speed measures. This type of component could explain why DSST scores account for nearly all of the age-related variance observed in episodic memory. Although our data are not able to address this issue, we can observe that when the episodic memory measure was used to predict age-related variance in DSST scores, it added a significant age-related variance explanation to executive functions and processing speed mediation (p < .001). After controlling for the number of years of education, executive functioning, perceptual speed and episodic memory together accounted for 90.5% of the age-related variance in DSST performance, age again contributing significantly to the variance. These preliminary additional results suggest the significant involvement of mnesic processes in the DSST. Finally, these findings suggest that the DSST should be used with caution, and highlight the fact that it is not a pure measure of processing speed. In sum, our findings extend the investigation of the age-related differences in episodic memory. They confirm previous results indicating the reliability of the DSST to explain age-related differ-

A. Baudouin et al. / Brain and Cognition 71 (2009) 240–245

ences in memory. They also indicate that this coding task involves several processes, namely executive functions and perceptual speed. Comparison of executive functions and perceptual processing speed to account for the age-related variance in episodic memory shows that executive functioning decline is a fundamental mediator of age-related memory differences. References Anderson, N. D., & Craik, F. I. M. (2000). Memory in the aging brain. In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 411–425). Oxford University Press. Baddeley, A. (1996). Exploring the central executive. The Quarterly Journal of Experimental Psychology, 49A, 5–28. Baddeley, A. (2000). The episodic buffer: A new component of working memory? Trends in Cognitive Sciences, 4(11), 417–423. Bell, S., Franzen, M., & Riley, R. (1999). An analysis of the modified digit symbol test of incidental learning procedures with other measures of memory functioning. Archives of Clinical Neuropsychology, 14(8), 703–704. Beres, C. A., & Baron, A. (1981). Improved digit symbol substitution by older women as a result of extended practice. Journal of Gerontology, 36(5), 591–597. Braver, T. S., & West, R. (2008). Working memory, executive control, and aging. In F. I. M. Craik & T. A. Salthouse (Eds.), The handbook of aging and cognition (3rd ed., pp. 311–372). New York: Psychology Press. Bryan, J., & Luszcz, M. A. (1996). Speed of information processing as a mediator between age and free recall performance. Psychology and Aging, 11, 3–9. Bugaiska, A., Clarys, D., Jarry, C., Taconnat, L., Tapia, G., Vanneste, S., et al. (2007). The effect of aging in recollective experience. The processing speed and executive functioning hypothesis. Consciousness and Cognition, 16(4), 797–808. Bunce, D. (2003). Cognitive support at encoding attenuates age differences in recollective experience among adults of lower frontal lobe function. Neuropsychology, 17(3), 353–361. Clarys, D., Isingrini, M., & Gana, K. (2002). Aging and episodic memory: Mediators of age differences in remembering and knowing. Acta Psychologica, 109(3), 315–329. Clarys, D., Souchay, C., Baudouin, A., Fay, S., Vanneste, S., Taconnat, L., et al. (2007). Contribution des fonctions exécutives et de la vitesse de traitement au vieillissement de la mémoire épisodique. L’année psychologique, 1, 15–38. Craik, F. I. M., & McDowd, J. M. (1987). Age differences in recall and recognition. Journal of Experimental Psychology: Learning, Memory and Cognition, 13, 474–479. 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. Daigneault, S., Braun, C. M. J., & Whitaker, H. A. (1992). Early effects of normal aging on perseverative and non-perseverative prefrontal measures. Developmental Neuropsychology, 8, 99–114. Doiseau, F., & Isingrini, M. (2005). Updating information in verbal working memory and executive functioning. Psychological Reports, 96, 67–76. Erber, J. T., Botwinick, J., & Storandt, M. (1981). The impact of memory on age differences in digit symbol performance. Journal of Gerontology, 36(5), 586–590. Folstein, M. F., Folstein, S. F., & McHugh, P. R. (1975). Mini-mental state: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189–198. Gardiner, J. M. (2001). Episodic memory and autonoetic consciousness: A first person approach. The Royal Society, 356, 1351–1361. Gardiner, J. M., & Richarson-Klavehn, A. (2000). Remembering and knowing. In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory. Oxford University Press. Glisky, E. L., Polster, M. R., & Routhieaux, B. C. (1995). Double dissociation between item and source memory. Neuropsychology, 9, 229–235. Glisky, E. L., Rubin, S. R., & Davidson, S. R. (2001). Source memory in older adults: An encoding or retrieval problem? Journal of Experimental Psychology: Learning Memory and Cognition, 27, 1131–1146. Head, D., Rodrigue, K. M., Kennedy, K. M., & Raz, N. (2008). Neuroanatomical and cognitive mediators of age-related differences in episodic memory. Neuropsychology, 22, 491–507. Isingrini, M., & Vazou, F. (1997). Relation between fluid and frontal lobe functioning in older adults. International Journal of Aging and Human Development, 45, 99–109.

245

Lindenberger, U., Mayr, U., & Kliegl, R. (1993). Speed and intelligence in old age. Psychology and Aging, 8, 207–220. Miyake, A., Friedman, N., Emerson, M., Witzki, A., Howerter, A., & Wager, T. (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. Nelson, H. E. (1976). A modified card sorting test sensitive to frontal lobe deficits. Cortex, 12, 313–324. Park, D. C., Smith, A. D., Lautenschlager, G., Earles, J. L., Frieske, D., Zwahr, M., et al. (1996). Mediators of long-term memory performance across the life span. Psychology and Aging, 11, 621–637. Parkin, A. J. (1997). Normal age-related memory loss and its relation to frontal lobe dysfunction. In P. Rabbitt (Ed.), Methodology of frontal and executive function (pp. 177–190). Hove: Psychology Press. Parkin, A. J., & Java, R. I. (2000). Determinants of age-related memory loss. In T. J. Perfect & E. A. Maylor (Eds.), Models of cognitive aging. Oxford University Press. Parkin, A. J., Walter, B. M., & Hunkin, N. M. (1995). Relationships between normal aging, frontal lobe function, and memory for temporal and spatial information. Neuropsychology, 9, 304–312. Perfect, T. J. (1997). Memory aging as frontal lobe dysfunction. In M. A. Conway (Ed.), Cognitive models of memory (pp. 315–340). Hove, UK: Psychology Press. Perrotin, A., Isingrini, M., Souchay, C., Clarys, D., & Taconnat, L. (2006). Episodic feeling-of-knowing accuracy and cued recall in the elderly: Evidence for double dissociation involving executive functioning and processing speed. Acta Psychologica, 122(1), 58–73. Piccinin, A. M., & Rabbitt, P. M. A. (1999). Contribution of cognitive abilities to performance and improvement on a substitution coding task. Psychology and Aging, 14(4), 539–551. Rabbitt, P. (1997). Introduction: Methodologies and models in the study of executive function. In P. Rabbitt (Ed.), Methodology of frontal and executive function. UK: Psychology Press. Raven, J. C. (1982). Revised manual for raven’s progressive matrices and vocabulary scales. Windser, UK: NFER-Nelson. Raz, N., & Rodrigue, K. R. (2006). Differential aging of the brain: Patterns, cognitive correlated and modifiers. Neuroscience and Biobehavioural Reviews, 30, 730–748. Rhodes, M. G. (2004). Age-related differences in performance on the Wisconsin Card Sorting Test: A meta-analytic review. Psychology and Aging, 19(3), 482–494. Salthouse, T. A. (1988). The role of processing resources in cognitive aging. In M. L. Howe & C. J. Brainerd (Eds.), Cognitive development in adulthood (pp. 185–239). New York: Springer-Verlag. Salthouse, T. A. (1992). What do adult age differences in the digit symbol substitution test reflect? Journal of Gerontology: Psychological Sciences, 47(3), 121–128. Salthouse, T. A. (1993). Speed mediation of adult age differences in cognition. Developmental Psychology, 29, 722–738. Salthouse, T. A. (1994). The nature of the influence of speed on adult age differences in cognition. Developmental Psychology, 30, 240–259. Salthouse, T. A. (1996). The processing speed theory of adult age differences in cognition. Psychological Review, 3, 403–428. Salthouse, T. A. (2000). Aging and measure of processing speed. Biological Psychology, 54, 35–54. Salthouse, T. A., & Babcock, R. L. (1991). Decomposing adult age differences in working memory. Developmental Psychology, 27, 763–776. Slivinski, M., & Buschke, H. (1997). Processing speed and memory in aging and dementia. Journal of Gerontology: Psychological Sciences, 52B, 308–318. Taconnat, L., Baudouin, A., Fay, S., Vanneste, S., Clarys, D., Tournelle, L., et al. (2006). Aging and implementation of encoding strategies: The role of executive functions. Neuropsychology, 20, 658–665. Taconnat, L., Clarys, D., Vanneste, S., Bouazzaoui, B., & Isingrini, M. (2007). Aging and strategic retrieval in a cued-recall test: The role of executive functions and fluid intelligence. Brain & Cognition, 64, 1–6. Tulving, E. (1985). Memory and consciousness. Canadian Psychologist, 26, 1–12. Wechsler, D. (1981). Wechsler adult intelligence scale-revised. New York: Psychological Corporation. Wechsler, D. (1987). Wechsler memory scale-revised. San Antonio, CA: Psychological Corporation. West, R. L. (1996). An application of prefrontal cortex function theory to cognitive aging. Psychological Bulletin, 120, 272–292. Zigmond, A. S., & Snaith, R. P. (1983). The hospital anxiety and depression scale. Acta Psychatrica Scandinavica, 67, 361–370.