Abstract

This study set out to establish the relationship between changes in episodic memory retrieval in normal aging on the one hand and gray matter volume and 18FDG uptake on the other. Structural MRI, resting-state 18FDG-PET, and an episodic memory task manipulating the depth of encoding and the retention interval were administered to 46 healthy subjects divided into three groups according to their age (young, middle-aged, and elderly adults). Memory decline was found not to be linear in the course of normal aging: Whatever the retention interval, the retrieval of shallowly encoded words was impaired in both the middle-aged and the elderly, whereas the retrieval of deeply encoded words only declined in the elderly. In middle-aged and elderly subjects, the reduced performance in the shallow encoding condition was mainly related to posterior mediotemporal volume and metabolism. By contrast, the impaired retrieval of deeply encoded words in the elderly group was mainly related to frontal and parietal regions, suggesting the adoption of inefficient strategic processes.

INTRODUCTION

Although the majority of cross-sectional studies have highlighted a linear decline in episodic memory with age, the remainder have reported less clear-cut results, as indeed have most longitudinal studies, revealing relatively little or no modification before middle age, but a significant decline in the elderly (Grady, Springer, Hongwanishkul, McIntosh, & Winocur, 2006; Rönnlund, Nyberg, Bäckman, & Nilsson, 2005; Salthouse, 2003; Salthouse, Atkinson, & Berish, 2003; Verhaeghen & Salthouse, 1997; but see also Park et al., 2002). These discrepancies may be due to the fact that different tasks draw on different memory processes. The effects of age on memory performance are determined by both the level of information processing at the encoding phase and the degree of environmental support during the retrieval phase. Given the age-related diminution of cognitive resources, sizeable memory decline may be expected in tasks requiring self-initiated processes that afford little environmental support for retrieval, such as free recall (Taconnat, Clarys, Vanneste, Bouazzaoui, & Isingrini, 2007; Whiting & Smith, 1997; Craik & McDowd, 1987; Craik & Byrd, 1982). Memory decline may also be linked to the inefficient initiation of deep processing during encoding. However, if deep encoding is induced (e.g., by inducing a semantic processing of the items; see Fay, Isingrini, Ragot, & Pouthas, 2005; Allan, Robb, & Rugg, 2000; Craik & Tulving, 1975; Craik & Lockhart, 1972, p. 675), one can observe an improvement in recall performance, particularly in elderly subjects who are then capable of achieving the same retrieval levels as younger subjects (Taconnat & Isingrini, 2004; Isingrini & Taconnat, 1997; Craik & Jennings, 1992).

Few neuroimaging studies have investigated the neural substrates of episodic memory taking into account the depth of encoding levels, particularly in the elderly. Generally speaking, young adults display greater activation in frontal areas for deeply encoded material both during encoding (Otten, Henson, & Rugg, 2002; Kapur et al., 1994) and retrieval (Rugg et al., 1998; Rugg, Fletcher, Frith, Frackowiak, & Dolan, 1997), as well as in medial-temporal regions during encoding (Mandzia, Black, McAndrews, Grady, & Graham, 2004; Otten et al., 2002) and retrieval (Mandzia et al., 2004; Rugg et al., 1997, 1998; Nyberg, McIntosh, Houle, Nilsson, & Tulving, 1996). More recently, other authors have placed the emphasis on the involvement of the parietal cortex in the effect of the levels of processing on episodic memory retrieval (Iidaka, Matsumoto, Nogawa, Yamamoto, & Sadato, 2006; Shannon & Buckner, 2004).

In their comparison of young and older subjects, Logan, Sanders, Snyder, Morris, and Buckner (2002) found left frontal underrecruitment in the older subjects when they had to intentionally memorize items (without any specific strategy). In contrast, when subjects had to deeply encode items using a strategy provided by the experimenters, no difference was found in left frontal activity, suggesting that underrecruitment can be reversed under more supportive encoding conditions. Moreover, these authors reported additional, nonselective frontal recruitment in the elderly group, notably in the deep encoding condition, which, according to them, reflected a breakdown in the appropriate selection of regions (Rosen et al., 2002; Grady, McIntosh, Rajah, Beig, & Craik, 1999 for similar results, but see also Stebbins et al., 2002). Age-related changes in the recruitment of frontal areas in episodic memory tasks have opened up a debate about its neurofunctional significance. Generally, in young subjects, the left frontal areas preferentially support encoding processes in episodic memory (as well as retrieval processes in semantic memory), whereas the right frontal areas support the retrieval of events (see the HERA model, Hemispheric Encoding Retrieval Asymmetry; Habib, Nyberg, & Tulving, 2003; Tulving, Kapur, Craik, Moscovitch, & Houle, 1994). Older adults, however, often recruit the frontal areas bilaterally, both in encoding and retrieval. This pattern of findings has been conceptualized in the HAROLD model (Hemispheric Asymmetry Reduction in older adults; Cabeza, 2002). This additional frontal activity may reflect compensatory processes and/or dedifferentiation (Cabeza, Anderson, Locantore, & McIntosh, 2002; Dolcos, Rice, & Cabeza, 2002), but Rajah and D'Esposito's (2005) meta-analysis revealed more subtle age effects in the frontal lobes, where changes in the prefrontal cortex, including either under- or overrecruitment, appeared to be region-specific (ventral, dorsal, or anterior). Other studies have revealed additional compensatory activity in the parahippocampal gyrus during successful retrieval in older adults (van der Veen, Nijhuis, Tisserand, Backes, & Jolles, 2006; Cabeza et al., 2004). Thus, functional age-related changes do occur, sometimes accompanied by differences in cognitive behavior, with regions being activated to a greater or lesser degree according to the task. This has led to the idea that age-related changes are more functionally than structurally based (Cabeza, Anderson, et al., 2002; Logan et al., 2002). In the same vein, Grady et al. (2006, see also Beason-Held, Kraut, & Resnick, 2008) believe that age-related reduced deactivation in the medial frontal, cingulate, and precuneus regions may be due to disruption of the default-mode network activity rather than to structural changes (these medial regions are activated in the resting state and display deactivation when specific cognitive processes are engaged for a given task; Greicius, Krasnow, Reiss, & Menon, 2003; Raichle et al., 2001).

However, structural deterioration does occur in normal aging, and numerous studies have looked for relationships between medial-temporal and frontal volume and memory performance (see Van Petten, 2004 for a review; Mungas et al., 2005). Van Petten (2004) and Van Petten et al.'s (2004) reviews reported findings from several studies of healthy volunteers. They showed that less than 10% of the correlations between hippocampal volume and memory scores that had been assessed proved to be significant and positive (mostly in elderly adults), and, more surprisingly, none of the correlations assessed between frontal volume and memory performance turned out to be significant, although age-related structural changes have consistently been found in this area (Kalpouzos et al., in press). Recently, Brickman, Habeck, Zarahn, Flynn, and Stern (2007) examined whether age-associated covariance topographies of gray and white matter density were reflected by age-related cognitive impairment. They found that declining gray matter (GM) networks involving the frontal, temporal, parietal, and cingulate cortices; the thalamus; and the posterior parahippocampal region were associated with free recall impairment. Thus, most studies have failed to show consistent hippocampal or frontal structure–memory relationships. Several reasons can be adduced. First, few studies have assessed middle age: Although some researchers have found age-related GM decline to be linear, particularly in the frontal lobes, the decline of other structures, such as the hippocampus, appears not to be (Allen, Bruss, Brown, & Damasio, 2005). Second, the insufficient sensitivity and specificity of the memory tests administered may have lessened the probability of highlighting structure–memory relationships, as several studies have used routine clinical tasks and composite scores which preclude the assessment of specific processes. Moreover, the hippocampus and the prefrontal cortex are thought to be involved in a number of different memory processes (Zimmerman et al., 2006; Schacter, Savage, Alpert, Rauch, & Albert, 1996), so a composite score could not be specifically correlated with one or the other structure. Also, different parts of these structures are involved in different memory processes, according to the HERA (Habib et al., 2003; Tulving et al., 1994) and HIPER (Hippocampal Encoding/Retrieval; Lepage, Habib, & Tulving, 1998) models. In effect, according to the latter, the anterior and posterior parts of the hippocampus are primarily involved in encoding and retrieval, respectively. The third methodological issue concerns the measurement of hippocampal and frontal volume: Most studies have used the region-of-interest (ROI) method, assessing the entire volume of the hippocampal formation and/or frontal lobe. However, as shown in voxel-based neuroimaging studies, age-related shrinkage is sometimes limited to subregions of the hippocampus/parahippocampal cortex (Kalpouzos et al., in press) and the frontal lobe (Grieve, Clark, Williams, Peduto, & Gordon, 2005). Moreover, the ROI procedure does not provide any opportunity to study other brain areas, even though these, too, may contribute to memory weakening.

In the current experiment, young, middle-aged, and elderly adults underwent an episodic memory task specifically designed to assess retrieval abilities following shallow and deep encoding. Our task also featured a retention interval factor (immediate vs. 1-week delayed recall). Little is known about the relationships between delayed recall and age effects and their neural substrates. In most of cognitive studies using intervals of retention shorter than an hour, authors did not find differences in the rate of forgetting across age. However, using longer delays, Tombaugh and Hubley (2001) showed that increasing age was associated with faster rates of forgetting of verbal material after one day of retention, but not beyond. Fjell et al. (2005), using visual material, disconfirmed the hypothesis that older subjects had a steeper rate of forgetting than younger. In neuroimaging studies, the retention interval is usually short, from seconds to an hour. Studying longer retention intervals is, nonetheless, relevant because consolidation in episodic memory takes place over several days. Such longer delays have been assessed in few studies by examining the correlations between brain volumes and performance. Despite the large variability in the retention intervals used across these studies (from an hour to 11 weeks), a relationship was found between the hippocampal size and the most delayed retention intervals, suggesting the importance of hippocampal integrity for successful delayed retrieval. For instance, Walhovd et al. (2004) assessed the correlations between hippocampal volume and recall scores obtained at three different retention intervals (5 min, 30 min, 11 weeks) in subjects aged 20 to 88 years. The results showed that both hippocampal volume and age predicted recall after weeks (see also Fjell et al., 2005; Golomb et al., 1994). Using the same procedure in healthy subjects spanning all the adulthood, Sullivan, Marsh, Mathalon, Lim, and Pfefferbaum (1995) found a relationship between hippocampal volume and delayed recall performance with nonverbal material, but not with the immediate scores, whereas the correlation between hippocampal volume and immediate verbal recall performance was significant (but not with the delayed scores). Globally, a relationship has been found between hippocampal volume and delayed memory performance, but the data remain uncertain. Uncapher and Rugg (2005) studied the neural substrates of encoding in episodic memory. Delayed recollection (2 days after the encoding phase) was associated with enhanced activity in the bilateral ventral inferior frontal gyrus at the time of encoding, whereas recollection 30 min after the encoding phase was associated with greater fusiform activity. Thus, little is known regarding the regions involved in episodic memory in function of the delay of recall; moreover, to our knowledge, no study has investigated yet how this variable may interact with the depth of encoding, that is, whether deep encoding is still beneficial with a longer retention interval, particularly in aging.

Our aim was to highlight age-related episodic memory impairment or preservation, and the structural and metabolic neural substrates responsible for cognitive decline or preservation in middle-aged and elderly adults. The introduction of a middle-aged group should help to better delineate the onset of episodic memory impairment in the course of normal aging, as studies of this age group are seldom, in contrast to direct comparisons between groups of young and old subjects. More precisely, we hypothesized different age effects in function of the depth of processing, such as a deep encoding would induce enhanced performance in the older groups, allowing thus to study the neural correlates of not only episodic memory impairment but also preserved processes. A secondary aim of this study was to test the hypothesis of an interaction effect between the retention interval and the depth of encoding, that is, to examine whether a deep processing at encoding remains as beneficial after a 1-week delay than in immediate recall. To deal with the abovementioned methodological issues, we adopted a voxel-based approach, using SPM5 software (www.fil.ion.ucl.ac.uk/spm/) to process structural MRI and (18F)2-Fluoro-2-Deoxy-d-Glucose–Positron Emission Tomography (18FDG-PET) data.

METHODS

Subjects

Forty-six healthy, right-handed subjects, all living at home, were enrolled in this study, in accordance with stringent prospective inclusion/exclusion criteria. They were assigned to three different groups according to their age (14 young adults aged 28.3 ± 6.1 years, 13 middle-aged ones aged 51.2 ± 6.6 years, and 19 elderly ones aged 67 ± 6.8 years), and all had a minimum educational background of 8 years of schooling. Detailed characteristics are given in Table 1 

Table 1

Population Characteristics and Memory Performance


Young Group
Middle-aged Group
Elderly Group
Number 14 13 19 
Women/men 6/8 10/3 9/10 
Age: mean ± SD (range) 28.3 ± 6.1 (20–38) 51.2 ± 6.6 (40–58) 67 ± 6.8 (60–83) 
Educational background*: mean ± SD (range) 13.4 ± 1.3 (12–15) 13.2 ± 2.3 (9–15) 11.5 ± 2.6 (8–17) 
Episodic memory scores: mean ± SD 
 Immediate recall–Shallow encoding 7.1 ± 1.8 5.5 ± 2.1 4.9 ± 2 
 Immediate recall–Deep encoding 9.9 ± 2.3 10.8 ± 2.3 8.5 ± 2.3 
 Delayed recall–Shallow encoding 2.4 ± 1.9 1.1 ± 1.2 0.9 ± 1.1 
 Delayed recall–Deep encoding 4.4 ± 3.2 3.5 ± 3 1.6 ± 1.6 

Young Group
Middle-aged Group
Elderly Group
Number 14 13 19 
Women/men 6/8 10/3 9/10 
Age: mean ± SD (range) 28.3 ± 6.1 (20–38) 51.2 ± 6.6 (40–58) 67 ± 6.8 (60–83) 
Educational background*: mean ± SD (range) 13.4 ± 1.3 (12–15) 13.2 ± 2.3 (9–15) 11.5 ± 2.6 (8–17) 
Episodic memory scores: mean ± SD 
 Immediate recall–Shallow encoding 7.1 ± 1.8 5.5 ± 2.1 4.9 ± 2 
 Immediate recall–Deep encoding 9.9 ± 2.3 10.8 ± 2.3 8.5 ± 2.3 
 Delayed recall–Shallow encoding 2.4 ± 1.9 1.1 ± 1.2 0.9 ± 1.1 
 Delayed recall–Deep encoding 4.4 ± 3.2 3.5 ± 3 1.6 ± 1.6 

SD = standard deviation.

*

Number of years of schooling. Difference between young and middle-aged groups, and between middle-aged and elderly groups were not significant (p > .05).

. A semistructured questionnaire was used by senior clinical neurologists to screen the subjects' health. They were not enrolled in these studies if there were any abnormalities in their clinical, MRI, and neuropsychological examinations. Inclusion/exclusion criteria are detailed in Kalpouzos et al.'s (in press) study. The Mattis Dementia Rating Scale (Mattis, 1976) was used for subjects older than 50 years to exclude those suspected of neurodegenerative pathology (mean ± SD = 141.4 ± 2.8). In no subject was there any evidence of significant cognitive decline beyond that expected for normal aging, and no subject complained of memory loss. All subjects underwent the structural MRI acquisition, the resting-state 18FDG-PET acquisition, and the cognitive assessment on different days within a maximum period of 2 months.

This protocol was approved by the regional ethics committee. Subjects gave their written consent to the procedure prior to the investigation.

Episodic Memory Assessment

We used the “Encoding, Storage, Retrieval” (ESR) paradigm devised by Eustache, Desgranges, and Lalevée (1998), which features two learning phases (shallow and incidental, deep and intentional) with two different word lists. This task is detailed elsewhere (Chételat et al., 2003; Eustache et al., 1998). Briefly, each list comprised 16 words, belonging to 16 different semantic categories. For the first list, subjects had to say whether the first and last letters of each orally presented word were in alphabetical order, and did not receive any instructions to memorize them. At the end of this incidental shallow encoding phase, free recall and visual recognition tasks were carried out. Immediately afterward, subjects were asked to memorize the words in the second list. In order to induce semantic processing, they had to generate orally a sentence that defined or described the orally presented word. Every two words, an immediate cued recall task was performed using a semantic category cue, in order to ensure that encoding had taken place and to reinforce its semantic nature. At the end of this 16-word intentional deep encoding phase, free recall and visual recognition tasks were carried out again. Delayed free recall and recognition tasks were also performed 1 week later. Thus, the encoding differed not only in the perceptual/semantic instructions but also in the incidental/intentional instructions. This made it possible to reinforce the depth-of-processing effects as the effects of automatic and externally driven or self-directed and controlled memory processes (i.e., nonstrategic vs. strategic encoding effects) were added to the perceptual and semantic processing, respectively. Only the free recall scores were analyzed because of ceiling effects in the recognition tasks. Two variables were taken into account: the retention interval (immediate free recall vs. delayed recall) and the nature of the encoding (shallow and incidental encoding vs. deep and intentional encoding).

MRI Data Acquisition and Processing

The T1-weighted volume MRI scan consisted of a set of 128 adjacent axial slices parallel to the AC–PC line, with a slice thickness of 1.5 mm and a voxel size of 0.94 × 0.94 mm, using the spoiled gradient-echo sequence (repetition time [TR] = 10.3 msec; echo time [TE] = 2.1 msec; field of view [FOV] = 24 × 18 cm; matrix = 256 × 192; 1.5-T Signa Advantage echospeed; General Electric). The standard correction for field inhomogeneities was applied.

The MRI data were analyzed using SPM5 and the automated VBM procedure, fully described elsewhere (Good et al., 2001). Briefly, this procedure consisted of the segmentation and normalization of the MRI images using the Montreal Neurological Institute (MNI) priors. The resulting optimized parameters were then applied to the original scans, which were segmented, modulated (to integrate volume changes), and smoothed with a 12-mm isotropic Gaussian kernel.

PET Data Processing

Data were collected using a high-resolution ECAT Exact HR+ PET scanner with an isotropic resolution of 4.6 × 4.2 × 4.2 mm (axial FOV = 158 mm). The subject's head was positioned on a headrest according to the cantho-meatal line and gently restrained with straps. 18FDG uptake was measured in the resting condition, with eyes closed, in a quiet, dark environment. A catheter was inserted into a vein in the arm to inject the radiotracer. Following 68Ga transmission scans, 3–5 mCi of 18FDG were injected as a bolus at Time 0, and a 10-min PET data acquisition began 50 min postinjection. Sixty-three planes were acquired with septa out (volume acquisition), using a voxel size of 2.2 × 2.2 × 2.43 mm (x, y, z). PET data were corrected for partial volume effect, using an optimized voxel-based technique fully implemented in partial volume effect lab software (see Kalpouzos et al., in press). Briefly, this technique makes it possible to correct for both the loss of GM activity due to spill-out onto non-GM tissues, and increases in GM activity, due to spill-in from adjacent white matter. In order to obtain accurate PET data, the images were subjected to coregistration onto their corresponding MR images and to spatial normalization on the MNI template. PET datasets were resliced (2 × 2 × 2 mm) and smoothed with a 14-mm isotropic Gaussian filter to blur individual variations in gyral anatomy and to increase the signal-to-noise ratio.

Statistical Analyses

Statistical analyses were processed in two steps. The first step consisted in the analysis of the behavioral data. Using Statistica software (version 7), a three-way repeated measures ANOVA was performed with the group as a between-subjects factor with three modalities (young, middle-aged, elderly), the retention interval as a within-subjects factor with two modalities (immediate, delayed), and the depth of encoding as a within-subjects factor with two modalities (incidental and shallow, intentional and deep).

The second step consisted in the analyses of the neuroimaging data. The strategy was to highlight the metabolic and structural brain correlates of the significant behavioral interaction effects. Therefore, the neuroimaging analyses were dependent on the behavioral findings. The general principle of highlighting neural substrates of cognitive processes is based on the correlational approach, which consisted in assessing correlations between brain data and cognitive scores. This method, which validity has been extensively corroborated in previous publications, allows to highlight the brain areas whose metabolic dysfunction or structural deterioration underlie cognitive deficiency (see, for instance, Rauchs et al., 2007; Chételat et al., 2003). As we hypothesized interaction effects between the three factors that have been manipulated in the present study (age, delay of recall, and depth of encoding), we used a design of SPM5 allowing to perform correlations between brain data and performance involving the factors of interest highlighted by the significant behavioral interactions (i.e., among one, two, or three groups), taking into account one or both memory factors (delay, depth). These analyses were performed separately for structural MRI and PET data using the flexible factorial design of SPM5, with groups as conditions (Condition 1: young; Condition 2: middle-aged; Condition 3: elderly), and recall scores as covariates (Covariate 1: immediate free recall following shallow encoding; Covariate 2: immediate free recall following deep encoding; Covariate 3: delayed free recall following shallow encoding; Covariate 4: delayed free recall following deep encoding). We then used the interaction option of this design between conditions and covariates to assess the significant interaction effects highlighted in the behavioral ANOVA. According to the behavioral results, the specific neuroimaging analyses and their interpretations are presented in the Results section.

The value of each voxel was divided by the mean value of all the voxels, in order to control for individual variations in total GM volume (MRI data) and for individual variations in total 18FDG uptake (PET data). Both MRI and PET data were masked so that only GM voxels of interest were considered in further analyses. This mask, implemented in the design via the explicit mask option of SPM5, was obtained by thresholding the MNI GM prior to an absolute value >0.35, corresponding to more than a 35% probability of belonging to GM (this GM threshold was determined as the best compromise for avoiding the inclusion of white matter voxels without excluding GM voxels of interest). The results are displayed at a statistical threshold of p < .005, uncorrected.

RESULTS

Behavioral Findings

The ANOVA revealed (1) a main effect of group [F(2, 43) = 6.8, p < .003], where the young group did not perform significantly better than the middle-aged one (p > .05), but the latter significantly outperformed the elderly one (p = .03); (2) a main effect of retention interval [F(1, 43) = 394.4, p < .001], where performance was better on immediate recall than on delayed recall; (3) a main effect of depth of encoding [F(1, 43) = 118.2, p < .001], with subjects obtaining better recall scores in the deep encoding condition than in the shallow one; (4) a significant interaction effect between the group and depth of encoding factors [F(2, 43) = 4.5, p < .02]; and (5) a significant interaction effect between the retention interval and depth of encoding factors [F(1, 43) = 28.8, p < .001]. Neither the Group × Retention interval interaction, nor the interaction between all three factors was significant [F < 1; F(2, 43) = 2.9, p = .068, respectively]. On the basis of the significant interaction between group and depth of encoding, planned comparisons were performed in order to identify specific effects (see Figure 1A,
Figure 1

Episodic memory retrieval performance. Interaction effects between (A) depth of encoding and age, and (B) depth of encoding and delay of recall.

Figure 1

Episodic memory retrieval performance. Interaction effects between (A) depth of encoding and age, and (B) depth of encoding and delay of recall.

). In the shallow encoding condition, the middle-aged subjects showed a significant decrease in the number of recalled words compared with the young group [F(1, 43) = 7.3, p < .01], as did the elderly group compared with the young one [F(1, 43) = 13.3, p < .01], whereas the decrease between the middle-aged and elderly subjects was not significant (F < 1). Conversely, in the deep encoding condition, there was no significant difference between the young and middle-aged groups (F < 1), whereas the elderly subjects displayed a significant decrease in performance compared with the middle-aged ones [F(1, 43) = 8.2, p < .007]. However, the elderly adults' recall performance significantly improved between the shallow and the deep encoding conditions [F(1, 43) = 29.5, p < .001]. Regarding the other significant interaction (Retention interval × Depth of encoding), in both immediate and delayed recall, there were significant differences between the shallow and deep encoding conditions [F(1, 43) = 122.2 and F(1, 43) = 31.1, p < .001, respectively], the difference being greater for immediate recall than for delayed recall (see Figure 1B).

Neural Substrates of Episodic Memory Retrieval

On the basis of the significant interaction effects obtained in the cognitive analysis above, we specifically addressed these interactions in relation to the neuroimaging data (structural MRI and 18FDG-PET), in order to highlight those areas that could mediate age-related memory decrease or preservation. We examined the structural and metabolic correlates of (1) the significant interaction between age and depth of encoding, and (2) the interaction between depth of encoding and retention interval. The main findings are summarized in Table 2 

Table 2

Summary Table of the Results: Episodic Memory Scores According to Age and their Structural and Metabolic Substrates


Depth of Encoding
Shallow
Deep
Groups 
Young subjects Performance: Young > Middle-aged and Elderly Performance: Young = Middle-aged 
Structural correlates: frontal bilateral cortex, left parietal cortex 
Middle-aged subjects Structural correlates: bilateral posterior mediotemporal areas, bilateral temporal cortex Metabolic correlates: left frontal cortex, right parietal cortex, bilateral anterior and middle mediotemporal areas 
Performance: Middle-aged > Elderly 
Elderly subjects Metabolic correlates: bilateral posterior mediotemporal areas, cerebellum Structural correlates: left frontal cortex 
Metabolic correlates: right frontal cortex, right parietal cortex 
 
Retention Interval 
Immediate recall Structural and metabolic correlates: occipital cortex (fusiform gyrus), cerebellum Structural correlates: frontal cortex 
Metabolic correlates: right frontal cortex, right parieto-temporal cortex 
Delayed recall Structural and metabolic correlates: bilateral posterior mediotemporal areas Structural correlates: right occipital cortex, cerebellum 
Metabolic correlates: right parieto-occipital cortex, SMA 

Depth of Encoding
Shallow
Deep
Groups 
Young subjects Performance: Young > Middle-aged and Elderly Performance: Young = Middle-aged 
Structural correlates: frontal bilateral cortex, left parietal cortex 
Middle-aged subjects Structural correlates: bilateral posterior mediotemporal areas, bilateral temporal cortex Metabolic correlates: left frontal cortex, right parietal cortex, bilateral anterior and middle mediotemporal areas 
Performance: Middle-aged > Elderly 
Elderly subjects Metabolic correlates: bilateral posterior mediotemporal areas, cerebellum Structural correlates: left frontal cortex 
Metabolic correlates: right frontal cortex, right parietal cortex 
 
Retention Interval 
Immediate recall Structural and metabolic correlates: occipital cortex (fusiform gyrus), cerebellum Structural correlates: frontal cortex 
Metabolic correlates: right frontal cortex, right parieto-temporal cortex 
Delayed recall Structural and metabolic correlates: bilateral posterior mediotemporal areas Structural correlates: right occipital cortex, cerebellum 
Metabolic correlates: right parieto-occipital cortex, SMA 

SMA = supplementary motor area.

.

Age and Depth of Encoding Interaction

We performed correlational analyses between neuroimaging data and scores obtained by the three groups in the shallow encoding condition (combining both delays of retention), as the planned comparisons of the interaction between age and depth of encoding factors showed decreased performance both in middle-aged and elderly subjects compared to their younger counterparts in this condition. This interaction also revealed preserved performance in the middle-aged group in the deep encoding modality, but reduced scores in the elderly group. To highlight the neural correlates of these two different age effects, we performed (1) a correlational analysis between neuroimaging data and performance of both young and middle-aged subjects in the deep encoding condition (combining both delays), and (2) between neuroimaging data and performance of both middle-aged and elderly subjects in the deep encoding condition (combining both delays). In case of behavioral age-related difference, we assumed that the highlighted regions mediate cognitive impairment because age is the factor which differentiates the three groups the most (as the subjects of the three groups were as far as possible matched for other factors, i.e., education, same strict including/excluding criteria regarding health).

Correlations between brain data and recall performance following shallow encoding in the young, middle-aged, and elderly groups
This analysis was performed in order to identify the brain regions which could mediate the memory deficits in the shallow encoding condition in both the middle-aged and elderly adults. The main regions whose GM volume and 18FDG uptake were correlated with the recall scores following shallow encoding were the bilateral posterior hippocampal and parahippocampal areas, including the fusiform and lingual gyri. Bilateral temporal GMv and left cerebellar metabolism were also correlated with these scores (Table 3,
Table 3

GMv and FDG Uptake Correlates of Shallow Encoding Condition in Young, Middle-aged, and Elderly Adults

MNI Coordinates
t
k
Labeling (BA)
x
y
z
[Young + Middle-aged + Elderly] Groups: GMv * [Immediate + Delayed] Recalls following Shallow Encoding 
46 −50 12 3.84 60 R Temporal mid (21) 
34 −34 −10 3.71 114 R Parahippocampal, hippocampus (20, 37) 
−38 −32 −12 3.70 122 L Lingual, parahippocampal, hippocampus (20, 37) 
14 −24 −38 3.68 36 R Cerebellum 
−46 −46 18 3.52 97 L Temporal sup, angular (39, 41) 
 
[Young + Middle-aged + Elderly] Groups: FDG Uptake * [Immediate + Delayed] Recalls following Shallow Encoding 
34 −34 −10 4.88 318 R Hippocampus, parahippocampal, fusiform (20, 37) 
−8 −42 −28 4.16 1797 L Cerebellum 
−30 −36 −12 3.58 239 L Hippocampus, parahippocampal, fusiform (20, 37) 
26 −56 −26 3.29 262 R Cerebellum  
−62 −32 3.14 69 L Temporal mid (21) 
26 −16 3.13 54 R Amygdala (34) 
−20 −54 −4 3.04 29 L Lingual (19) 
−26 −2 −14 2.97 65 L Amygdala (34) 
MNI Coordinates
t
k
Labeling (BA)
x
y
z
[Young + Middle-aged + Elderly] Groups: GMv * [Immediate + Delayed] Recalls following Shallow Encoding 
46 −50 12 3.84 60 R Temporal mid (21) 
34 −34 −10 3.71 114 R Parahippocampal, hippocampus (20, 37) 
−38 −32 −12 3.70 122 L Lingual, parahippocampal, hippocampus (20, 37) 
14 −24 −38 3.68 36 R Cerebellum 
−46 −46 18 3.52 97 L Temporal sup, angular (39, 41) 
 
[Young + Middle-aged + Elderly] Groups: FDG Uptake * [Immediate + Delayed] Recalls following Shallow Encoding 
34 −34 −10 4.88 318 R Hippocampus, parahippocampal, fusiform (20, 37) 
−8 −42 −28 4.16 1797 L Cerebellum 
−30 −36 −12 3.58 239 L Hippocampus, parahippocampal, fusiform (20, 37) 
26 −56 −26 3.29 262 R Cerebellum  
−62 −32 3.14 69 L Temporal mid (21) 
26 −16 3.13 54 R Amygdala (34) 
−20 −54 −4 3.04 29 L Lingual (19) 
−26 −2 −14 2.97 65 L Amygdala (34) 

k = cluster size > 20 voxels; BA = Brodmann's area. L = left; R = right; ant = anterior; inf = inferior; med = medial; mid = middle; orb = orbital; SMA = supplementary motor area; sup = superior; tri = triangularis.

, Figure 2A 
Figure 2

GMv and FDG uptake correlates of episodic memory abilities according to depth of encoding conditions. (A) Structural and metabolic correlates of episodic memory recall following shallow encoding in the young, middle-aged, and elderly groups. (B) Structural and metabolic correlates of episodic memory recall following deep encoding in the young and middle-aged groups. (C) Structural and metabolic correlates of episodic memory recall following deep encoding in the middle-aged and elderly groups.

Figure 2

GMv and FDG uptake correlates of episodic memory abilities according to depth of encoding conditions. (A) Structural and metabolic correlates of episodic memory recall following shallow encoding in the young, middle-aged, and elderly groups. (B) Structural and metabolic correlates of episodic memory recall following deep encoding in the young and middle-aged groups. (C) Structural and metabolic correlates of episodic memory recall following deep encoding in the middle-aged and elderly groups.

).
Correlations between brain data and recall performance following deep encoding in the young and middle-aged groups

This analysis was performed in order to identify the brain regions which might underlie memory capacities in the deep encoding condition—capacities which are preserved in middle-aged adults. The correlation between GM volume data and performance mainly involved the left (and, to a less extent, the right) superior parietal, bilateral frontal, and anterior/middle cingulate cortices. The analysis involving 18FDG uptake highlighted the left frontal cortex, the right parietal cortex, and the anterior (both hemispheres) and middle (right) hippocampus/parahippocampal cortex (Table 4,

Table 4

GMv and FDG Uptake Correlates of Deep Encoding in Young and Middle-aged Adults

MNI Coordinates
t
k
Labeling (BA)
x
y
z
[Young + Middle-aged] Groups: GMv * [Immediate + Delayed] Recalls following Deep Encoding 
−16 −64 48 4.81 96 L Parietal sup (7) 
−8 26 34 4.74 193 L Cingulate ant/mid, Frontal sup med (24, 32) 
26 32 30 4.14 220 R Frontal mid (9, 46) 
16 50 38 3.60 93 R Frontal sup (9) 
−8 −32 78 3.54 92 L Paracentral lobule (4) 
66 −30 40 3.53 30 R Supramarginal (40) 
−32 50 14 3.45 93 L Frontal mid (10, 46) 
−46 18 26 3.35 26 L Frontal inf tri (48) 
−24 40 −16 3.11 94 L Frontal mid orb (11) 
24 −90 22 3.00 37 R Occipital sup (18) 
−30 48 36 2.87 24 L Frontal mid (9, 46) 
 
[Young + Middle-aged] Groups: FDG Uptake * [Immediate + Delayed] Recalls following Deep Encoding 
28 −70 46 4.82 2788 R Parietal sup/inf, occipital sup/mid, precuneus/cuneus, postcentral, angular (7, 19, 39, 40) 
−34 30 −10 3.99 100 L Frontal inf orb (47) 
34 −22 −22 3.88 246 R Parahippocampal, hippocampus, fusiform (ant and mid) (20, 30, 36) 
−26 22 42 3.69 124 L Frontal mid (8, 9, 46) 
−40 −14 −30 3.51 179 L Temporal inf (20) 
−6 −2 50 3.49 270 L SMA, cingulate mid (6, 24) 
−42 16 26 3.39 33 L Frontal inf tri (48) 
12 20 54 3.38 64 R SMA, frontal sup (6, 8, 32) 
46 26 14 3.28 55 R Frontal inf tri (45) 
40 −72 −10 3.20 38 R Occipital inf (19) 
−30 52 3.12 28 L Frontal mid/mid orb/sup (10, 47) 
−14 −10 −20 2.92 21 L Hippocampus (anterior part) 
MNI Coordinates
t
k
Labeling (BA)
x
y
z
[Young + Middle-aged] Groups: GMv * [Immediate + Delayed] Recalls following Deep Encoding 
−16 −64 48 4.81 96 L Parietal sup (7) 
−8 26 34 4.74 193 L Cingulate ant/mid, Frontal sup med (24, 32) 
26 32 30 4.14 220 R Frontal mid (9, 46) 
16 50 38 3.60 93 R Frontal sup (9) 
−8 −32 78 3.54 92 L Paracentral lobule (4) 
66 −30 40 3.53 30 R Supramarginal (40) 
−32 50 14 3.45 93 L Frontal mid (10, 46) 
−46 18 26 3.35 26 L Frontal inf tri (48) 
−24 40 −16 3.11 94 L Frontal mid orb (11) 
24 −90 22 3.00 37 R Occipital sup (18) 
−30 48 36 2.87 24 L Frontal mid (9, 46) 
 
[Young + Middle-aged] Groups: FDG Uptake * [Immediate + Delayed] Recalls following Deep Encoding 
28 −70 46 4.82 2788 R Parietal sup/inf, occipital sup/mid, precuneus/cuneus, postcentral, angular (7, 19, 39, 40) 
−34 30 −10 3.99 100 L Frontal inf orb (47) 
34 −22 −22 3.88 246 R Parahippocampal, hippocampus, fusiform (ant and mid) (20, 30, 36) 
−26 22 42 3.69 124 L Frontal mid (8, 9, 46) 
−40 −14 −30 3.51 179 L Temporal inf (20) 
−6 −2 50 3.49 270 L SMA, cingulate mid (6, 24) 
−42 16 26 3.39 33 L Frontal inf tri (48) 
12 20 54 3.38 64 R SMA, frontal sup (6, 8, 32) 
46 26 14 3.28 55 R Frontal inf tri (45) 
40 −72 −10 3.20 38 R Occipital inf (19) 
−30 52 3.12 28 L Frontal mid/mid orb/sup (10, 47) 
−14 −10 −20 2.92 21 L Hippocampus (anterior part) 

k = cluster size > 20 voxels; BA = Brodmann's area. For abbreviations, refer to Table 3.

, Figure 2B).

Correlations between brain data and recall performance following deep encoding in the middle-aged and elderly groups

This analysis was performed in order to identify those brain regions which might be responsible for memory deficits in the deep encoding condition in the elderly group. The main regions where GM volume was correlated with the recall scores following deep encoding were the frontal areas (especially the left side), the middle and anterior cingulate cortex, and the left cerebellum. The main region whose metabolism was correlated with performance was the right parietal cortex. This analysis also highlighted the right frontal and middle cingulate (posterior segment) cortices (Table 5,

Table 5

GMv and FDG Uptake Correlates of Deep Encoding Condition in Middle-aged and Elderly Adults

MNI Coordinates
t
k
Labeling (BA)
x
y
z
[Middle-aged + Elderly] Groups: GMv * [Immediate + Delayed] Recalls following Deep Encoding 
−26 54 10 3.70 126 L Frontal mid, sup (10, 46) 
−76 −16 3.64 141 L Cerebellum, vermis 
14 52 38 3.56 40 R Frontal sup, sup med (9) 
−4 18 38 3.31 87 L Cingulate mid, frontal sup med (24, 32) 
30 42 32 3.28 46 R Frontal mid (9, 46) 
−2 40 12 2.95 80 L/R Cingulate ant (24, 32) 
 
[Middle-aged + Elderly] Groups: FDG Uptake * [Immediate + Delayed] Recalls following Deep Encoding 
42 −36 50 4.49 1410 R Parietal inf, postcentral (2) 
−2 −14 42 3.33 329 L/R Cingulate mid (24) 
12 54 34 3.26 107 R Frontal sup medial, sup (9, 46) 
−40 −46 46 3.17 186 L Parietal inf/sup (40) 
MNI Coordinates
t
k
Labeling (BA)
x
y
z
[Middle-aged + Elderly] Groups: GMv * [Immediate + Delayed] Recalls following Deep Encoding 
−26 54 10 3.70 126 L Frontal mid, sup (10, 46) 
−76 −16 3.64 141 L Cerebellum, vermis 
14 52 38 3.56 40 R Frontal sup, sup med (9) 
−4 18 38 3.31 87 L Cingulate mid, frontal sup med (24, 32) 
30 42 32 3.28 46 R Frontal mid (9, 46) 
−2 40 12 2.95 80 L/R Cingulate ant (24, 32) 
 
[Middle-aged + Elderly] Groups: FDG Uptake * [Immediate + Delayed] Recalls following Deep Encoding 
42 −36 50 4.49 1410 R Parietal inf, postcentral (2) 
−2 −14 42 3.33 329 L/R Cingulate mid (24) 
12 54 34 3.26 107 R Frontal sup medial, sup (9, 46) 
−40 −46 46 3.17 186 L Parietal inf/sup (40) 

k = cluster size > 20 voxels; BA = Brodmann's area. For abbreviations, refer to Table 3.

, Figure 2C).

Depth of Encoding and Retention Interval Interaction

Regarding the assessment of the neural correlates of the interaction between retention interval and depth of encoding factors, comparisons of correlations were intended to highlight the brain regions whose volume and metabolism might underlie the immediate and delayed recall performance, according to the depth of encoding (for instance, “deep encoding” > “shallow encoding” in immediate recall, i.e., searching for regions where correlations with “immediate recall, deep encoding” were significantly higher than correlations with “immediate recall, shallow encoding”).

In the retrieval of shallowly encoded items, cerebellar and occipital structural and metabolic integrity was associated with immediate recall scores, whereas posterior hippocampal and parahippocampal volume and metabolism were associated with delayed recall scores. Regarding the deeply encoded words, the volumes of the frontal cortex and the middle part of the hippocampus, and right-sided frontal and temporo-parietal metabolism were associated with performance on immediate recall, whereas posterior cortical volume and right parieto-occipital and supplementary motor area metabolism were associated with delayed recall (Table 6,
Table 6

GMv and FDG Uptake Correlates of the Interaction between Delay and Depth of Encoding Factors

MNI Coordinates
t
k
Labeling (BA)
x
y
z
GMv Correlates 
Immediate recall: shallow > deep encoding 
−12 −44 −14 4.10 1380 L/R Cerebellum, vermis, lingual, fusiform, parahippocampal (19, 30) 
−24 −36 −40 3.83 89 L Cerebellum  
16 −88 3.73 213 R Lingual, calcarine (17, 18) 
20 −80 30 3.26 79 R Occipital sup, cuneus (18, 19) 
−90 −30 3.14 34 R Cerebellum  
Immediate recall: deep > shallow encoding 
66 −26 40 4.71 598 R Supramarginal, postcentral (2, 3, 40, 43) 
−6 54 26 4.44 196 L Frontal sup med, mid, sup (9, 10, 32, 46) 
16 50 36 4.22 616 R Frontal mid, sup (9, 46) 
−36 −78 32 4.07 203 L Occipital mid, angular (19, 39) 
−46 16 24 3.63 33 L Frontal inf tri (48) 
20 58 −16 3.42 44 R Frontal mid orb (11) 
46 −48 12 3.34 23 R Temporal mid (21) 
46 −82 22 3.26 30 R Occipital mid (39) 
22 −22 −10 3.19 39 R Hippocampus  
−10 −28 40 3.19 35 L Cingulate mid (23) 
56 −52 38 3.18 50 R Parietal inf (39, 40) 
−62 −12 36 3.04 23 L Postcentral (43) 
−4 66 −8 3.01 118 L/R Frontal med orb (11) 
48 3.01 30 R Frontal sup med, cingulate ant (10) 
38 −10 14 2.99 42 R Insula (48) 
Delayed recall: shallow > deep encoding 
46 −50 12 5.25 168 R Temporal mid (21) 
−32 −48 −4 4.23 177 L Parahippocampal, hippocampus, lingual (37) 
32 −38 −8 4.20 229 R Parahippocampal, hippocampus (20, 37) 
−46 −52 26 3.54 53 L Angular (39, 41) 
50 −28 38 3.26 40 R Supramarginal, postcentral (2) 
−44 −6 44 3.21 20 L Precentral (6) 
−56 −26 20 3.18 42 L Supramarginal, temporal sup (42) 
−62 −16 38 3.04 31 L Postcentral (3, 43) 
Delayed recall: deep > shallow encoding 
16 −84 38 4.00 188 R Cuneus, occipital (18, 19) 
−64 −12 3.62 97 L Vermis, cerebellum  
−4 −28 78 3.30 31 L Paracentral lobule (4) 
18 −56 22 3.14 47 R Precuneus, cuneus (17, 23) 
−6 18 40 3.14 23 L Cingulate mid, frontal sup med (32) 
 
FDG Uptake Correlates 
Immediate recall: shallow > deep encoding 
−18 −34 −30 4.16 1994 L Cerebellum, fusiform, parahippocampal (18, 19, 30, 37) 
14 −24 66 3.32 84 R Precentral, paracentral lobule (4, 6) 
−30 18 12 3.17 199 L Insula, putamen (48) 
−10 −34 68 3.13 30 L Paracentral lobule (34) 
Immediate recall: deep > shallow encoding 
32 52 4.52 1071 R Frontal inf tri, orb, mid, sup (11, 45, 47) 
−30 52 −2 4.09 233 L Frontal sup, orb, mid (10, 11, 47) 
62 −18 24 3.93 1077 R Supramarginal, temporal mid/sup (2, 21, 22, 42, 48) 
−10 54 3.68 52 L Frontal sup med, med orb (10) 
62 20 3.24 99 R Precentral (6) 
Delayed recall: shallow > deep encoding 
34 −34 −10 5.32 394 R Hippocampus, parahippocampal, fusiform (20, 37) 
26 −16 4.08 264 R Amygdala, putamen, olfactory (34, 48) 
−64 −32 3.51 203 L Temporal mid/sup (21, 22) 
−36 −28 3.50 432 Vermis 
28 −64 −28 3.30 398 R Cerebellum 
−64 −18 38 3.26 43 L Postcentral, supramarginal (1, 43) 
−30 −40 −8 3.24 115 L Parahippocampal, hippocampus (37) 
−6 −70 −30 3.21 84 L Cerebellum 
−24 −78 −32 2.99 71 L Cerebellum 
Delayed recall: deep > shallow encoding 
−2 −2 58 5.14 1288 L/R SMA, cingulate mid (6, 24, 32) 
28 −72 46 5.10 1474 R Parietal sup, occipital sup/mid, precuneus, cuneus, angular (7, 19) 
−10 −42 64 3.37 195 L Precuneus, paracentral lobule (4, 5) 
16 −28 70 3.25 133 R Precentral, paracentral lobule (4, 6) 
−18 −78 42 3.16 99 L Occipital sup, parietal sup (7, 19) 
42 −12 42 3.13 25 R Precentral (3, 6) 
38 −30 56 3.03 47 R Postcentral (3) 
MNI Coordinates
t
k
Labeling (BA)
x
y
z
GMv Correlates 
Immediate recall: shallow > deep encoding 
−12 −44 −14 4.10 1380 L/R Cerebellum, vermis, lingual, fusiform, parahippocampal (19, 30) 
−24 −36 −40 3.83 89 L Cerebellum  
16 −88 3.73 213 R Lingual, calcarine (17, 18) 
20 −80 30 3.26 79 R Occipital sup, cuneus (18, 19) 
−90 −30 3.14 34 R Cerebellum  
Immediate recall: deep > shallow encoding 
66 −26 40 4.71 598 R Supramarginal, postcentral (2, 3, 40, 43) 
−6 54 26 4.44 196 L Frontal sup med, mid, sup (9, 10, 32, 46) 
16 50 36 4.22 616 R Frontal mid, sup (9, 46) 
−36 −78 32 4.07 203 L Occipital mid, angular (19, 39) 
−46 16 24 3.63 33 L Frontal inf tri (48) 
20 58 −16 3.42 44 R Frontal mid orb (11) 
46 −48 12 3.34 23 R Temporal mid (21) 
46 −82 22 3.26 30 R Occipital mid (39) 
22 −22 −10 3.19 39 R Hippocampus  
−10 −28 40 3.19 35 L Cingulate mid (23) 
56 −52 38 3.18 50 R Parietal inf (39, 40) 
−62 −12 36 3.04 23 L Postcentral (43) 
−4 66 −8 3.01 118 L/R Frontal med orb (11) 
48 3.01 30 R Frontal sup med, cingulate ant (10) 
38 −10 14 2.99 42 R Insula (48) 
Delayed recall: shallow > deep encoding 
46 −50 12 5.25 168 R Temporal mid (21) 
−32 −48 −4 4.23 177 L Parahippocampal, hippocampus, lingual (37) 
32 −38 −8 4.20 229 R Parahippocampal, hippocampus (20, 37) 
−46 −52 26 3.54 53 L Angular (39, 41) 
50 −28 38 3.26 40 R Supramarginal, postcentral (2) 
−44 −6 44 3.21 20 L Precentral (6) 
−56 −26 20 3.18 42 L Supramarginal, temporal sup (42) 
−62 −16 38 3.04 31 L Postcentral (3, 43) 
Delayed recall: deep > shallow encoding 
16 −84 38 4.00 188 R Cuneus, occipital (18, 19) 
−64 −12 3.62 97 L Vermis, cerebellum  
−4 −28 78 3.30 31 L Paracentral lobule (4) 
18 −56 22 3.14 47 R Precuneus, cuneus (17, 23) 
−6 18 40 3.14 23 L Cingulate mid, frontal sup med (32) 
 
FDG Uptake Correlates 
Immediate recall: shallow > deep encoding 
−18 −34 −30 4.16 1994 L Cerebellum, fusiform, parahippocampal (18, 19, 30, 37) 
14 −24 66 3.32 84 R Precentral, paracentral lobule (4, 6) 
−30 18 12 3.17 199 L Insula, putamen (48) 
−10 −34 68 3.13 30 L Paracentral lobule (34) 
Immediate recall: deep > shallow encoding 
32 52 4.52 1071 R Frontal inf tri, orb, mid, sup (11, 45, 47) 
−30 52 −2 4.09 233 L Frontal sup, orb, mid (10, 11, 47) 
62 −18 24 3.93 1077 R Supramarginal, temporal mid/sup (2, 21, 22, 42, 48) 
−10 54 3.68 52 L Frontal sup med, med orb (10) 
62 20 3.24 99 R Precentral (6) 
Delayed recall: shallow > deep encoding 
34 −34 −10 5.32 394 R Hippocampus, parahippocampal, fusiform (20, 37) 
26 −16 4.08 264 R Amygdala, putamen, olfactory (34, 48) 
−64 −32 3.51 203 L Temporal mid/sup (21, 22) 
−36 −28 3.50 432 Vermis 
28 −64 −28 3.30 398 R Cerebellum 
−64 −18 38 3.26 43 L Postcentral, supramarginal (1, 43) 
−30 −40 −8 3.24 115 L Parahippocampal, hippocampus (37) 
−6 −70 −30 3.21 84 L Cerebellum 
−24 −78 −32 2.99 71 L Cerebellum 
Delayed recall: deep > shallow encoding 
−2 −2 58 5.14 1288 L/R SMA, cingulate mid (6, 24, 32) 
28 −72 46 5.10 1474 R Parietal sup, occipital sup/mid, precuneus, cuneus, angular (7, 19) 
−10 −42 64 3.37 195 L Precuneus, paracentral lobule (4, 5) 
16 −28 70 3.25 133 R Precentral, paracentral lobule (4, 6) 
−18 −78 42 3.16 99 L Occipital sup, parietal sup (7, 19) 
42 −12 42 3.13 25 R Precentral (3, 6) 
38 −30 56 3.03 47 R Postcentral (3) 

k = cluster size > 20 voxels; BA = Brodmann's area. For abbreviations, refer to Table 3.

, Figure 3 
Figure 3

GMv and FDG uptake correlates of episodic memory according to the depth of encoding and the retention interval.

Figure 3

GMv and FDG uptake correlates of episodic memory according to the depth of encoding and the retention interval.

).

DISCUSSION

The present study sought to identify the neural substrates of episodic retrieval impairment in the middle-aged and the elderly, by manipulating two factors: depth of encoding and retention interval. In the first part of this discussion, we look at the effect of the depth of encoding in aging and its neural correlates, namely, (1) the parahippocampal regions in the shallow encoding condition, where middle-aged subjects showed decreasing performance, and (2) the frontal and parietal cortices as well as the anterior hippocampal formation in the deep encoding condition, where only the elderly group displayed a decline. In the second part, we briefly discuss the effect of the interaction between the depth of encoding and retention interval factors.

We found that deeply encoded items were recalled better than shallowly encoded words. This is consistent with the expected effects of the processing levels, in that “trace persistence is a function of depth of analysis, with deeper levels of analysis associated with more elaborate, longer lasting, and stronger traces” (Craik & Lockhart, 1972). This may be explained by an enhancement of the semantic elaboration relating to the information to be recovered. Our findings also highlighted a nonlinear progression of episodic memory decline in normal aging: In the middle-aged adults, the recall of shallowly encoded words was already impaired, whereas these subjects performed as well as the younger ones when semantic encoding was experimentally reinforced, suggesting preserved retrieval capacities in this condition, and/or enhanced encoding capacities. This semantic reinforcement at the encoding phase was sufficient to produce associated contextual features, enabling the middle-aged subjects to recollect them thereafter, and thus, retrieve the items. As far as the elderly subjects are concerned, they enhanced but failed to normalize their retrieval performance when deep encoding was induced, suggesting only partially successful compensatory mechanisms.

Structural and Metabolic Correlates of Episodic Memory Performance in Normal Aging

By contrast with the idea that age-related changes are more functionally than structurally based, our general assumption was that both neural changes would underlie cognitive alteration. In the same vein, we assumed that divergence of previous findings regarding the relationship highlighted throughout the literature between structural data and episodic memory scores could have resulted from the method (i.e., the use of an ROI procedure including the whole hippocampus or extended frontal areas), as different subregions of these areas may underlie different cognitive processes and are differentially affected by age. Indeed, using a methodology without a priori and two different encoding conditions, our results support the view that both structural and functional changes, especially in hippocampal subregions and specific frontal areas, underlie episodic memory changes.

The Retrieval of Shallowly Encoded Words

The significant correlations obtained by considering all three groups together underlined the importance of posterior mediotemporal structural and metabolic integrity for the retrieval of episodic memory items following shallow encoding. This finding suggests that a dysfunction in this region could mediate the memory deficit in middle-aged and elderly subjects in this condition. These results are in line with recent studies that have associated posterior parahippocampal activity with familiarity-based retrieval processes (Daselaar, Fleck, & Cabeza, 2006) or perception-based ones (Goh et al., 2004; Cabeza, Rao, Wagner, Mayer, & Schacter, 2001). In the shallow condition of our study, the material was solely subject to perceptual processing, and the participants were not given any instructions to memorize it. Therefore, the retrieval of the words was mainly perception- or familiarity-based, without any mandatory contextual recollection. This explanation also fits in with the fact that the posterior parahippocampal areas receive many direct inputs from perceptual regions (Yonelinas, 2002). Furthermore, our significant correlations concerned the lingual or fusiform cortices, which are mainly involved in visual processes. When van der Veen et al. (2006) assessed verbal episodic memory in young and elderly subjects, testing the effects of incidental and intentional encoding on retrieval performance, the younger adults, compared with the older, displayed additional activation of the middle occipital gyrus, especially in the incidental condition. This result, therefore, supports the hypothesis of quite an early age-related impairment of the neurocognitive processes engaged in shallow memory conditions.

Regarding the cerebellum, some authors have observed increased activity during encoding (Beason-Held, Golski, Kraut, Esposito, & Resnick, 2005; Otten et al., 2002) or retrieval (van der Veen et al., 2006; Cabeza, Dolcos, Graham, & Nyberg, 2002), or during both encoding and retrieval (Weis, Klaver, Reul, Elger, & Fernandez, 2004). Although its contribution to episodic memory has already been described, the precise nature of the cerebellum functions remains unclear. Fliessbach, Trautner, Quesada, Elger, and Weber (2007) recently conducted an incidental verbal memory fMRI experiment with three different encoding tasks manipulating the levels of processing. In an alphabetical decision encoding task similar to ours, the authors found that two regions in the left and the right superior cerebellum displayed a subsequent memory effect, suggesting specific neural processes for the encoding of perceptually processed items, and thus, supporting our findings.

The Retrieval of Deeply Encoded Words

Our findings concerning the deep encoding condition contrast with frequent failures to find any significant relationship between frontal GM volume and episodic memory performance (see Van Petten et al., 2004 for a review). We hypothesized that this failure of previous studies might be due to several methodological issues, notably the nature of the tasks and the way the neuroimaging data was processed (see Introduction). Taken together, the results of our task (specially designed to study different cognitive processes) and the voxel-based findings argue strongly in favor of a relationship between several specific anterior regions and performance in age-related episodic memory decline and preservation. Thus frontal structural integrity, may be a precondition to perform successfully an episodic memory task which includes deep semantic processes at encoding.

Many studies have shown that elderly people activate the frontal cortex bilaterally instead of unilaterally to compensate for their cognitive impairment (Cabeza, 2002; Dolcos et al., 2002). In our metabolic analysis notably, which included young and middle-aged subjects, the left frontal areas, which are known to play a role in semantic processing, were more closely correlated with recall performance following deep encoding than the right frontal areas, with middle-aged adults performing as accurately as younger ones. On the other hand, still considering metabolic findings, more pronounced correlations were found in the right frontal cortex than in the left in the analysis gathering the middle-aged and elderly groups. This result suggests limited neurocognitive compensatory mechanisms in the elderly group: Although they improved their performance in comparison with the shallow encoding condition, they failed to achieve middle-aged adults' scores. These findings fit in with Rajah and D'Esposito's (2005) meta-analysis showing that left dorsal and anterior prefrontal cortex activity may reflect effective functional compensation in elderly subjects, whereas right dorsal and anterior frontal cortex disruption contributes to age-related memory impairment.

As far as the parietal cortex is concerned, a new body of literature has recently emerged regarding its involvement in episodic memory (Babiloni et al., 2006; Iidaka et al., 2006; Naghavi & Nyberg, 2005; Wagner, Shannon, Kahn, & Buckner, 2005). Studies including both young and elderly subjects have found that the right side of this region is activated more often in young adults than in older ones, whose scores tend to be less accurate (see Persson et al., 2006). In the present study, we found right parietal metabolism to be correlated with performance following deep encoding in both analyses (i.e., young and middle-aged subjects, who displayed similar performance, and middle-aged and elderly groups where the former outperformed the latter). Thus, our study confirmed the function of the right parietal cortex in episodic memory performance, and beyond, our finding emphasized its role in the retrieval of deeply encoded words. A fronto-parietal network is thought to be involved in strategic processes during episodic retrieval (Nyberg, Forkstam, Petersson, Cabeza, & Ingvar, 2002) and would contribute to the maintenance of attention on internal mnemonic representations (see Wagner et al., 2005; Cabeza & Nyberg, 2000 for reviews). In the light of these proposals, and considering that the retrieval of deeply semantic encoded items would rely on strategic and attentional processes, our findings suggest that this attentional fronto-parietal network may be preserved in middle-aged subjects, but inaccurate in elderly subjects. Furthermore, as regard the laterality, we could relate these findings to studies which used the remember/know paradigm. They have highlighted a modulation function of the parietal cortex regarding the retrieval mode: The left fronto-parietal regions are more likely to underlie recollection than familiarity processes (Wheeler & Buckner, 2004; Henson, Rugg, Shallice, Josephs, & Dolan, 1999), whereas right parietal activity is likely to reflect familiarity mechanisms (Iidaka et al., 2006; Yonelinas, Otten, Shaw, & Rugg, 2005). Our findings might be interpreted as compensatory mechanisms, such as the younger groups might have enhanced their performance via familiarity-based processes in case they could not retrieve the items via recollection (indeed, the deep encoding condition permitted them to create retrieval cues because they elaborated sentences). By contrast, the elderly adults' compensatory mechanisms might not be sufficient to bring their performance up to the level of the younger groups. This interpretation of our results is to be considered with caution because the remember/know paradigm has not been applied. More specific investigations must be performed to test this assumption.

An interesting postero-medial portion of the parietal lobe, the precuneus, is also part of the network involved in episodic memory, as shown for the retrieval of source memory (for a review, see Cavanna & Trimble, 2006), autobiographical memories (Viard et al., 2007; Gilboa, Winocur, Grady, Hevenor, & Moscovitch, 2004), and rich episodic contextual information (Lundstrom, Ingvar, & Petersson, 2005). These findings can help us to interpret the correlation we observed between precuneus metabolism and performance in both young and middle-aged subjects, as many of the sentences they constructed were related to their own experiences. Similarly, during the debriefing, our subjects said that during the recall tests, they tried to remember the associations they had made during the deep encoding phase. Lastly, anterior hippocampal and parahippocampal (especially right-lateralized) metabolism was found to be linked with the retrieval of deeply encoded words in the analysis that enabled us to highlight those regions that might support preservation in the middle-aged group (i.e., the regions correlated with performance in young and middle-aged subjects). The anterior hippocampus is known to be preferentially activated in encoding processes (Lepage et al., 1998), suggesting that its involvement during deep encoding in middle-aged adults may have enhanced recall. More specific studies have also revealed its activity in the retrieval of recollection- rather than familiarity-based items (Yonelinas et al., 2005). Furthermore, the anterior hippocampal formation is known to be relatively spared by normal aging effects (Kalpouzos et al., in press), supporting the hypothesis of preserved mnemonic mechanisms linked with this specific area of the brain.

In sum, preserved retrieval processing in middle-aged adults following deep encoding is likely to be mediated by (1) effective semantic processes which depend on the left frontal cortex (Logan et al., 2002), (2) accurate encoding of the information relying on the anterior hippocampal areas, (3) accurate attentional processes, and perhaps multistrategic retrieval including recollection, familiarity, and reference to autobiographical memories supported by the parietal cortex. In the elderly, diminished performance in the deep encoding condition might be due to inadequate semantic elaborative mechanisms underlied by frontal regions, to reduced attentional processes and inaccurate compensatory mechanisms at recovery, underlied by the right parietal regions.

Depth of Encoding and Retention Intervals

In accordance with the literature, we did not find any effect of age on retention intervals, suggesting that after a 1-week interval, there were similar declines in the number of remembered items in all three groups (Fjell et al., 2005; Tombaugh & Hubley, 2001). In contrast, we found an interesting interaction between depth of encoding and retention interval on recall performance. Although the depth of encoding was still beneficial after a week, there was, nonetheless, a decline.

Numerous studies have highlighted the role of hippocampal volume in the retrieval of episodic information, particularly after long retention intervals (Fjell et al., 2005; Walhovd et al., 2004; Sullivan et al., 1995; Golomb et al., 1994). These findings are consistent with the fact that the hippocampal area is involved in consolidation processes, which take place over quite long periods (Viard et al., 2007; Nadel & Moscovitch, 1997). The interaction we observed between this factor and the depth of encoding suggests more subtle effects. In effect, although the structural and functional integrity of the posterior hippocampal and parahippocampal regions is required for delayed retrieval, this mainly concerns shallowly encoded words. This result confirms findings obtained in healthy young volunteers, showing a significant correlation between left posterior parahippocampal metabolism assessed during the incidental encoding of words and the number of words recalled after 24 hours (Alkire, Haier, Fallon, & Cahill, 1998). In addition, our findings concerning the interaction between depth of encoding and retention interval suggest that the cerebellum is involved in the retrieval of shallowly encoded material (as discussed above), especially when recall is immediate.

Lastly, as has already been discussed, we have established that the structural and functional integrity of the frontal cortex is required more for deeply encoded words than for shallowly encoded ones, and by taking the retention interval factor into account, we have also been able to highlight its involvement in immediate recall. Our data also underline the role of the parietal cortex in both the immediate and delayed recall of deeply encoded material. In order to confirm these findings, further studies are needed to probe the interactions between the depth of encoding and different retention intervals.

Conclusion

The main purpose of this study was to investigate the structural and metabolic neural substrates of episodic memory in normal aging, with an emphasis on depth of encoding effects. Three structures were strongly related to these cognitive processes: the frontal and parietal cortices and the mediotemporal areas. Structure and metabolism were both found to play a part in age-related cognitive changes. This general finding involves that structural age-related changes should be taken into account in activation paradigms, as they could be responsible for the changes observed in neurocognitive networks. In middle-aged and elderly adults, recall following shallow encoding was diminished compared with that of younger subjects, and this was mainly related to posterior mediotemporal volume and metabolism. When deep (semantic) encoding was experimentally induced, this group performed as well as the younger group. The regions related to the retrieval of deeply encoded words in the middle-aged subjects were the left frontal cortex, the anterior and middle parts of the hippocampal areas, and the parietal cortex, suggesting (1) successful semantic processing, (2) effective encoding in episodic memory, (3) accurate attentional processes, and perhaps the involvement of several retrieval strategies, such as recollection, familiarity, and the recall of autobiographical memories and contextual features. The elderly subjects performed less well following deep encoding. The regions related to this decrease were the frontal and parietal cortices, suggesting reduced attentional processes and the involvement of inadequate strategic processes.

Disclosure Statement

All authors disclose no actual or potential conflicts of interest with other people or organizations within 3 years of beginning the present work.

Acknowledgments

We thank Ms. F. Mézenge, Ms. M.-H. Noël, Ms. A. Pélerin, Mr. O. Tirel, Dr. J.-M. Constans, Mr. V. Beaudouin, Dr. A. Abbas, Mr. P. Conejero, Ms. N. Jacques for their help in this study. This research was funded by the G.I.S. Institut de la longévité et du vieillissement, INSERM.

Reprint requests should be sent to Béatrice Desgranges, INSERM—EPHE—Université de Caen/Basse-Normandie, U 923, Laboratoire de Neuropsychologie, CHU Côte de Nacre, 14033 Caen Cedex, France, or via e-mail: desgranges-b@chu-caen.fr.

REFERENCES

REFERENCES
Alkire
,
M. T.
,
Haier
,
R. J.
,
Fallon
,
J. H.
, &
Cahill
,
L.
(
1998
).
Hippocampal, but not amygdala, activity at encoding correlates with long-term, free recall of nonemotional information.
Proceedings of the National Academy of Sciences, U.S.A.
,
95
,
14506
14510
.
Allan
,
K.
,
Robb
,
W. G. K.
, &
Rugg
,
M. D.
(
2000
).
The effect of encoding manipulations on neural correlates of episodic retrieval.
Neuropsychologia
,
38
,
1188
1205
.
Allen
,
J. S.
,
Bruss
,
J.
,
Brown
,
C. K.
, &
Damasio
,
H.
(
2005
).
Normal neuroanatomical variation due to age: The major lobes and parcellation of the temporal region.
Neurobiology of Aging
,
26
,
1245
1260
.
Babiloni
,
C.
,
Vecchio
,
F.
,
Cappa
,
S.
,
Pasqualetti
,
P.
,
Rossi
,
S.
,
Miniussi
,
C.
,
et al
(
2006
).
Functional frontoparietal connectivity during encoding and retrieval processes follows HERA model. A high-resolution study.
Brain Research Bulletin
,
68
,
203
212
.
Beason-Held
,
L. L.
,
Golski
,
S.
,
Kraut
,
M. A.
,
Esposito
,
G.
, &
Resnick
,
S. M.
(
2005
).
Brain activation during encoding and recognition of verbal and figural material information in older adults.
Neurobiology of Aging
,
26
,
237
250
.
Beason-Held
,
L. L.
,
Kraut
,
M. A.
, &
Resnick
,
S. M.
(
2008
).
II. Temporal patterns of longitudinal change in aging brain function.
Neurobiology of Aging
,
29
,
497
513
.
Brickman
,
A. M.
,
Habeck
,
C.
,
Zarahn
,
E.
,
Flynn
,
J.
, &
Stern
,
Y.
(
2007
).
Structural MRI covariance patterns associated with normal aging and neuropsychological functioning.
Neurobiology of Aging
,
28
,
284
295
.
Cabeza
,
R.
(
2002
).
Hemispheric asymmetry reduction in older adults: The HAROLD model.
Psychology and Aging
,
17
,
85
100
.
Cabeza
,
R.
,
Anderson
,
N. D.
,
Locantore
,
J. K.
, &
McIntosh
,
R.
(
2002
).
Aging gracefully: Compensatory brain activity in high-performing older adults.
Neuroimage
,
17
,
1394
1402
.
Cabeza
,
R.
,
Daselaar
,
S. M.
,
Dolcos
,
F.
,
Prince
,
S. E.
,
Budde
,
M.
, &
Nyberg
,
L.
(
2004
).
Task-independent and task-specific age effects on brain activity during working memory, visual attention and episodic retrieval.
Cerebral Cortex
,
14
,
364
375
.
Cabeza
,
R.
,
Dolcos
,
F.
,
Graham
,
R.
, &
Nyberg
,
L.
(
2002
).
Similarities and differences in the neural correlates of episodic memory retrieval and working memory.
Neuroimage
,
16
,
317
330
.
Cabeza
,
R.
, &
Nyberg
,
L.
(
2000
).
Imaging cognition: II. An empirical review of 275 PET and fMRI studies.
Journal of Cognitive Neuroscience
,
12
,
1
47
.
Cabeza
,
R.
,
Rao
,
S. M.
,
Wagner
,
A. D.
,
Mayer
,
A. R.
, &
Schacter
,
D. L.
(
2001
).
Can medial temporal lobe regions distinguish true from false? An event-related functional MRI study of veridical and illusory recognition memory.
Proceedings of the National Academy of Sciences, U.S.A.
,
98
,
4805
4810
.
Cavanna
,
A. E.
, &
Trimble
,
M. R.
(
2006
).
The precuneus: A review of its functional anatomy and behavioural correlates.
Brain
,
129
,
564
583
.
Chételat
,
G.
,
Desgranges
,
B.
,
de la Sayette
,
V.
,
Viader
,
F.
,
Berkouk
,
K.
,
Landeau
,
B.
,
et al
(
2003
).
Dissociating atrophy and hypometabolism impact on episodic memory in mild cognitive impairment.
Brain
,
126
,
1955
1967
.
Craik
,
F. I. M.
, &
Byrd
,
M.
(
1982
).
Aging and cognitive deficits: The role of attentional resources.
In F. I. M. Craik & S. E. Trehub (Eds.),
Aging and cognitive processes
(pp.
191
211
).
New York
:
Plenum
.
Craik
,
F. I. M.
, &
Jennings
,
J. M.
(
1992
).
Human memory.
In F. I. M. Craik & T. A. Salthouse (Eds.),
The handbook of aging and cognition
(pp.
51
110
).
Hillsdale, NJ
:
Erlbaum
.
Craik
,
F. I. M.
, &
Lockhart
,
R. S.
(
1972
).
Levels of processing: A framework for memory research.
Journal of Verbal Learning and Verbal Behavior
,
11
,
671
684
.
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
.
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
.
Daselaar
,
S. M.
,
Fleck
,
M. S.
, &
Cabeza
,
R.
(
2006
).
Triple dissociation in the medial temporal lobes: Recollection, familiarity, and novelty.
Journal of Neurophysiology
,
96
,
1902
1911
.
Dolcos
,
F.
,
Rice
,
H. J.
, &
Cabeza
,
R.
(
2002
).
Hemispheric asymmetry and aging: Right hemisphere decline or asymmetry reduction.
Neuroscience and Biobehavioral Reviews
,
26
,
819
825
.
Eustache
,
F.
,
Desgranges
,
B.
, &
Lalevée
,
C.
(
1998
).
L'évaluation clinique de la mémoire.
Revue Neurologique (Paris)
,
154
,
18
32
.
Fay
,
S.
,
Isingrini
,
M.
,
Ragot
,
R.
, &
Pouthas
,
V.
(
2005
).
The effect of encoding manipulation on word-stem cued recall: An event-related potential study.
Cognitive Brain Research
,
24
,
615
626
.
Fjell
,
A. M.
,
Walhovd
,
K. B.
,
Reinvang
,
I.
,
Lundervold
,
A.
,
Dale
,
A. M.
,
Quinn
,
B. T.
,
et al
(
2005
).
Age does not increase rate of forgetting over weeks—Neuroanatomical volumes and visual memory across the adult life-span.
Journal of the International Neuropsychological Society
,
11
,
2
15
.
Fliessbach
,
K.
,
Trautner
,
P.
,
Quesada
,
C. M.
,
Elger
,
C. E.
, &
Weber
,
B.
(
2007
).
Cerebellar contributions to episodic memory encoding as revealed by fMRI.
Neuroimage
,
35
,
1330
1337
.
Gilboa
,
A.
,
Winocur
,
G.
,
Grady
,
C. L.
,
Hevenor
,
S. J.
, &
Moscovitch
,
M.
(
2004
).
Remembering our past: Functional neuroanatomy of recollection of recent and very remote personal events.
Cerebral Cortex
,
14
,
1214
1225
.
Goh
,
J. O. S.
,
Siong
,
S. C.
,
Park
,
D.
,
Gutchess
,
A.
,
Hebrank
,
A.
, &
Chee
,
M. W. L.
(
2004
).
Cortical areas involved in object, background, and object-background processing revealed with functional magnetic resonance adaptation.
Journal of Neuroscience
,
24
,
10223
10228
.
Golomb
,
J.
,
Kluger
,
A.
,
de Leon
,
M. J.
,
Ferris
,
S. H.
,
Convit
,
A.
,
Mittelman
,
M. S.
,
et al
(
1994
).
Hippocampal formation size in normal human aging: A correlate of delayed secondary memory performance.
Learning & Memory
,
1
,
45
54
.
Good
,
C. D.
,
Johnsrude
,
I. S.
,
Ashburner
,
J.
,
Henson
,
R. N. A.
,
Friston
,
K. J.
, &
Frackowiak
,
R. S. J.
(
2001
).
A voxel-based morphometric study of ageing in 465 normal adult human brains.
Neuroimage
,
14
,
21
36
.
Grady
,
C. L.
,
McIntosh
,
A. R.
,
Rajah
,
M. N.
,
Beig
,
S.
, &
Craik
,
F. I. M.
(
1999
).
The effects of age on the neural correlates of episodic encoding.
Cerebral Cortex
,
9
,
805
814
.
Grady
,
C. L.
,
Springer
,
M. V.
,
Hongwanishkul
,
D.
,
McIntosh
,
A. R.
, &
Winocur
,
G.
(
2006
).
Age-related changes in brain activity across the adult lifespan.
Journal of Cognitive Neuroscience
,
18
,
227
241
.
Greicius
,
M. D.
,
Krasnow
,
B.
,
Reiss
,
A. L.
, &
Menon
,
V.
(
2003
).
Functional connectivity in the resting brain: A network analysis of the default mode hypothesis.
Proceedings of the National Academy of Sciences, U.S.A.
,
100
,
253
258
.
Grieve
,
S. M.
,
Clark
,
C. R.
,
Williams
,
L. M.
,
Peduto
,
A. J.
, &
Gordon
,
E.
(
2005
).
Preservation of limbic and paralimbic structures in ageing.
Human Brain Mapping
,
25
,
391
401
.
Habib
,
R.
,
Nyberg
,
L.
, &
Tulving
,
E.
(
2003
).
Hemispheric asymmetries of memory: The HERA model revisited.
Trends in Cognitive Sciences
,
7
,
241
245
.
Henson
,
R. N. A.
,
Rugg
,
M. D.
,
Shallice
,
T.
,
Josephs
,
O.
, &
Dolan
,
R. J.
(
1999
).
Recollection and familiarity in recognition memory: An event-related functional magnetic resonance imaging study.
Journal of Neuroscience
,
19
,
3962
3972
.
Iidaka
,
T.
,
Matsumoto
,
A.
,
Nogawa
,
J.
,
Yamamoto
,
Y.
, &
Sadato
,
N.
(
2006
).
Frontoparietal network involved in successful retrieval from episodic memory. Spatial and temporal analyses using fMRI and ERP.
Cerebral Cortex
,
16
,
1349
1360
.
Isingrini
,
M.
, &
Taconnat
,
L.
(
1997
).
Aspects du vieillissement normal de la mémoire.
Psychologie Française
,
42
,
319
331
.
Kalpouzos
,
G.
,
Chételat
,
G.
,
Baron
,
J.-C.
,
Landeau
,
B.
,
Godeau
,
C.
,
Barré
,
L.
,
et al
(
in press
).
Voxel-based mapping of brain gray matter and metabolism preservation and decline in normal aging.
Neurobiology of Aging
.
Kapur
,
S.
,
Craik
,
F. I. M.
,
Tulving
,
E.
,
Wilson
,
A. A.
,
Houle
,
S.
, &
Brown
,
G. M.
(
1994
).
Neuroanatomical correlates of encoding in episodic memory: Levels of processing effect.
Proceedings of the National Academy of Sciences, U.S.A.
,
91
,
2008
2011
.
Lepage
,
M.
,
Habib
,
R.
, &
Tulving
,
E.
(
1998
).
Hippocampal PET activations of memory encoding and retrieval: The HIPER model.
Hippocampus
,
8
,
313
322
.
Logan
,
J. M.
,
Sanders
,
A. L.
,
Snyder
,
A. Z.
,
Morris
,
J. C.
, &
Buckner
,
R. L.
(
2002
).
Under-recruitment and nonselective recruitment: Dissociable neural mechanisms associated with aging.
Neuron
,
33
,
827
840
.
Lundstrom
,
B. N.
,
Ingvar
,
M.
, &
Petersson
,
K. M.
(
2005
).
The role of precuneus and left inferior frontal cortex during source memory episodic retrieval.
Neuroimage
,
27
,
824
834
.
Mandzia
,
J. L.
,
Black
,
S. E.
,
McAndrews
,
M. P.
,
Grady
,
C.
, &
Graham
,
S.
(
2004
).
fMRI differences in encoding and retrieval of pictures due to encoding strategy in the elderly.
Human Brain Mapping
,
21
,
1
14
.
Mattis
,
S.
(
1976
).
Mental status examination for organic mental syndrome in the elderly patient.
In L. Bellack & T. B. Karasu (Eds.),
Geriatrics psychiatry: A handbook for psychiatrists and primary care physicians
(pp.
77
101
).
New York
:
Grune & Stratton
.
Mungas
,
D.
,
Harvey
,
D.
,
Reed
,
B. R.
,
Jagust
,
W. J.
,
DeCarli
,
C.
,
Beckett
,
L.
,
et al
(
2005
).
Longitudinal volumetric MRI change and rate of cognitive decline.
Neurology
,
65
,
565
571
.
Nadel
,
L.
, &
Moscovitch
,
M.
(
1997
).
Memory consolidation, retrograde amnesia and the hippocampal complex.
Current Opinion in Neurobiology
,
7
,
217
227
.
Naghavi
,
H. R.
, &
Nyberg
,
L.
(
2005
).
Common fronto-parietal activity in attention, memory and consciousness: Shared demands on integration?
Consciousness and Cognition
,
14
,
390
425
.
Nyberg
,
L.
,
Forkstam
,
C.
,
Petersson
,
K. M.
,
Cabeza
,
R.
, &
Ingvar
,
M.
(
2002
).
Brain imaging of human memory systems: Between-systems similarities and within-system differences.
Cognitive Brain Research
,
13
,
281
292
.
Nyberg
,
L.
,
McIntosh
,
A. R.
,
Houle
,
S.
,
Nilsson
,
L.-G.
, &
Tulving
,
E.
(
1996
).
Activation of medial temporal structures during episodic memory retrieval.
Nature
,
380
,
715
717
.
Otten
,
L. J.
,
Henson
,
N. A.
, &
Rugg
,
M. D.
(
2002
).
Depth of processing effects on neural correlates of memory encoding: Relationship between findings from across- and within-task comparisons.
Brain
,
124
,
399
412
.
Park
,
D. C.
,
Lautenschlager
,
G.
,
Hedden
,
T.
,
Davidson
,
N. S.
,
Smith
,
A. D.
, &
Smith
,
P. K.
(
2002
).
Models of visuospatial and verbal memory across the adult life span.
Psychology and Aging
,
17
,
299
320
.
Persson
,
J.
,
Nyberg
,
L.
,
Lind
,
J.
,
Larsson
,
A.
,
Nilsson
,
L.-G.
,
Ingvar
,
M.
,
et al
(
2006
).
Structure–function correlates of cognitive decline in aging.
Cerebral Cortex
,
16
,
907
915
.
Raichle
,
M. E.
,
MacLeod
,
A. M.
,
Snyder
,
A. Z.
,
Powers
,
W. J.
,
Gusnard
,
D. A.
, &
Shulman
,
G. L.
(
2001
).
A default mode of brain function.
Proceedings of the National Academy of Sciences, U.S.A.
,
98
,
676
682
.
Rajah
,
M. N.
, &
D'Esposito
,
M.
(
2005
).
Region-specific changes in prefrontal function with age: A review of PET and fMRI studies on working and episodic memory.
Brain
,
128
,
1964
1983
.
Rauchs
,
G.
,
Piolino
,
P.
,
Mézenge
,
F.
,
Landeau
,
B.
,
Lalevée
,
C.
,
Pélerin
,
A.
,
et al
(
2007
).
Autonoetic consciousness in Alzheimer's disease: Neuropsychological and PET findings using an episodic learning and recognition task.
Neurobiology of Aging
,
28
,
1410
1420
.
Rönnlund
,
M.
,
Nyberg
,
L.
,
Bäckman
,
L.
, &
Nilsson
,
L.-G.
(
2005
).
Stability, growth, and decline in adult life span development of declarative memory: Cross-sectional and longitudinal data from a population-based study.
Psychology and Aging
,
20
,
3
18
.
Rosen
,
A. C.
,
Prull
,
M. W.
,
O'Hara
,
R.
,
Race
,
E. A.
,
Desmond
,
J. E.
,
Glover
,
G. H.
,
et al
(
2002
).
Variable effects of aging on frontal lobe contributions to memory.
NeuroReport
,
13
,
2425
2428
.
Rugg
,
M. D.
,
Fletcher
,
P. C.
,
Frith
,
C. D.
,
Frackowiak
,
R. S. J.
, &
Dolan
,
R. J.
(
1997
).
Brain regions supporting intentional and incidental memory: A PET study.
NeuroReport
,
8
,
1283
1287
.
Rugg
,
M. D.
,
Walla
,
P.
,
Schloerscheidt
,
A. M.
,
Fletcher
,
P. C.
,
Frith
,
C. D.
, &
Dolan
,
R. J.
(
1998
).
Neural correlates of depth of processing effects on recollection: Evidence from brain potential and positron emission tomography.
Experimental Brain Research
,
123
,
18
23
.
Salthouse
,
T. A.
(
2003
).
Memory aging from 18 to 80.
Alzheimer Disease and Associated Disorders
,
17
,
162
167
.
Salthouse
,
T. A.
,
Atkinson
,
T. M.
, &
Berish
,
D. E.
(
2003
).
Executive functioning as a potential mediator of age-related cognitive decline in normal adults.
Journal of Experimental Psychology: General
,
132
,
566
594
.
Schacter
,
D. L.
,
Savage
,
C. R.
,
Alpert
,
N. M.
,
Rauch
,
S. L.
, &
Albert
,
M. S.
(
1996
).
The role of hippocampus and frontal cortex in age-related memory changes: A PET study.
NeuroReport
,
7
,
1165
1169
.
Shannon
,
B. J.
, &
Buckner
,
R. L.
(
2004
).
Functional–anatomic correlates of memory retrieval that suggest non-traditional processing roles for multiple distinct regions within posterior parietal cortex.
Journal of Neuroscience
,
24
,
10084
10092
.
Stebbins
,
G. T.
,
Carrillo
,
M. C.
,
Dorfman
,
J.
,
Dirksen
,
C.
,
Desmond
,
J. E.
,
Turner
,
D. A.
,
et al
(
2002
).
Aging effects on memory encoding in the frontal lobes.
Psychology and Aging
,
17
,
44
55
.
Sullivan
,
E. V.
,
Marsh
,
L.
,
Mathalon
,
D. H.
,
Lim
,
K. O.
, &
Pfefferbaum
,
A.
(
1995
).
Age related decline in MRI volumes of temporal lobe gray matter but not hippocampus.
Neurobiology of Aging
,
16
,
591
606
.
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 and Cognition
,
64
,
1
6
.
Taconnat
,
L.
, &
Isingrini
,
M.
(
2004
).
Cognitive operations in the generation effect on a recall test: Role of aging and divided attention.
Journal of Experimental Psychology: Leaning, Memory, and Cognition
,
30
,
827
837
.
Tombaugh
,
T. N.
, &
Hubley
,
A. M.
(
2001
).
Rates of forgetting on three measures of verbal learning using retention intervals ranging from 20 min to 62 days.
Journal of the International Neuropsychological Society
,
7
,
79
91
.
Tulving
,
E.
,
Kapur
,
S.
,
Craik
,
F. I. M.
,
Moscovitch
,
M.
, &
Houle
,
S.
(
1994
).
Hemispheric encoding retrieval asymmetry in episodic memory: Positron emission tomography findings.
Proceedings of the National Academy of Sciences, U.S.A.
,
91
,
2016
2020
.
Uncapher
,
M. R.
, &
Rugg
,
M. D.
(
2005
).
Encoding and the durability of episodic memory: A functional magnetic resonance imaging study.
Journal of Neuroscience
,
25
,
7260
7267
.
van der Veen
,
F. M.
,
Nijhuis
,
F. A. P.
,
Tisserand
,
D. J.
,
Backes
,
W. H.
, &
Jolles
,
J.
(
2006
).
Effects of aging on recognition of intentionally and incidentally stored words: An fMRI study.
Neuropsychologia
,
44
,
2477
2486
.
Van Petten
,
C.
(
2004
).
Relationship between hippocampal volume and memory ability in healthy individuals across the lifespan: Review and meta-analysis.
Neuropsychologia
,
42
,
1394
1413
.
Van Petten
,
C.
,
Plante
,
E.
,
Davidson
,
P. S. R.
,
Kuo
,
T. Y.
,
Bajuscak
,
L.
, &
Glisky
,
E. L.
(
2004
).
Memory and executive function in older adults: Relationships with temporal and prefrontal gray matter volumes and white matter hyperintensities.
Neuropsychologia
,
42
,
1313
1335
.
Verhaeghen
,
P.
, &
Salthouse
,
T. A.
(
1997
).
Meta-analyses of age–cognition relations in adulthood: Estimates of linear and nonlinear age effects and structural models.
Psychological Bulletin
,
122
,
231
249
.
Viard
,
A.
,
Piolino
,
P.
,
Desgranges
,
B.
,
Chételat
,
G.
,
Lebreton
,
K.
,
Landeau
,
B.
,
et al
(
2007
).
Hippocampal activation for autobiographical memories over the entire lifetime in healthy aged subjects: An fMRI study.
Cerebral Cortex
,
17
,
2453
2467
.
Wagner
,
A. D.
,
Shannon
,
B. J.
,
Kahn
,
I.
, &
Buckner
,
R. L.
(
2005
).
Parietal lobe contributions to episodic memory retrieval.
Trends in Cognitive Sciences
,
9
,
445
453
.
Walhovd
,
K. B.
,
Fjell
,
A. M.
,
Reinvang
,
I.
,
Lundervold
,
A.
,
Fischl
,
B.
,
Quinn
,
B. T.
,
et al
(
2004
).
Size does matter in the long run. Hippocampal and cortical volume predict recall across weeks.
Neurology
,
63
,
1193
1197
.
Weis
,
S.
,
Klaver
,
P.
,
Reul
,
J.
,
Elger
,
C. E.
, &
Fernandez
,
G.
(
2004
).
Temporal and cerebellar brain regions that support both declarative memory formation and retrieval.
Cerebral Cortex
,
14
,
256
267
.
Wheeler
,
M. A.
, &
Buckner
,
R. L.
(
2004
).
Functional–anatomic correlates of remembering and knowing.
Neuroimage
,
21
,
1337
1349
.
Whiting
,
W. L.
, &
Smith
,
A. D.
(
1997
).
Differential age-related processing limitations in recall and recognition tasks.
Psychology and Aging
,
12
,
216
224
.
Yonelinas
,
A. P.
(
2002
).
The nature of recollection and familiarity: A review of 30 years of research.
Journal of Memory and Language
,
46
,
441
517
.
Yonelinas
,
A. P.
,
Otten
,
L. J.
,
Shaw
,
K. N.
, &
Rugg
,
M. D.
(
2005
).
Separating the brain regions involved in recollection and familiarity in recognition memory.
Journal of Neuroscience
,
25
,
2002
2008
.
Zimmerman
,
M. E.
,
Brickman
,
A. M.
,
Paul
,
R. H.
,
Grieve
,
S. M.
,
Tate
,
D. F.
,
Gunstad
,
J.
,
et al
(
2006
).
The relationship between frontal gray matter volume and cognition varies across the healthy adult lifespan.
American Journal of Geriatric Psychiatry
,
14
,
823
833
.