Abstract

Age-related declines in source memory have been observed for various stimuli and associated details. These impairments may be related to alterations in brain regions contributing to source memory via material-independent processes and/or regions specialized for processing specific materials. Using event-related functional magnetic resonance imaging, we investigate the effects of aging on source memory and associated neural activity for words and objects. Source accuracy was equally impaired in older adults for both materials. Imaging data revealed both groups recruited similar networks of regions to support source memory accuracy irrespective of material, including parietal and prefrontal cortices (PFC) and the hippocampus. Age-related decreases in material-independent activity linked to postretrieval monitoring were observed in right lateral PFC. Additionally, age-related increases in source accuracy effects were shown in perirhinal cortex, which were positively correlated with performance in older adults, potentially reflecting functional compensation. In addition to group differences in material-independent regions, age-related crossover interactions for material-dependent source memory effects were observed in regions selectively engaged by objects. These results suggest that older adults' source memory impairments reflect alterations in regions making material-independent contributions to source memory retrieval, primarily the lateral PFC, but may be further impacted by changes in regions sensitive to particular materials.

Introduction

Dual-process models of recognition memory suggest that previously experienced stimuli may be recognized either by recollection of contextual details about a prior episode or by familiarity for the stimulus in the absence of retrieval for contextual details (Mandler 1980; Yonelinas 2002). The contribution of these processes can be measured objectively, as with a “source memory” task, in which participants are asked to indicate which experimentally manipulated context or detail (e.g., color, encoding task, location, etc.) was associated with an item during its initial exposure (Johnson et al. 1993). Findings from numerous behavioral studies suggest that healthy older adults exhibit disproportionate impairments in source relative to item recognition, the latter of which can be supported by familiarity alone, for various types of materials (e.g., words, objects, neutral, and emotional) and source details (i.e., perceptual, conceptual, spatial, and temporal) (for reviews, see Spencer and Raz 1995; Yonelinas 2002; Kensinger 2009; Mitchell and Johnson 2009).

One potential explanation for these widespread deficits in source memory performance is that age-related changes may exist in brain areas that support source memory accuracy in a domain-general manner, such that altered processing would likely impair memory performance for various kinds of materials and associations. For example, functional magnetic resonance imaging (fMRI) studies have revealed that successful retrieval of source information is supported by neural activity in several regions including the medial temporal lobe (MTL), prefrontal cortex (PFC), and posterior parietal cortex (Davachi et al. 2003; Duarte et al. 2004; Ranganath et al. 2004; Kensinger and Schacter 2006; Diana et al. 2007; Vilberg and Rugg 2008; Mitchell and Johnson 2009 for review; Uncapher and Wagner 2009). Given the various types of stimuli and contexts that have been utilized in these studies, it is likely that these areas contribute to source recollection in a domain-general fashion (i.e., similar across item and source details). That is, some processes, such as the binding mechanisms attributed to the hippocampus, may contribute to successful recollection for a variety of details associated with various kinds of stimuli (Eichenbaum et al. 2007). There is evidence to suggest that, at retrieval, source memory may be supported strictly via material-independent processes for young adults (Duarte et al. 2011). Furthermore, previous aging studies have demonstrated that impairments in source memory accuracy may be related to changes in the lateral PFC at retrieval (Cabeza et al. 2002; Duarte et al. 2008; Rajah et al. 2009). Additionally, evidence has shown that changes in hippocampal activity at retrieval may also contribute to recollection deficits measured subjectively (Daselaar, Fleck, Dobbins, et al. 2006). However, as no previous aging studies have directly compared source memory-related activity as a function of material type, it remains unclear whether these and other age-related changes reflect dysfunction in domain-general processes, per se.

Another, nonmutually exclusive possibility for these seemingly ubiquitous source memory impairments is that age-related changes may be observed in several material-dependent processing regions. For example, a large region of extrastriate cortex called the lateral occipital complex (LOC) has been implicated in object-specific perception (Malach et al. 1995; Grill-Spector et al. 2001), as well as recollection of source details (location) about objects (Cansino et al. 2002; Sommer et al. 2005). Similarly, another region in the lateral fusiform gyrus is sensitive to visual, as opposed to auditory, presentation of word stimuli and may support processing of word forms (McCandliss et al. 2003) and recollection of word, as opposed to object, stimuli (Woodruff et al. 2005). Interestingly, age-related activity reductions in occipitotemporal cortical areas, including these extrastriate regions sensitive to object and word stimuli, have been shown in neuroimaging studies examining various kinds of cognitive tasks, including attention (Cabeza et al. 2004), visual perception (Iidaka et al. 2002; Davis et al. 2008), and episodic memory (Grady, McIntosh, and Craik 2003; Cabeza et al. 2004; Gutchess et al. 2005; Dennis et al. 2008). These results may suggest that older adults demonstrate source memory impairments for different kinds of stimuli in part because of less robust stimulus-specific representations.

To our knowledge, no previous aging fMRI studies have directly compared source memory activity for different categories of stimuli, and it is therefore not entirely clear whether source memory impairments commonly observed in older adults are related to dysfunction in material-independent, material-dependent processes or both. The current study was designed to address this issue. During study, participants made 1 of 2 semantic decisions about words or pictures of objects. At test, participants saw studied and unstudied words and objects, decided whether they had seen them previously, and judged in which source (semantic encoding task) the item was previously presented. Importantly, the source memory task included a response option that allowed participants to indicate that they did not know in which source context the recognized item was previously presented, as has been implemented in prior studies (Smith et al. 2004; Duarte et al. 2008, 2009; Duverne et al. 2008; Gottlieb et al. 2010). This procedure minimizes the potential dilution by guesses on source memory accuracy estimates and associated neural activity (Rugg and Morcom 2005); the impact of which may confound age-related differences in neural correlates of source memory accuracy (Li et al. 2004; Morcom et al. 2007; Duverne et al. 2008).

We hypothesized that:

  1. Material-independent source memory effects (accurate source trials vs. correctly rejected [CR] novel trials) were predicted in regions associated with domain-general recollection, including the hippocampus, posterior parietal cortex, and lateral PFC (see Wagner et al. 2005; Eichenbaum et al. 2007; Cabeza 2008; Mitchell and Johnson 2009 for reviews; Duarte et al. 2011). Furthermore, given the semantic nature of the encoding task (i.e., source), it was also likely that source memory effects would be observed in regions associated with controlled semantic retrieval such as the left ventrolateral and/or lateral temporal cortex (Dobbins and Wagner 2005; Badre and Wagner 2007).

  2. Older adults would show impairments in source memory accuracy for both words and objects, consistent with numerous previous studies (Raz 2000; Mitchell and Johnson 2009 for reviews). In correspondence with these impairments, material-independent old/new accuracy effects in the lateral PFC, and potentially the hippocampus, would be affected, consistent with previous evidence implicating alterations of both regions in age-related source memory deficits across different kinds of materials and source details implemented in these studies (Mitchell et al. 2006; Dennis et al. 2008; Duarte et al. 2008; Rajah et al. 2009). Age-related dysfunction in source memory activity in the lateral PFC would be consistent with the so-called “frontal aging hypothesis” (for reviews, see West 1996; Raz 2000), which suggests that the PFC, both structurally and functionally, is disproportionately affected by normal aging. Given that the lateral PFC has been implicated in the monitoring and evaluation of episodic information retrieved from MTL structures (Fletcher and Henson 2001; Simons and Spiers 2003; Dobbins and Han 2006) and selection of a subset of retrieved details in the face of competing alternatives (reviewed in Badre and Wagner 2007), processes that most certainly support source memory decisions, age-related alterations in the lateral PFC would seem a likely contributor to older adults' source memory impairments.

  3. Given emerging fMRI and event-related potential (ERP) evidence suggesting that the neural correlates of recollection are, in part, sensitive to the type of stimulus materials (Khader, Burke, et al. 2005; Khader, Heil, and Rosler 2005; Woodruff et al. 2005; Yick and Wilding 2008), source memory effects may be, at least in part, material-dependent. Specifically, anterior MTL, including the perirhinal cortex (PrC), and LOC regions previously implicated in object perception, recognition memory and source memory for objects (Malach et al. 1995; Grill-Spector et al. 2001; Cansino et al. 2002; Buckley and Gaffan 2006 for review; Lee et al. 2006; Barense et al. 2007, 2009; Awipi and Davachi 2008) and extrastriate (fusiform) and left-lateralized frontotemporal regions associated with visual word processing, including word reading, and word recollection (Fiez and Peterson 1998; Price 2000; Roskies et al. 2001; McCandliss et al. 2003; Woodruff et al. 2005) may demonstrate material-dependent source memory effects. To the extent that age-related changes exist in at least some of these material-sensitive regions (Iidaka et al. 2002; Cabeza et al. 2004; Davis et al. 2008), source memory accuracy may be impaired for words and objects, in part, because of weak or less differentiated material-dependent perceptual representations (Park et al. 2004; Gutchess et al. 2005; Chee et al. 2006; Dennis et al. 2008).

Materials and Methods

Participants

Sixteen young adults (YA) between 18 and 32 years of age and 14 older adults (OA) between 61 and 74 years of age were recruited from local universities, science and health fairs, the Georgia Institute of Technology alumni association and community solicitation. Two additional older adults were excluded for inadequate neuropsychological test performance, scoring more than 2 standard deviations below the age-adjusted norms for several tests. All participants were right-handed, native English speakers, with normal or corrected-to-normal vision (using MRI-compatible glasses when necessary). None reported a history of psychiatric or neurological disorders (e.g., stroke, epilepsy, multiple sclerosis, etc.), vascular disease, or psychoactive drug use. None of the participants were taking CNS-active medications or antihypertensive medications. All participants were paid $10 an hour for their time and signed consent forms approved by the Georgia Institute of Technology Institutional Review Board. Group characteristics are shown in Table 1. Groups did not differ for years of education or gender (P's > 0.43).

Table 1

Group characteristics

Measure YA (n = 16) OA (n = 14) 
Age 24.13 (4.00) 65.86 (3.86) 
Gender 9/16 female 6/14 female 
Education 17.28 (2.90) 18.0 (1.75) 
Letter fluency 51.62 (14.25) 57.21 (16.79) 
List recall (immediate) 11.25 (1.00) 10.93 (1.33) 
List recall (immediate, cued) 11.56 (0.89) 11.00 (1.11) 
List recall (delayed) 11.69 (0.87) 11.43 (0.76) 
List recall (delayed, cued) 11.81 (0.54) 11.43 (0.76) 
List recognition 12.00 (0.00) 12.00 (0.00) 
MAS digit span forward 6.69 (1.25) 6.57 (1.16) 
MAS digit span backward 5.81 (1.17) 4.71 (0.91)* 
Trails A (in seconds) 18.91 (5.22) 31.27 (9.11)* 
Trails B (in seconds) 40.12 (8.47) 67.91 (26.25)* 
Visual recognition 19.25 (1.18) 16.43 (1.55)* 
Delayed visual recognition 19.19 (1.04) 18.29 (1.59) 
Visual reproduction 9.13 (1.31) 5.64 (2.44)* 
Measure YA (n = 16) OA (n = 14) 
Age 24.13 (4.00) 65.86 (3.86) 
Gender 9/16 female 6/14 female 
Education 17.28 (2.90) 18.0 (1.75) 
Letter fluency 51.62 (14.25) 57.21 (16.79) 
List recall (immediate) 11.25 (1.00) 10.93 (1.33) 
List recall (immediate, cued) 11.56 (0.89) 11.00 (1.11) 
List recall (delayed) 11.69 (0.87) 11.43 (0.76) 
List recall (delayed, cued) 11.81 (0.54) 11.43 (0.76) 
List recognition 12.00 (0.00) 12.00 (0.00) 
MAS digit span forward 6.69 (1.25) 6.57 (1.16) 
MAS digit span backward 5.81 (1.17) 4.71 (0.91)* 
Trails A (in seconds) 18.91 (5.22) 31.27 (9.11)* 
Trails B (in seconds) 40.12 (8.47) 67.91 (26.25)* 
Visual recognition 19.25 (1.18) 16.43 (1.55)* 
Delayed visual recognition 19.19 (1.04) 18.29 (1.59) 
Visual reproduction 9.13 (1.31) 5.64 (2.44)* 

Note: Standard deviations in parentheses. All neuropsychological test scores are reported as raw scores. * = significantly different from Young (P < 0.05).

Neuropsychological Testing

All participants were administered a battery of standardized neuropsychological tests either directly following MRI scanning or within a few days of their scan. Tests sensitive to prodromal cognitive deficits were used to ensure that age-related differences in task performance were not due to older adults being in the early stages of dementia. The battery included tests of working and long-term memory, executive function, and visuospatial ability. Specifically, subtests from the Memory Assessment Scale test battery (Williams 1991) including: digit span forward and backward, list learning, word recognition, immediate word recall, delayed word recall, object recognition, object recall, object reproduction, and delayed object recognition; the Trail Making Tests A and B (Reitan and Wolfson 1985) and the Controlled Oral Word Association Test (Benton et al. 1983). Results are shown in Table 1.

Materials

Two hundred and forty words were selected from the Medical Research Council Psycholinguistic Database (http://www.psy.uwa.edu.au/mrcdatabase/uwa_mrc.htm). All words were noun referents of objects. Words between 4 and 8 letters, with a frequency of 10–50 per million in the English language and imageability ratings of between 500 and 700 were included (Kucera and Francis 1967). Two hundred and forty grayscale photographs of namable objects taken from the Hemera Technologies Photo-Objects DVDs were also used. All objects appeared on a white background. No word was also displayed as an object, and no object was also displayed as a word (i.e., the word “apple” and a picture of an apple). All stimuli subtended a maximum vertical and horizontal visual angle of up to 7.1.

Procedure

A practice version of the experiment was administered to participants outside of the scanner immediately prior to scanning. Participants were given an opportunity to practice until they felt comfortable with the tasks and could respond accurately within the allotted amount of time. Both study and test periods were scanned but only data from the test period are presented here. Stimuli were counterbalanced across participants such that each word and object served as both a studied and an unstudied item in the current experiment.

Each of the 3 study blocks, administered within a single scanning session, consisted of 80 trials: 40 objects and 40 words, presented in a pseudorandom order, with the condition that no more than 5 trials of the same type were presented sequentially. Study items were centrally presented for 3 s followed by a 500 ms central fixation cross. For each trial at study, participants were cued to make 1 of 2 semantic yes/no decisions. In one task, participants were asked to decide whether the item was living (i.e., the “Living?” task), and in the other task, they were asked to decide if the item would fit inside of a shoebox (i.e., the “Shoebox?” task). For half of the participants in each age group, “yes” and “no” responses were made with the index fingers of the left and right hands, respectively. Response mapping was counterbalanced across participants.

Study was followed immediately by 3 blocks of test, within the same scanning session as study. Each of the test blocks consisted of 160 trials including 80 studied and 80 unstudied items; with equal numbers of words and objects, presented in a pseudorandom order, with the condition that no more than 5 trials of the same type were presented sequentially. Study items were centrally presented for 4 s followed by a 500 ms central fixation cross. For each trial, participants were asked to indicate, by pressing 1 of 4 buttons, whether the item was a) previously studied and associated with the Living? task, b) previously studied and associated with the Shoebox? task, c) previously studied but the participant could not recollect the task in which the item was presented (don't know), or d) new and not previously studied. The don't know option was offered to reduce potential contamination by guessing in the source decision (Smith et al. 2004; Duarte et al. 2008, 2009; Duverne et al. 2008; Gottlieb et al. 2010). Participants responded with their left middle finger, left index finger, right index finger, and right middle finger for choices a–d, respectively. As with study, the order of these blocks and the response mapping were counterbalanced across participants. The Huynh–Feldt correction, reflected in the P values, was used in the behavioral analyses, where appropriate. Mixed-design analysis of variance (ANOVAs) and 2-tailed t-tests were used to analyze the neuropsychological and behavioral data.

fMRI Acquisition

Scanning was performed on a 3-T Siemens TIM Trio system. Functional data were acquired using a gradient echo pulse sequence (32 transverse slices oriented along the anterior–posterior commissural axis with a 30 degree upward tilt to avoid the eyes, repetition time of 2 s, echo time of 30 ms, 3 × 3 × 3.5 mm voxels, 0.8-mm interslice gap). Three study blocks of 155 volumes and 3 tests blocks of 375 volumes were acquired. The first 5 volumes of each block were discarded to allow for equilibration effects. A high-resolution T1-weighted magnetization-prepared rapid acquisition gradient echo (MPRAGE) image was collected for normalization (see below).

fMRI Analysis

Data were analyzed with SPM8 (SPM8, http://www.fil.ion.ucl.ac.uk/spm/software/spm8/). Images were corrected for differences in slice timing acquisition using the middle slice of each volume as the reference, spatially realigned, and resliced with respect to the first volume of the first block. Each participant's MPRAGE scan was coregistered to the mean echo planar imaging (EPI), produced from spatial realignment. Each coregistered structural scan was then segmented using the diffeomorphic anatomical registration through exponentiated lie algebra (DARTEL) SPM 8 toolbox (Ashburner 2007) (DARTEL is a suite of tools fully integrated with SPM 8, which the SPM 8 manual recommends over optimized normalization, to achieve sharper nonlinear registration, for intersubject alignment. This method also achieves better localization of fMRI activations in Montreal Neurological Institute [MNI] space. This method has been used successfully in several previous studies with various healthy and neurological populations (Yassa and Stark 2009; Pereira et al. 2010). Comparison of the current results produced by DARTEL with the standard unified segmentation normalization procedure revealed largely similar patterns of activity for the contrasts assessed in this study.). Briefly, the gray and white matter segmented images were used to create a study-specific template using the DARTEL toolbox and the flow fields containing the deformation parameters to this template for each subject were used to normalize each participant's realigned and resliced EPIs to MNI space. Normalized EPI images were written to 2 × 2 × 2 mm and smoothed with an 8 mm full-width at half-maximum isotropic Gaussian kernel. The EPI data were then high-pass filtered to a maximum of 1/128 Hz and grand mean scaled to 100.

Statistical analysis was performed in 2 stages. In the first stage, neural activity was modeled by a sequence of delta functions at onset of the various event types and convolved with a canonical hemodynamic response function. The time courses were down sampled to the middle slice to form the covariates for the general linear model. For each participant and block, 6 covariates representing residual movement-related artifacts, determined by the spatial realignment step, were included in the first-level model to capture residual (linear) movement artifacts. Voxel-wise parameter estimates for these covariates were obtained by restricted maximum-likelihood estimation, using a temporal high-pass filter (cutoff 128 s) to remove low-frequency drifts and modeling temporal autocorrelation across scans with an AR(1) process.

Contrasts of the parameter estimates for each participant were submitted to the second stage of analysis (treating participants as a random effect). A mixed ANOVA model was created for the test period that allowed us to examine both within group effects and group interactions. The 3 × 2 × 2 model included factors of Condition (source correct [SC], source incorrect/don't know [SINCDK] source, CR unstudied item), Material (words, objects), and Group (young, old). Incorrect source and don't know source trial types were combined to form a category representing instances where the item was recognized but correct source information was unavailable, as has been done previously (Duarte et al. 2008, 2009; Duverne et al. 2008; Gottlieb et al. 2010). There were insufficient numbers of incorrect responses to studied (misses) and unstudied (false alarms) items for all participants to examine separately and so they were not included in the ANOVA. Memory decisions were collapsed across the study task (living/shoebox). Covariates modeling the mean across conditions for each participant were also added to each model, to remove between-subject variance of no interest. A weighted least squares estimation procedure was used to correct for inhomogeneity of covariance across within-group conditions and inhomogeneity of variance across groups.

The SPM for the main effect of Condition (across groups) was masked exclusively with the SPMs for the Group-by-Condition (across material), Condition-by-Material (within group), and Group-by-Condition-by-Material interactions, using a liberal uncorrected threshold of P < 0.05 for the masks in order to restrict memory effects to those “common” (i.e., similar size) across material types and groups (Note that a liberal threshold for an exclusive mask is more conservative in excluding regions from the masked SPM.). The SPM for the main effect of Material (across conditions and groups) was similarly exclusively masked Group-by-Material interactions. Inclusive masks were applied to determine the overlap between these regions associated with material-dependent processing (regardless of memory judgment) and material-dependent memory effects. Inclusive masking was applied using an uncorrected threshold of P < 0.01 for the mask. All masked as well as unmasked contrasts were evaluated using 2-tailed T-contrasts under an uncorrected alpha level of 0.001 (each direction of t-test at 0.0005) and a minimum cluster size of 5 contiguous voxels. In addition to these whole-brain analyses, we conducted regions of interest (ROIs) analyses using regions from prior studies that had clear anatomical delineation and about which we had a priori hypotheses, specifically the hippocampus and parahippocampal cortex. ROI analyses were examined using a family-wise error corrected threshold of P < 0.05, using bilateral masks from the automatic anatomical labeling of the MNI brain and the small-volume correction (SVC). The SVC approach reveals peak voxels within the ROI masks that are reliable for the effect of interest, thus, peak voxels that survived the corrected threshold are reported (Cluster size is not easily reported for the MTL ROIs given that the SVC cluster size [k] equals the SVC search volume in voxels associated with the MTL masks.). These ROI analyses were subjected to the same inclusive and exclusive masking procedures as described above for the whole-brain analyses.

For both whole-brain and ROI analyses, simple effect SPMs were performed to elucidate the source of interactions (e.g., Young > Old: SC > SINCDK) and to ensure that main effects were reliable for each group and were conducted using the same whole-brain or SVC procedure (for MTL regions). Importantly, given that these simple effect comparisons for a particular region were made independently to the initial contrast, they were not statistically biased.

Maxima of significant clusters were localized on individual normalized structural images. Neural activity from these maxima was plotted for SC, SINCDK, and CR conditions. Neural activity reflected the parameter estimates for the convolved regressors and had arbitrary units.

Results

Neuropsychological Test Results

Group characteristics and raw scores for neuropsychological tests are shown for both groups in Table 1. While the young group tended to exhibit numerically better performance for most of the tests, performance on Trails A & B, digit span backwards, object recognition, and object reproduction were the only tests that significantly differed between groups (t's > 2.839, P's < 0.008).

Behavioral Results

The mean proportion of correct, incorrect, and don't know source judgments (SCs) to studied items and of new judgments made to studied (misses) and unstudied items (correct rejections [CRs]) and corresponding reaction times (RTs) are shown for both groups in Table 2, along with all performance indices. An ANOVA of correction rejection rates employing factors of Material (words, objects) and Group revealed a main effect of Material (F1,28 = 25.60, P < 0.001), confirming higher CR rates for objects than words for both groups and a main effect of Group (F1,28 = 4.50, P = 0.04), showing that young adults had a higher CR rates than older adults. Given the reduced CR rates, or higher false alarm rates, in the older adults, we wanted to determine whether response biases were more liberal in older than in young adults. To this end, we calculated Br estimates of bias for each material type and group according to: Br = p(f1alse alarms)/(1 − (p(hits) − p(false alarms))), after adjusting hit and false alarm rates according to the formula ((number of hits or false alarms, respectively, + 0.5)/(number of old or new items, respectively, + 1)) (see Snodgrass and Corwin 1988). These estimates are shown in Table 2. An ANOVA of these Br estimates employing factors of Material (words, objects) and Group (young, old) revealed a marginal main effect of Material (F1,28 = 4.28, P = 0.05) and a marginal interaction (F1,28 = 3.63, P = 0.07). The main effect reflects the more liberal response bias for words than for objects across groups while the interaction confirms that older adults were more liberal than the young in their responses to objects (t28 = 2.24, P = 0.03) but not words (t28 = 0.30, P = 0.77).

Table 2

Response rates, performances indices, and response times to studied and unstudied items at test

 YA
 
OA
 
 Words Objects Words Objects 
Response rates     
    Studied items     
        Correct 0.53 (0.15) 0.51 (0.14) 0.46 (0.13) 0.51 (0.12) 
        Incorrect 0.05 (0.03) 0.07 0 (.05) 0.13 (0.05) 0.19 (0.09) 
        Don't know 0.22 (0.11) 0.25 (0.15) 0.16 (0.17) 0.14 (0.17) 
        Miss 0.20 (0.11) 0.17 (0.08) 0.25 (0.12) 0.16 (0.08) 
    Unstudied items     
        CRs 0.84 (0.11) 0.93 (0.06) 0.79 (0.11) 0.85 (0.10) 
Performance indices     
    Item recognition (Pr-Item) 0.63 (0.11) 0.76 (0.08) 0.54 (0.12) 0.68 (0.12) 
    Corrected source (Pr-Source) 0.82 (0.08) 0.76 (0.18) 0.56 (0.11) 0.50 (0.18) 
    Source recognition (Psr) 0.48 (0.15) 0.42 (0.16) 0.38 (0.12) 0.35 (0.15) 
    Bias (Br) 0.44 (0.26) 0.29 (0.22) 0.46 (0.18) 0.46 (0.19) 
Response times     
    Studied items     
        Correct 1739 (248) 1739 (298) 2305 (468) 2201 (523) 
        Incorrect 2134 (383) 2010 (405) 2501 (530) 2386 (483) 
        Don't know 2220 (429) 2166 (472) 2588 (438) 2268 (534) 
        Miss 1777 (420) 1637 (296) 2040 (550) 1947 (598) 
Unstudied items     
    CRs 1481 (330) 1264 (240) 1782 (552) 1604 (447) 
 YA
 
OA
 
 Words Objects Words Objects 
Response rates     
    Studied items     
        Correct 0.53 (0.15) 0.51 (0.14) 0.46 (0.13) 0.51 (0.12) 
        Incorrect 0.05 (0.03) 0.07 0 (.05) 0.13 (0.05) 0.19 (0.09) 
        Don't know 0.22 (0.11) 0.25 (0.15) 0.16 (0.17) 0.14 (0.17) 
        Miss 0.20 (0.11) 0.17 (0.08) 0.25 (0.12) 0.16 (0.08) 
    Unstudied items     
        CRs 0.84 (0.11) 0.93 (0.06) 0.79 (0.11) 0.85 (0.10) 
Performance indices     
    Item recognition (Pr-Item) 0.63 (0.11) 0.76 (0.08) 0.54 (0.12) 0.68 (0.12) 
    Corrected source (Pr-Source) 0.82 (0.08) 0.76 (0.18) 0.56 (0.11) 0.50 (0.18) 
    Source recognition (Psr) 0.48 (0.15) 0.42 (0.16) 0.38 (0.12) 0.35 (0.15) 
    Bias (Br) 0.44 (0.26) 0.29 (0.22) 0.46 (0.18) 0.46 (0.19) 
Response times     
    Studied items     
        Correct 1739 (248) 1739 (298) 2305 (468) 2201 (523) 
        Incorrect 2134 (383) 2010 (405) 2501 (530) 2386 (483) 
        Don't know 2220 (429) 2166 (472) 2588 (438) 2268 (534) 
        Miss 1777 (420) 1637 (296) 2040 (550) 1947 (598) 
Unstudied items     
    CRs 1481 (330) 1264 (240) 1782 (552) 1604 (447) 

Note: Standard deviations in parentheses.

Item recognition accuracy was estimated by the Pr (henceforth “Pr-Item”) measure of discriminability (p(hits) – p(false alarms)) (Snodgrass and Corwin 1988). Source recognition accuracy was estimated by Pr (henceforth “Pr-Source”) as well, excluding “don't know” responses (Pr = p(correct) − p(incorrect)), in replication of our previous studies (Duarte et al. 2009, 2011). Additionally, we have also included Psr estimates, derived from a single high threshold model (Snodgrass and Corwin 1988) consistent with other previous aging studies (e.g., Duverne et al. 2008); where Psr = (p(correct) − 0.5(1 − p(don't know)))/(1 − (0.5 (1 − p(don't know)))). While Pr-Source provides an index of a participant's accuracy for choosing the correct source out of all their source attempts, Psr provides an index of a participant's accuracy for choosing the correct source where the contribution of lucky guesses on source decisions has been removed. Thus, a participant with a conservative bias who does not often “guess” on the source decision opting instead for don't know responses may have a high Pr-Source and but a low Psr estimate. Importantly, all item and source accuracy estimates for each material type and group were significantly greater than chance (0%) (t's > 10.2, P's < 0.001). These accuracy estimates are shown in Table 2.

A Material (words, objects) × Group ANOVA for item memory estimates revealed main effects of Group (F1,28 = 5.93, P = 0.02) and Material (F1,28 = 54.6, P < 0.001). These results showed that item recognition was greater for objects than for words for both groups and greater for young than older adults for both words and objects. A similar ANOVA was conducted for Pr-Source estimates of source memory accuracy. This ANOVA revealed main effects of Group (F1,28 = 41.7, P < 0.001) and Material (F1,28 = 4.86, P = 0.04). In contrast to item memory, source memory accuracy was greater for words than for objects in both age groups but impaired in older adults relative to the young overall. The same ANOVA was conducted for Psr estimates and revealed marginal main effects of Group (F1,28 = 3.14, P = 0.09) and Material (F1,28 = 3.70, P = 0.06), suggesting similar, though less robust, source memory accuracy differences to those described above for Pr-Source estimates. As can be seen in Table 2, the group differences in source memory accuracy were less pronounced for Psr than Pr-Source measures because Psr estimates were substantially lower than Pr-Source estimates for the young, due to the increased tendency of young adults to use the “don't know source” response option.

Mean RTs for the recognition judgments made for studied and unstudied items at test are shown in Table 2. A Material × Response (SC, SINCDK source, CRs) × Group ANOVA for these RTs yielded main effects of Group (F1,28 = 9.32, P = 0.005), Material (F1,28 = 29.44, P < 0.001), and Response (F2,56 = 78.59, P < 0.001). As shown in the table, the main effect of Group indicates that older participants were slower to respond to all items than young adults while the main effect of material reflects the generally faster RTs for objects than for words across groups. Pairwise comparisons revealed that SC responses were faster than SINCDK responses (t's > 2.18, P's < 0.05) and that CR responses were faster than SC responses (t's > 3.05, P's < 0.008) for each material type and age group.

fMRI Results

To identify brain regions related to source memory, we first examined “old/new effects” via a contrast between studied items associated with correct SCs and new items that were CR consistent with some previous studies (Morcom et al. 2007; Duverne et al. 2008). We present neural activity that was 1) common to both groups and material types; 2) common to both material types but different between the groups; 3) common to both groups but different between material types; and 4) different between groups and material type, with common activity defined using exclusive masking (see Materials and Methods). An overview of the material-independent old/new effects (both common to and different between groups) may be seen in Figure 1. Additionally, because the contrast between SC and CR trials although robust, likely reflects source recollection in addition to familiarity-based recognition memory more generally, we also assessed recollection-specific effects via a contrast between studied items associated with correct SCs and those associated with incorrect or don't know SCs (SINCDK). Results are described as follows: 1) regions showing material-independent old/new effects, common to and different between groups; 2) regions showing material sensitivity regardless of memory judgment, common to and different between groups; 3) regions showing material-dependent old/new effects common to and different between groups; and 4) regions showing recollection-specific activity common to/different between group.

Figure 1.

Old/new effects, (SC vs. CR). Activity common to groups and age-related interactions are shown. Select regions are displayed on the 3D rendered MNI reference brain (P < 0.0005, uncorrected, with a 5 voxel extent; exclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Figure 1.

Old/new effects, (SC vs. CR). Activity common to groups and age-related interactions are shown. Select regions are displayed on the 3D rendered MNI reference brain (P < 0.0005, uncorrected, with a 5 voxel extent; exclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Material-Independent Old/New Effects

Common to Groups

Several regions, including bilateral portions of the parietal cortex (inferior parietal lobule), lateral PFC, as well as left lateral temporal cortex, exhibited greater activity for SC than CR items, across material types (Table 3 and Fig. 2). In contrast, a few regions, most notably bilateral portions of the anterior hippocampus, showed greater activity for CR than SC items (Table 3 and Fig. 2).

Table 3

Regions showing material-independent old/new effects common to both age groups

Contrast Region L/R MNI coordinates (x, y, zBA T score Cluster size 
SC > CR Middle occipital cortex −34, −64, 40 11.40 5636 
Inferior parietal lobule 4060, 50 39 11.04  
Angular gyrus 38, −60, 44 39 6.69 564 
Inferior frontal gyrus −50, 32, 20 45 8.48 2799 
Middle frontal gyrus 44, 10, 40 44 5.03 59 
 42, 34, 22 45 4.63 89 
 30, 8, 60 4.34 18 
Superior frontal gyrus 32, −4, 64 3.65 
Supplemental motor area −4, 22, 50 7.50 609 
Superior medial frontal gyrus 6, 30, 44 8 7.44  
Precentral gyrus 26, −8, 56 4.18 28 
Middle cingulate −2, −20, 32 23 5.04 
Precuneus −14, −66, 64  7.12 12 
Insula 32, 26, 0 47 6.51 208 
Inferior temporal cortex −58, −48, −10 37 5.63 269 
Cerebellum −32, −44, −36  4.25 24 
CR > SC 
Anterior hippocampus −24, −8, −18 20/35 5.25 46 
 26, −6, −18 20 4.81 133 
Superior medial frontal cortex 8, 58, 30 10 4.57 107 
 −8, 52, 18 32 3.56 
Superior temporal cortex 54, −42, 20 42 4.54 409 
 −60, −4, 8 48 3.60 40 
Supramarginal gyrus −66, −26, 22 48 4.22 39 
Middle temporal cortex −62, −10, −16 21 4.18 117 
 54, −16, −8 22 4.09 88 
 Rolandic operculum 52, −4, 12 48 3.82 
Contrast Region L/R MNI coordinates (x, y, zBA T score Cluster size 
SC > CR Middle occipital cortex −34, −64, 40 11.40 5636 
Inferior parietal lobule 4060, 50 39 11.04  
Angular gyrus 38, −60, 44 39 6.69 564 
Inferior frontal gyrus −50, 32, 20 45 8.48 2799 
Middle frontal gyrus 44, 10, 40 44 5.03 59 
 42, 34, 22 45 4.63 89 
 30, 8, 60 4.34 18 
Superior frontal gyrus 32, −4, 64 3.65 
Supplemental motor area −4, 22, 50 7.50 609 
Superior medial frontal gyrus 6, 30, 44 8 7.44  
Precentral gyrus 26, −8, 56 4.18 28 
Middle cingulate −2, −20, 32 23 5.04 
Precuneus −14, −66, 64  7.12 12 
Insula 32, 26, 0 47 6.51 208 
Inferior temporal cortex −58, −48, −10 37 5.63 269 
Cerebellum −32, −44, −36  4.25 24 
CR > SC 
Anterior hippocampus −24, −8, −18 20/35 5.25 46 
 26, −6, −18 20 4.81 133 
Superior medial frontal cortex 8, 58, 30 10 4.57 107 
 −8, 52, 18 32 3.56 
Superior temporal cortex 54, −42, 20 42 4.54 409 
 −60, −4, 8 48 3.60 40 
Supramarginal gyrus −66, −26, 22 48 4.22 39 
Middle temporal cortex −62, −10, −16 21 4.18 117 
 54, −16, −8 22 4.09 88 
 Rolandic operculum 52, −4, 12 48 3.82 

L, left; R, right; B, bilateral; BA, Brodmann's area. Regions in italics represent subclusters.

Figure 2.

Material-independent old/new effects. Activity common to groups and age-related interactions are shown. Select regions are displayed on the MNI reference brain. Plots show parameter estimates for the event-related response at the peak maxima of the selected regions for SC and CR trial types. Error bars depict standard error of the mean across participants, for each group. (P < 0.0005, uncorrected, with a 5 voxel extent; exclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Figure 2.

Material-independent old/new effects. Activity common to groups and age-related interactions are shown. Select regions are displayed on the MNI reference brain. Plots show parameter estimates for the event-related response at the peak maxima of the selected regions for SC and CR trial types. Error bars depict standard error of the mean across participants, for each group. (P < 0.0005, uncorrected, with a 5 voxel extent; exclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Group Differences

Group differences were observed in several regions showing greater activity for SC than for CR items in the old only, the majority of which displayed no reliable effects in the young, including lateral and medial parietal cortex (see Table 4 and Fig. 2). However, the largest precuneus region (−4, −45, 48; BA 7) showed a crossover interaction where activity in young adults displayed a pattern of CR > SC. Furthermore, one region where CR > SC activity was observed in the young but there was no reliable effect in the old was the left medial orbitofrontal cortex (Table 4 and Fig. 2).

Table 4

Regions showing material-independent old/new effects different between age groups

Contrast Region L/R MNI coordinates (x, y, zBA T score Cluster size 
OA > YA       
 SC > CR Superior parietal cortex 18, −74, 56 4.16 80 
Superior frontal gyrus −18, 10, 60 4.06 97 
Precuneus −4, −46, 48 3.93 122 
 −10, −58, 62 3.75 49 
 6, −66, 60 3.53 
Precentral gyrus 28, −6, 48 3.88 25 
Supplemental motor area −14, −8, 64 3.76 26 
YA > OA       
    CR > SC Medial orbitofrontal gyrus −10, 38, −12 11 3.75 25 
Contrast Region L/R MNI coordinates (x, y, zBA T score Cluster size 
OA > YA       
 SC > CR Superior parietal cortex 18, −74, 56 4.16 80 
Superior frontal gyrus −18, 10, 60 4.06 97 
Precuneus −4, −46, 48 3.93 122 
 −10, −58, 62 3.75 49 
 6, −66, 60 3.53 
Precentral gyrus 28, −6, 48 3.88 25 
Supplemental motor area −14, −8, 64 3.76 26 
YA > OA       
    CR > SC Medial orbitofrontal gyrus −10, 38, −12 11 3.75 25 

L, left; R, right; B, bilateral; BA, Brodmann's area.

Given the “overrecruitment” of these parietal regions by older adults compared with the young, we wanted to find out if there was a direct relationship between the size of these old/new effects and our 3 measures of behavioral performance (Pr-Item, Pr-Source, and Psr), which would be consistent with theories of functional compensation in older adults (Rajah and D'Esposito 2005). Activity from the peak voxel coordinates for these parietal regions (Table 4) were selected and correlational analyses were conducted with the various behavioral measures. Results showed that these regions were not significantly correlated with any measure of behavioral performance (r's < 0.22, P's > 0.45). It should be noted that the sample size in the present study is relatively small, thus the lack of significant correlations may simply be the product of low power.

Material Effects

Common to Groups

Prior to exploring material-dependent memory-related activity, we first established evidence of material-sensitive processing effects by contrasting the average of the mean event-related responses for all studied items correctly recognized, regardless of source memory judgment, and CR unstudied items, for each material type. Our intention with these analyses was merely to verify that the word and object stimuli used in the current study elicited activity in the a priori predicted brain regions discussed in the introduction. The overlap between these material-sensitive effects and material-dependent source memory effects was assessed via inclusive masking (see Materials and Methods) to determine whether a subset of regions sensitive to word and object stimuli also demonstrate source accuracy effects for these stimuli (see below).

Regions exhibiting greater activity for objects included bilateral occipitotemporal cortex and bilateral anterior MTL, including the hippocampi, amygdala, and PrC. Greater activity for words than for objects was found in bilateral middle temporal gyri and left ventrolateral PFC (see Fig. 3).

Figure 3.

Regions exhibiting differential activity for words and objects, collapsed across memory judgment, are displayed on the 3D rendered MNI reference brain. Activity common to both groups and activity showing age-related interactions are shown (P < 0.0005, uncorrected, with a 5 voxel extent; exclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Figure 3.

Regions exhibiting differential activity for words and objects, collapsed across memory judgment, are displayed on the 3D rendered MNI reference brain. Activity common to both groups and activity showing age-related interactions are shown (P < 0.0005, uncorrected, with a 5 voxel extent; exclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Group Differences

Young adults showed greater object-specific activity in bilateral fusiform gyri and the right middle occipital gyrus and greater word-specific activity in the left middle temporal gyrus than did older adults (see Fig. 3).

Material-Dependent Old/New Effects

Common to Groups

In general, few regions exhibited material-dependent memory activity. No regions showed old/new activity that was greater for words than for objects. Only one region, the right cuneus (see Table 5), showed activity that was greater for objects than words, (SC > CR items), but inclusive masking revealed this did not overlap with the object processing regions shown in Figure 3.

Table 5

Regions showing material-dependent old/new effects common to both age groups

Contrast Region L/R MNI coordinates (x, y, zBA T score Cluster size 
Common to Groups       
    Objects > Words       
        SC > CR Cuneus 12, −68, 34  4.00 43 
Contrast Region L/R MNI coordinates (x, y, zBA T score Cluster size 
Common to Groups       
    Objects > Words       
        SC > CR Cuneus 12, −68, 34  4.00 43 

L, left; R, right; B, bilateral; BA, Brodmann's area.

Group Differences

A few regions, including the right inferior frontal gyrus and bilateral portions of the middle occipital cortex, within the LOC, showed reliable crossover effects in which young adults showed SC > CR effects for objects but not words, while older adults showed SC > CR effects for words but not objects (see Table 6). Furthermore, inclusive masking confirmed that these regions overlapped with object processing regions shown in Figure 3. The overlap of these old/new effects and object processing effects can be seen in Figure 4.

Table 6

Regions showing material-dependent old/new effects different between age groups

Contrast Region L/R MNI coordinates (x, y, zBA T score Cluster size 
YA > OA       
    Objects > Words       
        SC > CR Inferior frontal gyrus 48, 24, 30 48 4.29 74 
Precentral gyrus −56, 14, 34 44 3.86 18 
 −44, −4, 40 3.61 
Middle occipital cortex 34, −76, 38 19 3.59 16 
 −28, −74, 36 19 3.49 
Insula −34, 18, 8 48 3.46 
Contrast Region L/R MNI coordinates (x, y, zBA T score Cluster size 
YA > OA       
    Objects > Words       
        SC > CR Inferior frontal gyrus 48, 24, 30 48 4.29 74 
Precentral gyrus −56, 14, 34 44 3.86 18 
 −44, −4, 40 3.61 
Middle occipital cortex 34, −76, 38 19 3.59 16 
 −28, −74, 36 19 3.49 
Insula −34, 18, 8 48 3.46 

L, left; R, right; B, bilateral; BA, Brodmann's area.

Figure 4.

Material-dependent old/new effects. Regions more sensitive to objects than words (see Fig. 3) are also shown to highlight the overlap between perceptual processing and memory effects. Select regions are displayed on the MNI reference brain. Plots show parameter estimates for the event-related response at the peak maxima of the selected regions for SC and CR trial types. Error bars depict standard error of the mean across participants, for each group. (P < 0.0005, uncorrected, with a 5 voxel extent; inclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Figure 4.

Material-dependent old/new effects. Regions more sensitive to objects than words (see Fig. 3) are also shown to highlight the overlap between perceptual processing and memory effects. Select regions are displayed on the MNI reference brain. Plots show parameter estimates for the event-related response at the peak maxima of the selected regions for SC and CR trial types. Error bars depict standard error of the mean across participants, for each group. (P < 0.0005, uncorrected, with a 5 voxel extent; inclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Recollection-Specific Effects

As noted earlier, one caveat of assessing source recognition with the contrast between SC and CR items (old/new effects) is that it likely captures both recollection and familiarity-based activity. We were interested in the effects of age and material on recollection-specific processing and to this end, we compared activity between studied items associated with correct SCs and studied items that were associated with incorrect or don't know source judgments (SINCDK). We present a few regions of interest below (for full analysis, see Supplementary Tables).

Material-Independent Effects

Common to groups.

ROI analyses of the MTL (see Materials and Methods) revealed activity common to groups that was greater for SC than SINCDK items in the left posterior hippocampus (−30, −36, 0; T = 3.45; BA 37) (see Fig. 5). The reverse contrast revealed right-lateralized PFC regions where activity was greater for SINCDK than SC items including the inferior frontal operculum (44, 16, 6; T = 4.13; BA 48) (Fig. 5).

Figure 5.

Material-independent source memory effects, specific to recollection (SC vs. SINCDK). Activity common to groups and age-related interactions are shown. Select regions are displayed on the MNI reference brain. Plots show parameter estimates for the event-related response at the peak maxima of the selected regions for SC and SINCDK trial types. Error bars depict standard error of the mean across participants, for each group. (P < 0.0005, uncorrected, with a 5 voxel extent; P < 0.05, FWE-corrected for the MTL regions; exclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Figure 5.

Material-independent source memory effects, specific to recollection (SC vs. SINCDK). Activity common to groups and age-related interactions are shown. Select regions are displayed on the MNI reference brain. Plots show parameter estimates for the event-related response at the peak maxima of the selected regions for SC and SINCDK trial types. Error bars depict standard error of the mean across participants, for each group. (P < 0.0005, uncorrected, with a 5 voxel extent; P < 0.05, FWE-corrected for the MTL regions; exclusive masking conducted as described in Materials and Methods). YA, young; OA, old.

Group differences.

No regions showed greater source recollection (i.e., SC > SINCDK) effects for young adults than older adults. However, older adults showed greater source recollection effects than young adults in the MTL. Specifically, ROI analyses revealed greater activity for SC than SINCDK items in the right PrC (32, −6, −28; T = 3.38; BA 36) and (30, 0, −32; T = 3.35; BA 36) with no reliable differences between conditions in the young, as can be seen in Figure 5. In contrast, young adults showed greater activity for SINCDK than SC items in right-lateralized PFC regions, including the anterior insula/inferior orbitofrontal cortex (30, 30, 0; T = 4.36; BA 47) (shown in Fig. 5) where no effects were observed in the old (Additionally, we performed a subsidiary analysis for subgroups of participants from each age group that were roughly matched for source accuracy, in order to address the possibility that the observed age group differences in memory-related activity are the result of performance rather than age differences, per se [Rugg and Morcom 2005]. A subset of young (n = 7) and of old (n = 7) were used for this analysis, based on their Pr-Source accuracy estimates, such that they would be more closely matched than were the larger age groups. Source accuracy estimates across material type did not reliably differ between these age groups [Young: 68.4%, Old: 60.0%; t12 = 1.82, P = 0.09]. fMRI results showed similar age-related differences as those observed for the larger age groups in material-independent effects, most notably age-related decreases in source monitoring effects [i.e., SINCDK > SC] in the anterior insula/inferior orbitofrontal cortex [30, 30, 0; T = 4.85] and age-related increases in activity related to source recollection [i.e., SC > SINCDK] in the PrC [30, 0, −32; T = 4.75].).

To determine if the overrecruitment of the PrC by older adults might be consistent with theories of functional compensation, as we describe above, activity from the peak voxel coordinate for the PrC region shown in Figure 5 (30, 0, −32; BA 36) was extracted and correlated with our 3 behavioral measures. Activity was positively correlated with the Pr-Source measure (across material type) in older adults (r = 0.76, P = 0.001) but not in the young (r = 0.001, P = 0.996), as shown in Figure 6. The positive correlation was reliable in the old even after excluding the one outlier participant (r = 0.58, P = 0.04).

Figure 6.

Correlations between source memory accuracy and activity in the PrC for both groups. YA, young; OA, old.

Figure 6.

Correlations between source memory accuracy and activity in the PrC for both groups. YA, young; OA, old.

Material-Dependent Effects

No regions exhibited material-dependent recollection-specific activity that overlapped with material processing regions in either group.

Age Group Differences in Lateralization of Source Memory Effects

Additionally, we sought to determine whether there was any evidence of reduced lateralization in PFC memory-related activity in older adults relative to the young, as suggested by previous research (e.g., Cabeza 2002; Grady, McIntosh, Beig, et al. 2003; Morcom et al. 2003; Davis et al. 2008; Duverne et al. 2009b). Specifically, parameter estimates were extracted from voxels for PFC regions showing old/new effects common to groups ([{−50, 32, 20; BA 45}; {44, 10, 40; BA 44}; {42, 34, 22; BA 45}; { 30, 8, 60; BA 8}]) and the one region displaying age-related decreases in CR > SC (−10, 38, −12; BA 11) along with corresponding homologous voxels in the opposite hemisphere. These parameter estimate values were submitted to ANOVAs employing factors of Group, Condition (SC, CR), and Hemisphere. No interactions between these factors were observed for these PFC regions (F1,28's < 0.29, P's > 0.59). Similarly, parameter estimates were extracted from PFC regions showing source recollection effects common to both age groups ([{44, 16, 6; BA 45/48}; {−30, 50, 16; BA 46}]), as well as the right inferior orbitofrontal region that displayed age-related decreases in SINCDK > SC effects (30, 30, 0; BA 47), along with each corresponding homologous voxel and submitted to Group × Condition (SC, SINCDK) × Hemisphere ANOVAs. No interactions between these factors were identified for any region (F1,28's < 1.45, P's > 0.23). Collectively, these results suggest that there were no age-related differences in the lateralization of source memory-related activity in the PFC.

Discussion

The purpose of the present study was to determine the extent to which source memory impairments, commonly observed in older adults for numerous kinds of materials, are related to age-related alterations in material-dependent or material-independent processes during retrieval. To our knowledge, this is the first study to examine age-related changes in source memory retrieval and associated neural activity as a function of material type. As predicted, several regions, including portions of the posterior parietal cortex, lateral PFC made material-independent contributions to source memory effects, for both young and older adults. Additionally, both age groups recruited the hippocampus in a recollection-specific manner, with no group differences in this activity. Older adults also displayed overrecruitment in several regions adjacent to parietal and frontal areas similarly recruited by both groups, as well as the anterior MTL (PrC), which was correlated with better source memory accuracy in the older adults only. Furthermore, older adults demonstrated source memory impairments for both words and objects and underrecruitment of material-independent source monitoring activity in the right lateral PFC, generally consistent with the frontal aging hypothesis. Finally, in addition to age-related activity reductions in left lateral temporal and extrastriate regions, associated with word- and object-specific processing, respectively, age-related interactions, in the form of crossover effects, were observed for material-dependent source memory effects in a subset of these material-sensitive processing regions. We discuss each of these results and their implications below.

Behavioral Performance

Our behavioral findings indicate source memory impairments, for both words and objects, in older adults, consistent with the commonly observed pattern in healthy aging (Spencer and Raz 1995; Yonelinas 2002; Mitchell and Johnson 2009 for reviews). We (Duarte et al. 2006, 2008) and others (Mark and Rugg 1998; Ciaramelli and Ghetti 2007) have suggested that objective measures of recollection, as via source memory judgments, are functionally dissociable from subjective measures, such as calculated according to the “remember-know” procedure (Tulving 1985). Subjective measures are more inclusive, in that any detail about a prior event may support recollective judgments, whereas objective measures are restricted to the experimentally manipulated context, and there is only one right answer. Although we have previously found that high-functioning older adults, matched in global recognition to the level of the young, may exhibit intact subjective measures, both high and low functioning older adults demonstrated equivalent impairments in objective recollection, which we suggested may reflect a more pervasive deficit in aging (Duarte et al. 2006, 2008). The current results, showing equivalent source memory impairments for both words and objects, offer support for this hypothesis.

Although material type did not influence the effects of aging on source and/or item memory impairments, item memory for objects was better than for words in both groups. This finding is most likely due to the “picture superiority effect” in which pictures are better remembered than other stimuli, a phenomenon which research suggests does not decline with age (Winograd et al. 1982). By contrast, source memory estimates were slightly greater for words than for objects. The most likely explanation for this result is that the attention grabbing influence of the object stimuli resulted in a slight item–source memory tradeoff. Although relatively small, this accuracy difference between source estimates must be noted as a potential caveat for the interpretation of material-dependent source memory effects, particularly for regions that did not overlap with regions associated with material-specific processing per se. Thus, only material-dependent memory effects also overlapping with material-specific processing areas are described below.

Material-Independent Source Memory Effects

Both groups showed old/new effects in largely left-lateralized posterior parietal cortex and lateral PFC, independent of material; consistent with many previous studies suggesting that these regions contribute to source memory retrieval (Dobbins et al. 2002; Lundstrom et al. 2005; Swick et al. 2006; see Cabeza 2008; Vilberg and Rugg 2008 for reviews; Duverne et al. 2009a; Donaldson et al. 2010). Additionally, results revealed activity specific to recollection in the posterior hippocampus in both age groups, consistent with numerous previous studies implicating this region in recollection rather than recognition more generally (Eichenbaum et al. 2007; Cabeza 2008; Vilberg and Rugg 2008 for reviews; Duarte et al. 2011). It is noteworthy that neuroanatomical evidence from non-human primates (Clower et al. 2001; Lavenex et al. 2002) and functional connectivity studies in humans (Vincent et al. 2006) suggest that the lateral parietal cortex is highly connected with the MTL, including the hippocampus. The present findings may therefore suggest that in healthy older adults with even moderate source memory impairment, or with no measurable impairment (Morcom et al. 2007; Duverne et al. 2008), a hippocampal–parietal network supporting recollection, for various kinds of materials and associations, may be relatively unaffected by aging. Collectively, these data add support to some recent findings (Morcom et al. 2007; Duverne et al. 2008) suggesting that source retrieval-related activity may be largely age invariant.

These results are inconsistent, however, with one previous study finding impairments in recollection concomitant with reductions in recollection-related activity in the hippocampus and lateral parietal cortex (Daselaar, Fleck, Dobbins, et al. 2006). One potential explanation for these discrepant results may be that older adults' probability of successful recollection was lower in this previous study. Given that this previous study measured recollection subjectively, in contrast to the objective source memory measure used here, it is difficult to compare recollection estimates between these studies. It should be noted, however, that we have also failed to find age-related reductions in recollection-related activity in the hippocampus using a subjective measure (Duarte et al. 2008). Collectively, these results, in addition to the finding in the current study of age-equivalency of new > old effects in the anterior hippocampus, may suggest that underrecruitment of the MTL in older adults is at best, an inconsistent finding.

While the networks supporting source memory retrieval were largely similar between groups, there were also some interesting group differences. As was also observed in a previous episodic retrieval study (Morcom et al. 2007), older adults showed greater source memory effects (old > new) in several, primarily posterior parietal regions adjacent to those similarly recruited by both groups. Morcom and colleagues suggested that this enhanced activation may reflect a decline in neural efficiency in the old, whereby enhanced activity in a common network of retrieval-supportive regions is necessary to support memory performance that is equivalent to that of the young (Morcom et al. 2007). The subsequent findings of Duverne et al. (2008) cast doubt on this conclusion, suggesting instead that the pattern of overrecruitment in the study by Morcom et al. (2007) reflected the reliance on different kinds of information to support recollection by the young and old. Specifically, Duverne et al. (2008) employed a source task that relied primarily on perceptual processing, whereas either perceptual or conceptual information could have supported source recollection in the study by Morcom et al. (2007). In the current study, the source retrieval task emphasized retrieval of conceptual information (i.e., the semantic encoding task). Thus, it is difficult to see how recollection of different types of information by the young and old could explain the current pattern of age-related overrecruitment. Although additional studies are needed to determine the conditions under which this pattern is observed, the current findings are most consistent with the age-related neural inefficiency hypothesis. However, it should be noted that it is unclear whether this overrecruitment is in fact compensatory in nature or reflects general neural inefficiency. Our findings extend these previous results by showing that this pattern of overrecruitment is independent of material type and occurs even when performance is reduced in the old. One consistent finding with both of these previous studies, however, is the lack of evidence to support the HAROLD model of aging, where age-related overrecruitment, particularly in PFC, manifests as a reduction in lateralization of activity (Cabeza 2002; Cabeza et al. 2002; Dolcos et al. 2002 for review; Morcom et al. 2003; Cabeza et al. 2004; Grady et al. 2005; Gutchess et al. 2005). Although there are likely many differences across studies that may explain this discrepancy, including the fact that many of these previous studies did not examine neural correlates of source memory accuracy but rather item recognition or retrieval attempts, the current data suggest that a greater bilateral distribution of frontal activity is not necessarily the rule in aging.

One region that was uniquely recruited by the old to support successful source recollection was the PrC. Moreover, this activity was positively correlated with source accuracy performance exclusively in the older adults, consistent with the compensation hypothesis (Cabeza 2002; Cabeza et al. 2002; Rajah and D'Esposito 2005 for review). This finding is reminiscent of a previous study in which an age-related increase in perirhinal cortical activity was observed during retrieval (Daselaar, Fleck, and Cabeza 2006). However, in the study by Daselaar and colleagues, 2006, this increase in perirhinal activity was observed for a familiarity memory contrast, not recollection and indeed, findings from neuroimaging (Henson et al. 2003; Ranganath et al. 2004; Montaldi et al. 2006) and human lesion studies (Bowles et al. 2007) suggest that the PrC supports familiarity rather than recollection. One speculative interpretation of the perirhinal overrecruitment is that older adults relied upon familiarity to support their source memory judgments to a greater extent than the young, who showed no source memory activity in this region. Without a direct measure of familiarity, as via remember/know (Tulving 1985) or receiver operating characteristics (ROC) (Yonelinas 1994) methods, however, we cannot fully determine the perirhinal over-recruitment reflects a greater contribution of familiarity to source recognition in the old relative to young adults. An alternative explanation is that, as discussed in the introduction, the PrC may support source recollection for objects (Awipi and Davachi 2008), presumably because of the role of the PrC in object perception (Buckley and Gaffan 2006 for review; Lee et al. 2006; Barense et al. 2007, 2009). Although perirhinal activity was greater for objects than words in both age groups (regardless of memory judgment), it is possible that older adults may have relied upon PrC-mediated perceptual processes for both objects and the imagined representations of the concrete nouns when making source accuracy decisions as a compensatory strategy for insufficiently recruited PFC-mediated executive processes (see below). Future aging studies that investigate source accuracy for additional kinds of stimuli, namely nonobject stimuli, are needed to assess the validity of this hypothesis.

Consistent with our prediction that changes in PFC activity may contribute to source memory deficits in older adults, material-independent source memory effects in right-lateralized PFC regions were reduced in older adults. This underrecruitment is similar to some recent evidence showing age-related reductions in lateral PFC activity associated with source memory retrieval together with impaired source memory accuracy (Mitchell et al. 2006; Duarte et al. 2008; Rajah et al. 2009). Specifically, activity was greater for trials for which correct source information was not recollected. This pattern is largely consistent with evidence suggesting the right lateral PFC, including the regions identified here, may be involved in inhibition of irrelevant information (Aron et al. 2004) and postretrieval monitoring and repeated evaluation of information retrieved from MTL (Henson et al. 1999, 2000; Simons and Spiers 2003; Donaldson et al. 2010). The finding of longer RTs for the unsuccessful source trials in both groups is consistent with this hypothesis, given that monitoring may have been engaged to a greater extent when retrieval failed to produce the desired source information. If the right lateral PFC contributes to source retrieval via the inhibition and monitoring mechanisms described above, it may be the case that age-related underrecruitment of these executive processes may lead to impaired source memory accuracy for various kinds of content. The present findings are consistent with previous fMRI evidence from our group suggesting that age-related decreases in PFC activity may be more apparent for objective than subjective measures of recollection (Duarte et al. 2008). We have argued that PFC-mediated executive processes, like monitoring, may be engaged to a greater extent for objective tests of recollection, as subjective recollection measures can be supported by any number of retrieved details, limiting the need for extended postretrieval evaluation. Collectively, these data offer support for the frontal aging hypothesis (West 1996; Raz 2000 for reviews), suggesting that age-related dysfunction in the PFC underlies many of the cognitive deficits associated with aging, including impaired source memory retrieval.

It should be noted that the present imaging results may reflect a combination of age differences as well as performance differences. That is, as performance declines, the proportion of guessing may increase and dilute memory-related neural activity, exaggerating the degree of group differences, particularly age-related underrecruitment (Rugg and Morcom 2005). We attempted to mitigate against this confound by allowing participants to indicate that they did not know the source, so that they were not forced to guess. Nonetheless, source accuracy was still impaired in the old. Despite this performance difference, many of our results, including the largely overlapping patterns of activity between age groups and the overrecruitment of the MTL, are similar to those of previous source retrieval studies in which source accuracy was matched between age groups (Morcom et al. 2007; Duverne et al. 2008). In a further attempt to disentangle age effects from performance effects, we performed a subsidiary analysis for subgroups of participants from each age group that were roughly matched for source accuracy, across material type. Interestingly, even when performance was matched, age-related underrecruitment of the right PFC region implicated in postretrieval monitoring and overrecruitment of the PrC was observed. Thus, these observed age-group differences may in fact be related to the effects of aging on source memory retrieval. However, future research explicitly matching performance between age groups will be necessary to verify these results.

Material-Dependent Source Memory Effects

As predicted based on previous perceptual discrimination (Malach et al. 1995; Grill-Spector et al. 2001; Buckley and Gaffan 2006; Lee et al. 2006, 2008; Barense et al. 2009) and memory studies (Malach et al. 1995; Grill-Spector et al. 2001; Pihlajamaki et al. 2005; Litman et al. 2009; O'Neil et al. 2009), anterior MTL regions, including the hippocampus and PrC, exhibited greater activity for objects regardless of memory judgment for both young and older adults, in addition to portions of the LOC, fusiform gyrus, and right-lateralized frontal regions. Similarly, left-lateralized frontotemporal regions associated with visual word perception and memory (Fiez and Peterson 1998; Price 2000 for review; Roskies et al. 2001) exhibited activity specifically for words for both age groups. Consistent with our predictions, a subset of these regions demonstrated material-dependent source memory effects. Most notably, activity in the LOC was associated with source memory retrieval (old/new effect) specifically for objects in the young. However, there was a crossover effect in this region, with older adults showing old > new activity for words but not objects. It is not immediately clear what might explain this pattern in the older adults. One possibility is that, given that older adults demonstrated underrecruitment of object-specific activity in posterior cortex (described below), this crossover pattern of activity may underlie their memory impairments and more liberal response bias for these stimuli. Given that we were unable to detect a direct relationship between this activity and source accuracy for objects in the older adults, further work is necessary to assess the validity of this hypothesis.

Overall, there was little material-dependent source memory activity and none related to recollection specifically that also overlapped with material-specific processing regions. This finding appears contradictory to previous fMRI and ERP evidence revealing content-dependent neural correlates of recollection (Wheeler and Buckner 2004; Khader, Burke, et al. 2005; Khader, Heil, and Rosler 2005; Woodruff et al. 2005; Johnson and Rugg 2007; Johnson et al. 2008; for review Rugg et al. 2008). One important difference between these previous studies and the current design is that recollection was measured subjectively, via “remember” judgments, in which any recollected detail, including material-specific perceptual features, could have supported remember judgments. In the current study, the source retrieval task may have biased attention toward conceptual, rather than perceptual, representations of the stimuli, therefore increasing the likelihood that the sought after conceptual encoding context would be recollected, consistent with the source monitoring framework (Johnson et al. 1993). We suggest that material-dependent recollection effects would be more evident if the demands of the source retrieval task explicitly required recovery of information about each study item's perceptual features or alternatively, if recollection were measured subjectively. Indeed, our finding of material-dependent source memory effects exclusively for the old > new contrast, where noncriterial recollection of perceptual details may have been more likely to contribute, is consistent with this hypothesis.

Finally, in addition to the material-sensitive processing effects that were similar for both age groups, older adults showed weaker word- and object-sensitive activity in the left lateral temporal cortex and bilateral fusiform gyri, respectively. These results are similar to previous episodic memory studies showing age-related underrecruitment of material-sensitive posterior cortical areas (Grady, McIntosh, and Craik 2003; Cabeza et al. 2004; Gutchess et al. 2005; Dennis et al. 2008). Although, as discussed above, most of these word and object-sensitive regions were not directly modulated by source memory accuracy, it remains possible that underrecruitment of posterior cortical regions results in impoverished item-specific representations, contributing to older adults' memory impairments for these stimuli and associated source details. These impoverished representations, particularly for objects, may have further contributed to older adults' tendency to endorse novel objects as having been studied, as indicated by their more liberal response bias than that of the young specifically for objects.

Conclusions

In conclusion, results from the present study suggest that age-related impairments in source memory accuracy for a variety of materials and source associations may be related to disruptions in regions that contribute to episodic retrieval in a material-independent manner, namely the lateral PFC. However, we argue that material-specific memory deficits may be further impacted by altered patterns of source memory activity in material-dependent processing regions, particularly for objects, as well as impoverished material-specific representations for words and objects and perhaps other kinds of stimuli in posterior cortex. Our results are generally consistent with the idea that source memory impairments may be pervasive in aging due to altered executive control processes subserved by the PFC, consistent with the frontal aging hypothesis (see West 1996; Raz 2000 for reviews) but offer no support for the proposal that disruption in the hippocampus contributes to these memory deficits. Given the present evidence of MTL-mediated functional compensation and lack of frontal overrecruitment, an important issue for future research is to determine the conditions under which age-related increases in MTL or PFC activity are observed and either contribute to or mitigate age-related memory decline.

Funding

This study was supported, in part, by the National Institute on Aging training (grant 5T32AG000175-23).

Supplementary Material

Supplementary material can be found at: http://www.cercor.oxfordjournals.org/

Conflict of Interest : None declared.

References

Aron
AR
Robbins
TW
Poldrack
RA
Inhibition and the right inferior frontal cortex
Trends Cogn Sci
 , 
2004
, vol. 
8
 (pg. 
170
-
177
)
Ashburner
J
A fast diffeomorphic image registration algorithm
Neuroimage
 , 
2007
, vol. 
38
 (pg. 
95
-
113
)
Awipi
T
Davachi
L
Content-specific source encoding in the human medial temporal lobe
J Exp Psychol Learn Mem Cogn
 , 
2008
, vol. 
34
 (pg. 
769
-
779
)
Badre
D
Wagner
AD
Left ventrolateral prefrontal cortex and the cognitive control of memory
Neuropsychologia
 , 
2007
, vol. 
45
 (pg. 
2883
-
2901
)
Barense
MD
Gaffan
D
Graham
KS
The human medial temporal lobe processes online representations of complex objects
Neuropsychologia
 , 
2007
, vol. 
45
 (pg. 
2963
-
2974
)
Barense
MD
Henson
RN
Lee
AC
Graham
KS
Medial temporal lobe activity during complex discrimination of faces, objects, and scenes: effects of viewpoint
Hippocampus
 , 
2009
, vol. 
20
 (pg. 
389
-
401
)
Benton
AL
Hamsher
SKd
Sivan
AB
Multilingual aplasia examination
 , 
1983
Iowa City (IA)
AJA Associates
Bowles
B
Crupi
C
Mirsattari
SM
Pigott
SE
Parrent
AG
Pruessner
JC
Yonelinas
AP
Kohler
S
Impaired familiarity with preserved recollection after anterior temporal-lobe resection that spares the hippocampus
Proc Natl Acad Sci U S A
 , 
2007
, vol. 
104
 (pg. 
16382
-
16387
)
Buckley
MJ
Gaffan
D
Perirhinal cortical contributions to object perception
Trends Cogn Sci
 , 
2006
, vol. 
10
 (pg. 
100
-
107
)
Cabeza
R
Hemispheric asymmetry reduction in older adults: the HAROLD model
Psychol Aging
 , 
2002
, vol. 
17
 (pg. 
85
-
100
)
Cabeza
R
Role of parietal regions in episodic memory retrieval: the dual attentional processes hypothesis
Neuropsychologia
 , 
2008
, vol. 
46
 (pg. 
1813
-
1827
)
Cabeza
R
Anderson
ND
Locantore
JK
McIntosh
AR
Aging gracefully: compensatory brain activity in high-performing older adults
Neuroimage
 , 
2002
, vol. 
17
 (pg. 
1394
-
1402
)
Cabeza
R
Daselaar
SM
Dolcos
F
Prince
SE
Budde
M
Nyberg
L
Task-independent and task-specific age effects on brain activity during working memory, visual attention and episodic retrieval
Cereb Cortex
 , 
2004
, vol. 
14
 (pg. 
364
-
375
)
Cansino
S
Maquet
P
Dolan
RJ
Rugg
MD
Brain activity underlying encoding and retrieval of source memory
Cereb Cortex
 , 
2002
, vol. 
12
 (pg. 
1048
-
1056
)
Chee
MW
Goh
JO
Venkatraman
V
Tan
JC
Gutchess
A
Sutton
B
Hebrank
A
Leshikar
E
Park
D
Age-related changes in object processing and contextual binding revealed using fMR adaptation
J Cogn Neurosci
 , 
2006
, vol. 
18
 (pg. 
495
-
507
)
Ciaramelli
E
Ghetti
S
What are confabulators' memories made of? A study of subjective and objective measures of recollection in confabulation
Neuropsychologia
 , 
2007
, vol. 
45
 (pg. 
1489
-
1500
)
Clower
DM
West
RA
Lynch
JC
Strick
PL
The inferior parietal lobule is the target of output from the superior colliculus, hippocampus, and cerebellum
J Neurosci
 , 
2001
, vol. 
21
 (pg. 
6283
-
6291
)
Daselaar
SM
Fleck
MS
Cabeza
R
Triple dissociation in the medial temporal lobes: recollection, familiarity, and novelty
J Neurophysiol
 , 
2006
, vol. 
96
 (pg. 
1902
-
1911
)
Daselaar
SM
Fleck
MS
Dobbins
IG
Madden
DJ
Cabeza
R
Effects of healthy aging on hippocampal and rhinal memory functions: an event-related fMRI study
Cereb Cortex
 , 
2006
, vol. 
16
 (pg. 
1771
-
1782
)
Davachi
L
Mitchell
JP
Wagner
AD
Multiple routes to memory: distinct medial temporal lobe processes build item and source memories
Proc Natl Acad Sci U S A
 , 
2003
, vol. 
100
 (pg. 
2157
-
2162
)
Davis
SW
Dennis
NA
Daselaar
SM
Fleck
MS
Cabeza
R
Que PASA? The posterior-anterior shift in aging
Cereb Cortex
 , 
2008
, vol. 
18
 (pg. 
1201
-
1209
)
Dennis
NA
Hayes
SM
Prince
SE
Madden
DJ
Huettel
SA
Cabeza
R
Effects of aging on the neural correlates of successful item and source memory encoding
J Exp Psychol Learn Mem Cogn
 , 
2008
, vol. 
34
 (pg. 
791
-
808
)
Diana
RA
Yonelinas
AP
Ranganath
C
Imaging recollection and familiarity in the medial temporal lobe: a three-component model
Trends Cogn Sci
 , 
2007
, vol. 
11
 (pg. 
379
-
386
)
Dobbins
IG
Foley
H
Schacter
DL
Wagner
AD
Executive control during episodic retrieval: multiple prefrontal processes subserve source memory
Neuron
 , 
2002
, vol. 
35
 (pg. 
989
-
996
)
Dobbins
IG
Han
S
Cue- versus probe-dependent prefrontal cortex activity during contextual remembering
J Cogn Neurosci
 , 
2006
, vol. 
18
 (pg. 
1439
-
1452
)
Dobbins
IG
Wagner
AD
Domain-general and domain-sensitive prefrontal mechanisms for recollecting events and detecting novelty
Cereb Cortex
 , 
2005
, vol. 
15
 (pg. 
1768
-
1778
)
Dolcos
F
Rice
HJ
Cabeza
R
Hemispheric asymmetry and aging: right hemisphere decline or asymmetry reduction
Neurosci Biobehav Rev
 , 
2002
, vol. 
26
 (pg. 
819
-
825
)
Donaldson
DI
Wheeler
ME
Petersen
SE
Remember the source: dissociating frontal and parietal contributions to episodic memory
J Cogn Neurosci
 , 
2010
, vol. 
22
 (pg. 
377
-
391
)
Duarte
A
Henson
R
Graham
KS
Stimulus content and the neural correlates of item recognition and source memory
Brain Res
 , 
2011
, vol. 
1373
 (pg. 
110
-
123
)
Duarte
A
Henson
RN
Graham
KS
The effects of aging on the neural correlates of subjective and objective recollection
Cereb Cortex
 , 
2008
, vol. 
18
 (pg. 
2169
-
2180
)
Duarte
A
Henson
RN
Knight
RT
Emery
T
Graham
KS
Orbito-frontal cortex is necessary for temporal context memory
J Cogn Neurosci
 , 
2009
, vol. 
22
 (pg. 
1819
-
1831
)
Duarte
A
Ranganath
C
Trujillo
C
Knight
RT
Intact recollection memory in high-performing older adults: ERP and behavioral evidence
J Cogn Neurosci
 , 
2006
, vol. 
18
 (pg. 
33
-
47
)
Duarte
A
Ranganath
C
Winward
L
Hayward
D
Knight
RT
Dissociable neural correlates for familiarity and recollection during the encoding and retrieval of pictures
Brain Res Cogn Brain Res
 , 
2004
, vol. 
18
 (pg. 
255
-
272
)
Duverne
S
Habibi
A
Rugg
MD
Regional specificity of age effects on the neural correlates of episodic retrieval
Neurobiol Aging
 , 
2008
, vol. 
29
 (pg. 
1902
-
1916
)
Duverne
S
Motamedinia
S
Rugg
MD
Effects of age on the neural correlates of retrieval cue processing are modulated by task demands
J Cogn Neurosci
 , 
2009
, vol. 
21
 (pg. 
1
-
17
)
Duverne
S
Motamedinia
S
Rugg
MD
The relationship between aging, performance, and the neural correlates of successful memory encoding
Cereb Cortex
 , 
2009
, vol. 
19
 (pg. 
733
-
744
)
Eichenbaum
H
Yonelinas
AR
Ranganath
C
The medial temporal lobe and recognition memory
Annu Rev Neurosci
 , 
2007
, vol. 
30
 (pg. 
123
-
152
)
Fiez
JA
Peterson
SE
Neuroimaging studies of word reading
Proc Natl Acad Sci U S A
 , 
1998
, vol. 
95
 (pg. 
914
-
921
)
Fletcher
PC
Henson
RN
Frontal lobes and human memory: insights from functional neuroimaging
Brain
 , 
2001
, vol. 
124
 (pg. 
849
-
881
)
Gottlieb
LJ
Uncapher
MR
Rugg
MD
Dissociation of the neural correlates of visual and auditory contextual encoding
Neuropsychologia
 , 
2010
, vol. 
48
 (pg. 
137
-
144
)
Grady
CL
McIntosh
AR
Beig
S
Keightley
ML
Burian
H
Black
SE
Evidence from functional neuroimaging of a compensatory prefrontal network in Alzheimer's disease
J Neurosci
 , 
2003
, vol. 
23
 (pg. 
986
-
993
)
Grady
CL
McIntosh
AR
Craik
FI
Age-related differences in the functional connectivity of the hippocampus during memory encoding
Hippocampus
 , 
2003
, vol. 
13
 (pg. 
572
-
586
)
Grady
CL
McIntosh
AR
Craik
FI
Task-related activity in prefrontal cortex and its relation to recognition memory performance in young and old adults
Neuropsychologia
 , 
2005
, vol. 
43
 (pg. 
1466
-
1481
)
Grill-Spector
K
Kourtzi
Z
Kanwisher
N
The lateral occipital complex and its role in object recognition
Vision Res
 , 
2001
, vol. 
41
 (pg. 
1409
-
1422
)
Gutchess
AH
Welsh
RC
Hedden
T
Bangert
A
Minear
M
Liu
LL
Park
DC
Aging and the neural correlates of successful picture encoding: frontal activations compensate for decreased medial-temporal activity
J Cogn Neurosci
 , 
2005
, vol. 
17
 (pg. 
84
-
96
)
Henson
R
Shallice
T
Dolan
RJ
Right prefrontal cortex and episodic memory retrieval: a functional MRI test of the monitoring hypothesis
Brain
 , 
1999
, vol. 
122
 
Pt 7
(pg. 
1367
-
1381
)
Henson
RN
Cansino
S
Herron
JE
Robb
WG
Rugg
MD
A familiarity signal in human anterior medial temporal cortex?
Hippocampus
 , 
2003
, vol. 
13
 (pg. 
301
-
304
)
Henson
RN
Rugg
MD
Shallice
T
Dolan
RJ
Confidence in recognition memory for words: dissociating right prefrontal roles in episodic retrieval
J Cogn Neurosci
 , 
2000
, vol. 
12
 (pg. 
913
-
923
)
Iidaka
T
Okada
T
Murata
T
Omori
M
Kosaka
H
Sadato
N
Yonekura
Y
Age-related differences in the medial temporal lobe responses to emotional faces as revealed by fMRI
Hippocampus
 , 
2002
, vol. 
12
 (pg. 
352
-
362
)
Johnson
JD
Minton
BR
Rugg
MD
Content dependence of the electrophysiological correlates of recollection
Neuroimage
 , 
2008
, vol. 
39
 (pg. 
406
-
416
)
Johnson
JD
Rugg
MD
Recollection and the reinstatement of encoding-related cortical activity
Cereb Cortex
 , 
2007
, vol. 
17
 (pg. 
2507
-
2515
)
Johnson
MK
Hashtroudi
S
Lindsay
DS
Source monitoring
Pschol Rev
 , 
1993
, vol. 
114
 (pg. 
3
-
28
)
Kensinger
EA
How emotion affects older adults' memories for event details
Memory
 , 
2009
, vol. 
17
 (pg. 
208
-
219
)
Kensinger
EA
Schacter
DL
Amygdala activity is associated with the successful encoding of item, but not source, information for positive and negative stimuli
J Neurosci
 , 
2006
, vol. 
26
 (pg. 
2564
-
2570
)
Khader
P
Burke
M
Bien
S
Ranganath
C
Rosler
F
Content-specific activation during associative long-term memory retrieval
Neuroimage
 , 
2005
, vol. 
27
 (pg. 
805
-
816
)
Khader
P
Heil
M
Rosler
F
Material-specific long-term memory representations of faces and spatial positions: evidence from slow event-related brain potentials
Neuropsychologia
 , 
2005
, vol. 
43
 (pg. 
2109
-
2124
)
Kucera
H
Francis
W
Computational analysis of present-day American English
 , 
1967
Providence (RI)
Brown University Press
Lavenex
P
Suzuki
WA
Amaral
DG
Perirhinal and parahippocampal cortices of the macaque monkey: projections to the neocortex
J Comp Neurol
 , 
2002
, vol. 
447
 (pg. 
394
-
420
)
Lee
AC
Bandelow
S
Schwarzbauer
C
Henson
RN
Graham
KS
Perirhinal cortex activity during visual object discrimination: an event-related fMRI study
Neuroimage
 , 
2006
, vol. 
33
 (pg. 
362
-
373
)
Lee
AC
Scahill
VL
Graham
KS
Activating the medial temporal lobe during oddity judgment for faces and scenes
Cereb Cortex
 , 
2008
, vol. 
18
 (pg. 
683
-
696
)
Li
J
Morcom
AM
Rugg
MD
The effects of age on the neural correlates of successful episodic retrieval: an ERP study
Cogn Affect Behav Neurosci
 , 
2004
, vol. 
4
 (pg. 
279
-
293
)
Litman
L
Awipi
T
Davachi
L
Category-specificity in the human medial temporal lobe cortex
Hippocampus
 , 
2009
, vol. 
19
 (pg. 
308
-
319
)
Lundstrom
BN
Ingvar
M
Petersson
KM
The role of precuneus and left inferior frontal cortex during source memory episodic retrieval
Neuroimage
 , 
2005
, vol. 
27
 (pg. 
824
-
834
)
Malach
R
Reppas
JB
Benson
RR
Kwong
KK
Jiang
H
Kennedy
WA
Ledden
PJ
Brady
TJ
Rosen
BR
Tootell
RB
Object-related activity revealed by functional magnetic resonance imaging in human occipital cortex
Proc Natl Acad Sci U S A
 , 
1995
, vol. 
92
 (pg. 
8135
-
8139
)
Mandler
G
Recognising: the judgment of previous occurrence
Pschol Rev
 , 
1980
, vol. 
87
 (pg. 
252
-
271
)
Mark
RE
Rugg
MD
Age effects on brain activity associated with episodic memory retrieval. An electrophysiological study
Brain
 , 
1998
, vol. 
121
 
Pt 5
(pg. 
861
-
873
)
McCandliss
BD
Cohen
L
Dehaene
S
The visual word form area: expertise for reading in the fusiform gyrus
Trends Cogn Sci
 , 
2003
, vol. 
7
 (pg. 
293
-
299
)
Mitchell
KJ
Johnson
MK
Source monitoring 15 years later: what have we learned from fMRI about the neural mechanisms of source memory?
Psychol Bull
 , 
2009
, vol. 
135
 (pg. 
638
-
677
)
Mitchell
KJ
Raye
CL
Johnson
MK
Greene
EJ
An fMRI investigation of short-term source memory in young and older adults
Neuroimage
 , 
2006
, vol. 
30
 (pg. 
627
-
633
)
Montaldi
D
Spencer
TJ
Roberts
N
Mayes
AR
The neural system that mediates familiarity memory
Hippocampus
 , 
2006
, vol. 
16
 (pg. 
504
-
520
)
Morcom
AM
Good
CD
Frackowiak
RS
Rugg
MD
Age effects on the neural correlates of successful memory encoding
Brain
 , 
2003
, vol. 
126
 (pg. 
213
-
229
)
Morcom
AM
Li
J
Rugg
MD
Age effects on the neural correlates of episodic retrieval: increased cortical recruitment with matched performance
Cereb Cortex
 , 
2007
, vol. 
17
 (pg. 
2491
-
2506
)
O'Neil
EB
Cate
AD
Kohler
S
Perirhinal cortex contributes to accuracy in recognition memory and perceptual discriminations
J Neurosci
 , 
2009
, vol. 
29
 (pg. 
8329
-
8334
)
Park
DC
Polk
TA
Park
R
Minear
M
Savage
A
Smith
MR
Aging reduces neural specialization in ventral visual cortex
Proc Natl Acad Sci U S A
 , 
2004
, vol. 
101
 (pg. 
13091
-
13095
)
Pereira
JM
Xiong
L
Acosta-Cabronero
J
Pengas
G
Williams
GB
Nestor
PJ
Registration accuracy for VBM studies varies according to region and degenerative disease grouping
Neuroimage
 , 
2010
, vol. 
49
 (pg. 
2205
-
2215
)
Pihlajamaki
M
Tanila
H
Kononen
M
Hanninen
T
Aronen
HJ
Soininen
H
Distinct and overlapping fMRI activation networks for processing of novel identities and locations of objects
Eur J Neurosci
 , 
2005
, vol. 
22
 (pg. 
2095
-
2105
)
Price
CJ
The anatomy of language: contributions from functional neuroimaging
J Anat
 , 
2000
, vol. 
197
 
Pt 3
(pg. 
335
-
359
)
Rajah
MN
D'Esposito
M
Region-specific changes in prefrontal function with age: a review of PET and fMRI studies on working and episodic memory
Brain
 , 
2005
, vol. 
128
 (pg. 
1964
-
1983
)
Rajah
MN
Languay
R
Valiquette
L
Age-related changes in prefrontal cortex activity are associated with behavioural deficits in both temporal and spatial context memory retrieval in older adults
Cortex
 , 
2009
, vol. 
46
 (pg. 
535
-
549
)
Ranganath
C
Yonelinas
AP
Cohen
MX
Dy
CJ
Tom
SM
D'Esposito
M
Dissociable correlates of recollection and familiarity within the medial temporal lobes
Neuropsychologia
 , 
2004
, vol. 
42
 (pg. 
2
-
13
)
Raz
N
Craik
FIM
Salthouse
TA
Aging of the brain and its impact on cognitive performance: integration of structural and functional findings
Handbook of aging and cognition
 , 
2000
2nd ed. Mahwah, NJ: Erlbaum. p. 1-90
Reitan
R
Wolfson
D
The Halstead-Reitan neuropsychological test battery: therapy and clinical assessment. Tucson (AZ): Neuropsychological Press
1985
Roskies
AL
Fiez
JA
Balota
DA
Raichle
ME
Petersen
SE
Task-dependent modulation of regions in the left inferior frontal cortex during semantic processing
J Cogn Neurosci
 , 
2001
, vol. 
13
 (pg. 
829
-
843
)
Rugg
MD
Johnson
JD
Park
H
Uncapher
MR
Encoding-retrieval overlap in human episodic memory: a functional neuroimaging perspective
Prog Brain Res
 , 
2008
, vol. 
169
 (pg. 
339
-
352
)
Rugg
MD
Morcom
AM
Cabeza
R
Nyberg
L
Park
DC
The relationship between brain activity, cognitive performance and aging: the case of memory
Cognitive neuroscience of aging
 , 
2005
Oxford
Oxford University Press
(pg. 
132
-
154
)
Simons
JS
Spiers
HJ
Prefrontal and medial temporal lobe interactions in long-term memory
Nat Rev Neurosci
 , 
2003
, vol. 
4
 (pg. 
637
-
648
)
Smith
AP
Dolan
RJ
Rugg
MD
Event-related potential correlates of the retrieval of emotional and nonemotional context
J Cogn Neurosci
 , 
2004
, vol. 
16
 (pg. 
760
-
775
)
Snodgrass
J
Corwin
J
Pragmatics of measuring recognition memory: applications to dementia and amnesia
J Exp Psychol
 , 
1988
, vol. 
116
 (pg. 
34
-
50
)
Sommer
T
Rose
M
Weiller
C
Buchel
C
Contributions of occipital, parietal and parahippocampal cortex to encoding of object-location associations
Neuropsychologia
 , 
2005
, vol. 
43
 (pg. 
732
-
743
)
Spencer
WD
Raz
N
Differential effects of aging on memory for content and context: a meta-analysis
Psychol Aging
 , 
1995
, vol. 
10
 (pg. 
527
-
539
)
Swick
D
Senkfor
AJ
Van Petten
C
Source memory retrieval is affected by aging and prefrontal lesions: behavioral and ERP evidence
Brain Res
 , 
2006
, vol. 
1107
 (pg. 
161
-
176
)
Tulving
E
Memory and consciousness
Canadian psychol
 , 
1985
, vol. 
26
 (pg. 
1
-
12
)
Uncapher
MR
Wagner
AD
Posterior parietal cortex and episodic encoding: insights from fMRI subsequent memory effects and dual-attention theory
Neurobiol Learn Mem
 , 
2009
, vol. 
91
 (pg. 
139
-
154
)
Vilberg
KL
Rugg
MD
Memory retrieval and the parietal cortex: a review of evidence from a dual-process perspective
Neuropsychologia
 , 
2008
, vol. 
46
 (pg. 
1787
-
1799
)
Vincent
JL
Snyder
AZ
Fox
MD
Shannon
BJ
Andrews
JR
Raichle
ME
Buckner
RL
Coherent spontaneous activity identifies a hippocampal-parietal memory network
J Neurophysiol
 , 
2006
, vol. 
96
 (pg. 
3517
-
3531
)
Wagner
AD
Shannon
BJ
Kahn
I
Buckner
RL
Parietal lobe contributions to episodic memory retrieval
Trends Cogn Sci
 , 
2005
, vol. 
9
 (pg. 
445
-
453
)
West
RL
An application of prefrontal cortex function theory to cognitive aging
Psychol Bull
 , 
1996
, vol. 
120
 (pg. 
272
-
292
)
Wheeler
ME
Buckner
RL
Functional-anatomic correlates of remembering and knowing
Neuroimage
 , 
2004
, vol. 
21
 (pg. 
1337
-
1349
)
Williams
J
Memory assessment scales professional manual
 , 
1991
Odessa (Ukraine)
Psychological Assessment Resources
Winograd
E
Smith
AD
Simon
EW
Aging and the picture superiority effect in recall
J Gerontol
 , 
1982
, vol. 
37
 (pg. 
70
-
75
)
Woodruff
CC
Johnson
JD
Uncapher
MR
Rugg
MD
Content-specificity of the neural correlates of recollection
Neuropsychologia
 , 
2005
, vol. 
43
 (pg. 
1022
-
1032
)
Yassa
MA
Stark
CE
A quantitative evaluation of cross-participant registration techniques for MRI studies of the medial temporal lobe
Neuroimage
 , 
2009
, vol. 
44
 (pg. 
319
-
327
)
Yick
YY
Wilding
EL
Material-specific neural correlates of memory retrieval
Neuroreport
 , 
2008
, vol. 
19
 (pg. 
1463
-
1467
)
Yonelinas
AP
Receiver-operating characteristics in recognition memory: evidence for a dual-process model
J Exp Psychol Learn Mem Cogn
 , 
1994
, vol. 
20
 (pg. 
1341
-
1354
)
Yonelinas
AP
The nature of recollection and familiarity: a review of 30 years of research
J Mem Lang
 , 
2002
, vol. 
46
 (pg. 
441
-
517
)