Skip to main content
NCBI home page
Human Brain Mapping logoLink to Human Brain Mapping
. 2016 Mar 16;37(6):2348–2361. doi: 10.1002/hbm.23178

Aging modulates the oscillatory dynamics underlying successful working memory encoding and maintenance

Amy L Proskovec 1,2, Elizabeth Heinrichs‐Graham 2,3, Tony W Wilson 2,3,4,✉
PMCID: PMC4867257  NIHMSID: NIHMS764788  PMID: 26991358

Abstract

Working memory is central to the execution of many daily functions and is typically divided into three phases: encoding, maintenance, and retrieval. While working memory performance has been repeatedly shown to decline with age, less is known regarding the underlying neural processes. We examined age‐related differences in the neural dynamics that serve working memory by recording high‐density magnetoencephalography (MEG) in younger and older adults while they performed a modified, high‐load Sternberg working memory task with letters as stimuli. MEG data were evaluated in the time‐frequency domain and significant oscillatory responses were imaged using a beamformer. A hierarchical regression was performed to investigate whether age moderated the relationship between oscillatory activity and accuracy on the working memory task. Our results indicated that the spatiotemporal dynamics of oscillatory activity in language‐related areas of the left fronto‐temporal cortices were similar across groups. Age‐related differences emerged during early encoding in the right‐hemispheric homologue of Wernicke's area. Slightly later, group differences emerged in the homologue of Broca's area and these persisted throughout memory maintenance. Additionally, occipital alpha activity during maintenance was stronger, occurred earlier, and involved more cortical tissue in older adults. Finally, age significantly moderated the relationship between accuracy and neural activity in the prefrontal cortices. In younger adults, as prefrontal activity decreased, accuracy tended to increase. Our results are consistent with predictions of the compensation‐related utilization of neural circuits hypothesis (CRUNCH). Such differences in the oscillatory dynamics could reflect compensatory mechanisms, which would aid working memory performance in older age. Hum Brain Mapp 37:2348–2361, 2016. © 2016 Wiley Periodicals, Inc.

Keywords: magnetoencephalography, compensation, oscillation, Broca's, Wernicke's, prefrontal cortex, alpha, CRUNCH

INTRODUCTION

Working memory refers to the on‐line temporary storage and manipulation of information to be employed in ongoing processing, and is central to the execution of a variety of daily functions. It is typically broken down into three phases: encoding, maintenance, and retrieval. Encoding refers to the loading of internal and external information into memory, while maintenance is the process whereby this information is actively retained and rehearsed. Retrieval reflects the active recall and use of this information towards the performance of a specific cognitive activity. Working memory processing has been linked to activations in many neuroanatomical locations, including fronto‐parietal networks, occipital cortices, and the cerebellum [Cabeza and Nyberg, 2000; Rottschy et al., 2012]. These neuroanatomical regions can be segregated into those that are associated with specific working memory paradigms (e.g., n‐back, Sternberg), specific types of stimuli (e.g., visual, acoustic), and/or specific phases of working memory (e.g., encoding). There are also a number of studies that evaluate activation during overall performance, without distinguishing phases such as encoding and maintenance, and other studies that have utilized oscillatory analyses to examine activity in different frequency bands during working memory processing. Given this, there is some variability in the literature regarding the brain regions that are critical for working memory [Rottschy et al., 2012].

Recent magnetoencephalography (MEG) work has shown increased theta activity in midline frontal regions, with concomitant decreases in high‐beta/low‐gamma activity during performance of both n‐back and Sternberg tasks [Brookes et al., 2011]. This beta/gamma decrease was additionally observed in left fronto‐temporal cortices, but only during the Sternberg task. Another MEG study also reported increased frontal theta activity at the sensor level during a Sternberg task [Jensen and Tesche, 2002]. Unfortunately, these studies were unable to clearly distinguish whether the oscillatory responses reflected encoding or maintenance operations, as both sequentially presented stimuli, which causes encoding and maintenance operations to occur in parallel. However, a recent MEG investigation used a modified Sternberg task that involved simultaneously presenting the encoding set within a visual array [Heinrichs‐Graham and Wilson, 2015], thereby permitting analytical distinction between the three phases of working memory. They found strong decreases in alpha/beta activity in left fronto‐temporal cortices during encoding and maintenance operations. This study also reported strong alpha/beta decreases in bilateral occipital cortices beginning immediately after onset of the encoding array, which rapidly dispelled during early maintenance and transformed into a strong increase in a narrower alpha band [Heinrichs‐Graham and Wilson, 2015]. These results corroborate other studies that have reported increases in parieto‐occipital alpha activity during working memory maintenance [Bonnefond and Jensen, 2012, 2013; Jensen et al., 2002; Jiang et al., 2015; Tuladhar et al., 2007].

The aforementioned studies all investigated healthy young adults in their analyses and far less is known about working memory networks in older participants. Such studies are of major interest, as working memory performance has been shown to decline with age [Park et al., 2002; Salthouse et al., 1991]. Several neuroimaging studies, typically utilizing functional magnetic resonance imaging (fMRI) or positron emission tomography (PET), have addressed this topic and generally reported age‐related differences in prefrontal cortex (PFC) activation [Grady, 2012; Spreng et al., 2010]. However, some studies reported stronger and/or greater volumes of activation in older relative to younger adults [Emery et al., 2008; Fakhri et al., 2012; Reuter‐Lorenz et al., 2000], whereas others reported the opposite pattern [Cabeza et al., 1997; Rypma and D'Esposito, 2000]. Several studies have also found a combination of hyper‐activation and hypo‐activation, depending on the brain region, in older relative to younger adults, that varied with memory load [Cappell et al., 2010; Mattay, et al., 2006]. There are also studies that have connected behavioral parameters to age‐specific response patterns [Rypma et al., 2007], and at least one study found increased activation in older relative to younger adults in some areas of the PFC, but a reverse pattern in other areas of the PFC [Spreng et al., 2010]. Beyond the PFC, increased left cerebellar and bilateral occipital activations have been reported in older relative to younger adults during working memory maintenance [Bennett et al., 2013]. Finally, several electrophysiological studies of aging have been conducted. One study demonstrated increased activity in younger relative to older adults in posterior frontal regions (Solesio‐Jofre et al., 2011), whereas another reported greater alpha activity in right temporal electrodes, decreased theta activity in posterior electrodes, and increased low‐beta activity in central electrodes of older relative to younger participants [Karrasch et al., 2004]. Of note, one study used MEG and time‐domain analyses [Solesio‐Jofre et al., 2011], whereas the other used EEG and an oscillatory approach [Karrasch et al., 2004]; thus, direct comparison of these studies is difficult.

In the current study, we examined age‐related differences in the neural dynamics of working memory processing using spatially‐resolved MEG. As stated above, it is widely known that working memory performance declines with age [Park et al., 2002; Salthouse et al., 1991], but the mechanisms involved remain less understood. Furthermore, of the handful of neuroimaging studies available, none have evaluated the temporal dynamics of activity within the encoding and maintenance phases of working memory. Thus, in this study, we used the temporal precision of MEG to identify the neural oscillatory dynamics of working memory performance in younger and older adults by imaging a sequential series of nonoverlapping time bins that spanned the encoding and maintenance phases. Our primary hypotheses were that older participants would exhibit greater oscillatory activity relative to younger participants during encoding and throughout maintenance. Specifically, we expected older adults to engage bilateral PFC and parieto‐occipital cortices more strongly during these time windows compared to younger adults. Additionally, we hypothesized that age would moderate the relationship between oscillatory activity and behavioral working memory performance.

METHODS AND MATERIALS

Subject Selection

We studied 15 older (mean age: 60.2‐years‐old, SD: 8.2, range: 50 − 75) and 16 younger (mean age: 26.0‐years‐old, SD: 2.5, range: 19 − 30) males. All participants were healthy, Caucasian, and recruited from the local community. Exclusionary criteria included any medical illness affecting CNS function, neurological or psychiatric disorder, history of head trauma, current substance abuse, and the MEG Laboratory's standard exclusion criteria (e.g., dental braces, metal implants, battery operated implants, and/or any type of ferromagnetic implanted material). After a complete description of the study was given to participants, written informed consent was obtained following the guidelines of the University of Nebraska Medical Center's Institutional Review Board, which approved the study protocol.

Experimental Paradigm

During MEG recording, participants were seated in a nonmagnetic chair within a magnetically‐shielded room and instructed to fixate on a crosshair presented centrally for 1.0 s. A grid containing six letters was then presented for 2.0 s (encoding). The letters then disappeared from the grid and 3.0 s later (maintenance phase) a single “probe” letter appeared for 0.9 s (retrieval phase). Participants were instructed to respond with a button press as to whether the probe letter was one of the six letters previously presented. The crosshair was present throughout the entire task to reduce the likelihood of eye saccades. Each trial lasted 6.9 s, including a 1.0 s pre‐stimulus fixation; see Figure 1 for an illustration of the overall task design. Each participant completed 128 trials and the task lasted ∼14 min.

Figure 1.

Figure 1

Open in a new tab

Modified Sternberg working memory paradigm. Each trial was comprised of four phases: (1) a fixation phase which lasted for 1.0 s, (2) an encoding phase (red text) in which a grid containing six letters was presented for 2.0 s, (3) a maintenance phase (blue text) in which the letter stimuli were removed from the grid for 3.0 s, and (4) a retrieval phase in which one probe letter was presented for 0.9 s and the participant responded as to whether it had been included in the previous encoding set. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

MEG Data Acquisition

Recordings were conducted in a one‐layer magnetically shielded room with active shielding engaged. With an acquisition bandwidth of 0.1–330 Hz, neuromagnetic responses were sampled continuously at 1 kHz using an Elekta MEG system with 306 magnetic sensors (Elekta, Helsinki, Finland). MEG data from each subject were individually corrected for head motion and subjected to noise reduction using the signal space separation method with a temporal extension [Taulu and Simola, 2006; Taulu et al., 2005].

MEG Coregistration and Structural MRI Processing

Preceding MEG measurement, four coils were attached to the subject's head and localized, together with the three fiducial points and scalp surface, with a 3‐D digitizer (Fastrak 3SF0002, Polhemus Navigator Sciences, Colchester, VT). Once the subject was positioned for MEG recording, an electric current with a unique frequency label (e.g., 322 Hz) was fed to each of the coils. This induced a measurable magnetic field and allowed each coil to be localized in reference to the sensors throughout the recording session. Since coil locations were also known in head coordinates, all MEG measurements could be transformed into a common coordinate system. With this coordinate system, each participant's MEG data were coregistered with structural T1‐weighted MRI data prior to source space analyses using BESA MRI (Version 2.0; BESA GmbH, Gräfelfing, Germany). Structural MRI data were aligned parallel to the anterior and posterior commissures and were transformed into standardized space after beamforming.

MEG Time‐Frequency Transformation and Statistics

Cardio‐artifacts were removed from the data using signal‐space projection (SSP), which was accounted for during source reconstruction [Uusitalo and Ilmoniemi, 1997]. The continuous magnetic time series was divided into epochs of 6.9‐s duration, with the baseline being defined as −0.4 to 0.0 s before initial stimulus onset (see Fig. 1). Epochs containing artifacts were rejected based on a fixed threshold method, supplemented with visual inspection. Artifact‐free epochs were transformed into the time‐frequency domain using complex demodulation with a resolution of 1.0 Hz and 50 ms [Papp and Ktonas, 1977], and the resulting spectral power estimations per sensor were averaged over trials to generate time‐frequency plots of mean spectral density. These sensor‐level data were normalized by dividing the power value of each time‐frequency bin by the respective bin's baseline power, which was calculated as the mean power during the −0.4 to 0.0 s time period.

The time‐frequency windows used for imaging were determined by statistical analysis of the sensor‐level spectrograms across the entire array of gradiometers during the 5‐s “encoding” and “maintenance” time window. Each data point in the spectrogram was initially evaluated using a mass univariate approach based on the general linear model (GLM). To reduce the risk of false positive results while maintaining reasonable sensitivity, a two stage procedure was followed to control for Type 1 error. In the first stage, one‐sample t‐tests were conducted on each data point and the output spectrogram of t‐values was thresholded at P < 0.05 to define time‐frequency bins containing potentially significant oscillatory deviations across all participants. In stage two, time‐frequency bins that survived the threshold were clustered with temporally and/or spectrally neighboring bins that were also significant at the (P < 0.05) threshold, and a cluster value was derived by summing all of the t‐values of all data points in the cluster. Nonparametric permutation testing was then used to derive a distribution of cluster‐values and the significance level of the observed clusters (from stage one) were tested directly using this distribution [Ernst, 2004; Maris and Oostenveld, 2007]. For each comparison, at least 10,000 permutations were computed to build a distribution of cluster values. Based on these analyses, only the time‐frequency windows that contained significant oscillatory events across all younger and older participants during the encoding and maintenance phases were subjected to the beamforming (i.e., imaging) analysis. Thus, a purely data‐driven approach was utilized when selecting the time‐frequency windows to be imaged. This objective approach was employed due to the notable variability in the time‐frequency windows analyzed across various neurophysiological studies of working memory.

MEG Source Imaging and Statistics

Cortical networks were imaged through an extension of the linearly constrained minimum variance vector beamformer [Gross et al., 2001; Hillebrand et al., 2005; Liljestrom et al., 2005; Van Veen et al., 1997], which employs spatial filters in the frequency domain to calculate source power for the entire brain volume. The single images are derived from the cross spectral densities of all combinations of MEG gradiometers averaged over the time‐frequency range of interest, and the solution of the forward problem for each location on a grid specified by input voxel space. Following convention, the source power in these images was normalized per participant using a separately averaged pre‐stimulus noise period of equal duration and bandwidth [Hillebrand et al., 2005]. MEG pre‐processing and imaging used the Brain Electrical Source Analysis (version 6.0) software.

Normalized source power was computed for the selected time‐frequency bands over the entire brain volume per participant at 4.0 x 4.0 x 4.0 mm resolution. Preceding statistical analysis, each participant's functional images were transformed into standardized space using the transform that was previously applied to the structural images. The resulting 3D maps of brain activity were averaged across participants in each group to assess the neuroanatomical basis of significant oscillatory responses identified through the sensor‐level analysis. In addition, these images were statistically evaluated using a random effects, mass univariate approach based on the GLM. The effect of group (i.e., older vs. younger) was determined using two‐tailed independent‐samples t‐tests per time‐frequency bin. All output statistical maps were displayed as a function of alpha level, thresholded at P < 0.005, and adjusted for multiple comparisons using a spatial extent threshold (i.e., cluster restriction; k = 200 contiguous voxels) based on the theory of Gaussian random fields [Poline et al., 1995; Worsley et al., 1999; Worsley et al., 1996]. Of note, we also conducted nonparametric permutation testing to control for Type 1 error in the group comparison SPMs, using a cluster‐based method similar to that performed on the sensor‐level spectrograms, and our results were virtually identical (see Supporting Information).

Regression Analysis

Hierarchical regression analysis was conducted to examine statistical relationships between behavioral performance, oscillatory activity, and age. For each participant, accuracy was computed by dividing the number of correct trials by the total number of trials completed. The amplitude of the peak group difference voxel was obtained by identifying the time bin containing the largest difference in neural activation between older and younger participants and extracting the value of this voxel for each participant. Peak voxel value and age were centered and accuracy was regressed on both variables in the first block, and their interaction term in the second block. To understand the nature of the interaction, post‐hoc probing was executed using the approach developed by Aiken and West [1991].

RESULTS

Behavioral Analysis

All 31 participants successfully completed the task. Both older and younger participants performed well, accurately identifying the probe 81.32% (SD = 8.04%) and 84.30% (SD = 6.66%), respectively, of all trials. Only correct trials were used for analysis. There were no significant differences in accuracy between older and younger participants t(29) = 1.12, P = 0.270.

Sensor‐Level Analysis

Statistical analyses of the time‐frequency spectrograms revealed a significant cluster of alpha/low‐beta (9 − 16 Hz) oscillatory activity in central and posterior gradiometers across all participants (P < 0.001, corrected; Fig. 2). This alpha/beta activity began 0.2 s after onset of the encoding grid, was sustained throughout the encoding phase, and terminated at 2.5 s (i.e., 0.5 s into the maintenance phase). Additionally, a significant cluster of alpha (9 − 12 Hz) activity was observed during the maintenance phase in roughly the same group of sensors (P < 0.001, corrected; Fig. 2). This alpha activity began at 3.0 s (i.e., 1.0 s into the maintenance phase), was sustained throughout the remainder of the maintenance phase, and sharply dissipated early in the retrieval phase. Since these responses spanned across the encoding and maintenance phases, respectively, we split them into 0.4 s nonoverlapping time bins and subjected each time‐frequency window to beamformer analysis in each participant.

Figure 2.

Figure 2

Open in a new tab

Time‐frequency spectrogram with time shown on the x‐axis and frequency (Hz) denoted on the y‐axis. Percent power change was computed by dividing the mean power of each time‐frequency bin by the respective bin's baseline power (−0.4 to 0 s) and multiplying this value by 100. The color legend is displayed to the right. Data represent an averaged peak sensor, collapsed across age groups, located near the parieto‐occipital cortex. The chip where this sensor was precisely located is shown in the 2‐dimensional plot of the sensor array that appears on the far right. Strong alpha/beta desynchronization was observed shortly after the encoding grid was presented, and this evolved into a narrower alpha synchronization during maintenance. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Beamformer Analysis

To evaluate the neural dynamics serving working memory performance, we initially examined the time course of activity in each group (Fig. 3). These data indicated a strong decrease in alpha/beta activity beginning early in the encoding phase in the bilateral occipital cortices and cerebellum, which rapidly spread to left superior temporal regions, supramarginal, and inferior frontal gyri. Such decreases in alpha/beta activity were sustained throughout the encoding phase in left fronto‐temporal cortices, with the frontal activity beginning to dissipate early in the maintenance phase. Meanwhile, alpha activity in the left superior temporal cortices extended throughout the maintenance phase. Generally, the spatiotemporal dynamics of left fronto‐temporal activity were similar in younger and older adults (Fig. 3).

Figure 3.

Figure 3

Open in a new tab

Group mean beamformer images (pseudo‐t) at 70% transparency for younger (top) and older (bottom) adults are displayed across all time bins spanning the encoding (red labels) and maintenance (blue labels) phases. Across age groups, there was a strong sustained decrease in alpha/beta oscillatory activity in left hemispheric brain regions throughout most of the encoding and maintenance phases. This decrease began in posterior regions early in the encoding phase, spread anterior to include left fronto‐temporal areas, including language areas, during the latter half of encoding and early maintenance period, and dissipated slightly throughout the remainder of the maintenance phase. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

In regard to group differences, as hypothesized, we found significant age‐related effects in alpha/beta activity in the right PFC throughout the majority of the encoding and maintenance phases (P < 0.005, corrected; Fig. 4 and Table 1). During the latter half of encoding, older adults exhibited a largely sustained decrease in alpha/beta activity in the right inferior frontal gyrus (IFG) stretching from 1.0 to 3.0 s, which was clearly absent in younger adults. Later in the time course, we found a similar pattern of decreased alpha activity in the right IFG, right superior temporal gyrus (STG), and right middle frontal gyrus of older participants from 3.8 to 4.2 s (P < 0.005, corrected; Fig. 4). Finally, such decreased alpha activity in the older group was sustained throughout the remainder of the maintenance phase (i.e., until 5.0 s) in the right IFG/STG cluster. Additionally, in congruence with our hypotheses, we observed stronger alpha activity in the bilateral parieto‐occipital cortices of older participants during early maintenance (i.e., 2.6 − 3.0 s), with alpha transitioning from a desynchronization (decrease relative to baseline) to a synchronization (increase) in younger participants during this same temporal period (P < 0.005, corrected; see Fig. 5). Analysis of the dynamics indicated that younger participants eventually exhibited an increase in parieto‐occipital alpha during the maintenance phase, but this increase occurred earlier and encompassed more cortical tissue in older relative to younger participants.

Figure 4.

Figure 4

Open in a new tab

Significant group differences (P < .005) in alpha/beta oscillatory activity in the right hemisphere are displayed across all time bins spanning the encoding (red labels) and maintenance (blue labels) phases. Age‐related differences in the right angular gyrus began immediately following the onset of the encoding grid and spread to include right supramarginal gyrus (i.e., homologue of Wernicke's area) before terminating in the latter half of the encoding phase (see Fig. 6 for complimentary views). These group differences were attributable to an early decrease in alpha/beta activity in younger adults, which was not observed in older adults. Additionally, age‐related differences in the right inferior frontal gyrus (i.e., homologue of Broca's area) began in the latter half of the encoding phase at 1.0 s, and persisted throughout the majority of the remaining time bins, spreading to include superior temporal regions during the latter half of the maintenance phase. These differences reflect decreased alpha/beta activity in older adults, with stable activity in younger adults. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Table 1.

Group differences in neural oscillatory activity

Anatomical Label Peak Coordinates (x,y,z) Direction of Difference t‐value
Encoding 1
L Calcarine −10, −88, 11 ERD Amplitude: Young > Old 4.77
L Middle Occipital Gyrus −42, −75, 24 ERD Amplitude: Young > Old 3.46
R Middle Temporal Gyrus 56, −53, 10 ERD Amplitude: Young > Old 3.22
R Inferior Temporal Sulcus 61, −20, −17 ERD Amplitude: Young > Old 3.52
R Cerebellum 19, −86, −30 ERD Amplitude: Young > Old 3.72
R Superior Medial Frontal 12, 52, 42 ERD Amplitude: Young > Old 3.63
Encoding 2
R Middle Temporal Gyrus 53, −63, 4 ERD Amplitude: Young > Old 3.54
R Supramarginal Gyrus 49, −50, 18 ERD Amplitude: Young > Old 3.40
L Superior Temporal Gyrus −52, −39, 12 ERD Amplitude: Young > Old 3.88
L Supramarginal Gyrus −51, −41, 24 ERD Amplitude: Young > Old 3.97
L Inferior Temporal −46, −3, −26 ERD Amplitude: Young > Old 3.38
Encoding 3
R Inferior Frontal Gyrus 53, 1, 19 ERD Amplitude: Young < Old −3.45
Encoding 4
R Inferior Frontal Gyrus 44, −4, 21 ERD Amplitude: Young < Old −3.50
Transition
R Inferior Frontal Gyrus 40, −3, 20 ERD Amplitude: Young < Old −3.33
Maintenance 1
L Inferior Temporal −40, −33, −15 ERD Amplitude: Young < Old −3.20
L Supramarginal Gyrus −61, −29, 22 ERD Amplitude: Young < Old −3.36
R Inferior Frontal Gyrus 49, −6, 21 ERD Amplitude: Young < Old −3.73
Maintenance 2
L Inferior Occipital −11, −95, −9 ERS Amplitude: Young < Old 4.06
L Parietal‐Occipital Sulcus −11, −84, 34 ERS Amplitude: Young < Old 4.48
R Superior Parietal 8, −67, 57 ERS Amplitude: Young < Old 4.85
R Inferior Frontal Gyrus 54, 4, 16 ERD Amplitude: Young < Old −3.86
Maintenance 5
R Prefrontal Cortex 47, 2, 47 ERD Amplitude: Young < Old −3.74
R Inferior Frontal Gyrus 52, 7, 4 ERD Amplitude: Young < Old −3.31
R Inferior Frontal Gyrus 44, 14, 23 ERD Amplitude: Young < Old −3.85
Maintenance 6
R Inferior Frontal Gyrus 53, 8, 2 ERD Amplitude: Young < Old −3.26
Maintenance 7
R Superior Temporal Gyrus 53, 6, −4 ERD Amplitude: Young < Old −3.96
Open in a new tab

Note. N = 31. All peaks significant at p < .005, corrected. ERD and ERS refer to event‐related desynchronization and synchronization, respectively.

Figure 5.

Figure 5

Open in a new tab

Group mean beamformer images (pseudo‐t) for younger and older adults are displayed on the top and middle rows, respectively, for the first three time bins of the maintenance phase. The bottom row displays significant group differences (P < 0.005, corrected) in alpha oscillatory activity. Younger adults showed a shift from a bilateral decrease in alpha activity in the occipital cortices to a right‐lateralized increase in alpha activity in parieto‐occipital areas. Conversely, older adults demonstrated a sustained increase in alpha activity across bilateral occipital cortices, with the increased alpha occurring earlier and involving more cortical tissue in older adults relative to younger adults. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

We also found significant age‐related differences beyond those which were anticipated. Specifically, during early encoding, younger adults exhibited decreased alpha/beta activity in the right inferior temporal sulcus and temporoparietal area (0.2 − 0.6 s; P < 0.005, corrected; Fig. 4), as well as a much stronger decrease in alpha/beta activity in bilateral occipital regions and cerebellar cortices during most of the encoding phase (0.2–1.4 s; P < 0.005, corrected). Younger adults also had a stronger decrease in alpha/beta activity during encoding in bilateral supramarginal and angular gyri (Fig. 6), left posterior superior temporal regions (0.2–1.0 s), and the left inferior temporal sulcus/gyrus (0.6–1.0 s; P < 0.005, corrected). Interestingly, slightly later during the early maintenance phase, older adults had a stronger decrease in oscillatory activity in largely the same left supramarginal and posterior temporal regions (P < 0.005, corrected).

Figure 6.

Figure 6

Open in a new tab

Significant group differences (P < 0.005) in alpha/beta oscillatory activity in bilateral supramarginal and angular gyri (i.e., Wernicke's area and its right hemisphere homologue) during early encoding (0.6–1.0 s). The left panel shows a sagittal view of left hemispheric differences, while the right panel depicts a coronal view of the simultaneous bilateral differences (see Fig. 4, time bin 0.6–1.0 s for a sagittal view of right hemispheric differences). These group differences were attributable to a decrease in alpha/beta activity in younger adults, which was not observed in older adults. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Regression Analysis

To identify whether age and region‐specific oscillatory activity were predictive of behavioral performance, multiple regression was performed. For this analysis, we focused on the right PFC, as the largest group difference occurred in this region during the 4.6 − 5.0 s time window (see Fig. 4), and oscillatory differences in this brain area were likely our most important finding. To this end, we extracted the value of the peak voxel (coordinates: 53, 11, 2) for each participant. These data and participant age were statistically centered and prepared for regression analysis (see Methods).

The interaction between age and neural activity in the right PFC for accuracy was significant, ΔR 2 = .13, ΔF = 4.87, P = 0.036, β = 0.40, b = 0.03, t(27) = 2.21, CI [0.002, 0.052], above and beyond the effects of age and peak voxel value (see Table 2). Age significantly moderated the relationship between right PFC activity and accuracy, and this interaction accounted for 13% of the variance in accuracy. The full model accounted for 30% of the variance in accuracy, R 2 = 0.30, F(3, 27) = 3.79, P = 0.022.

Table 2.

Hierarchical multiple regression results for accuracy on peak voxel value (centered), age (centered), and the interaction term

Model b SE t β F R2 ΔF ΔR 2 95% CI
Additive Model:
Intercept 82.86 1.25 66.13*** 2.85 .17 (80.29, 85.42)
Peak Voxel −0.30 0.24 −1.25 −.26 (−0.78, 0.19)
Age −0.20 0.08 −2.39* −.50 (−0.37, −0.03)
Interaction Model:
Intercept 84.63 1.42 59.44*** 3.79* .30 4.87* .13* (81.71, 87.56)
Peak Voxel −0.53 0.24 −2.16* −.47 (−1.03, −0.03)
Age −0.23 0.08 −2.87** −.57 (−0.40, −0.07)
Interaction 0.03 0.01 2.21* .40 (0.002, 0.052)
Open in a new tab

N = 31. CI, confidence interval. *P < 0.05. ** P < 0.01. *** P < 0.001.

Post‐hoc probing of the interaction demonstrated that at younger ages, right PFC activity significantly and uniquely predicted accuracy, β = −0.91, b = −1.02, t(27) = −2.58, P = 0.016, CI [−1.83, −0.21] (see Table 3), such that as peak alpha activity decreased in this brain region, accuracy on the working memory task was expected to increase (see Fig. 7). At older ages, the relationship between right PFC activity and accuracy was not significant, β = −0.03, b = −0.04, t(27) = −0.15, P = 0.884, CI [−0.55, 0.48] (see Table 3).

Table 3.

Multiple regression results for accuracy on peak voxel value (centered), age below or age above, and the interaction term

Model b SE t Β F R2 95% CI
Younger Age:
Intercept 88.86 2.14 41.53*** 3.79* 0.30 (84.47, 93.24)
Peak Voxel −1.02 0.40 −2.58* −0.91 (−1.83, −0.21)
AgeBelow −0.23 0.08 −2.87** −0.57 (−0.40, −0.07)
InteractBelow 0.03 0.01 2.21* 0.70 (0.002, 0.052)
Older Age:
Intercept 80.41 1.95 41.25*** 3.79* 0.30 (76.41, 84.41)
Peak Voxel −0.04 0.25 −0.15 −0.03 (−0.55, 0.48)
AgeAbove −0.23 0.08 −2.87** −0.57 (−0.40, −0.07)
InteractAbove 0.03 0.01 2.21* 0.45 (0.002, 0.052)
Open in a new tab

N = 31. CI, confidence interval. *P < 0.05. **P < 0.01. ***P < 0.001.

Figure 7.

Figure 7

Open in a new tab

Line graph depicting the interaction between age and right prefrontal cortex (PFC; i.e., homologue of Broca's area) oscillatory activity for accuracy on the working memory task. Peak voxel amplitude is depicted on the x‐axis, with decreased activity signifying lower peak voxel values and increased activity signifying higher peak voxel values. Accuracy is on the y‐axis in percent units. Younger ages (i.e., one standard deviation below age) are depicted by the black solid line, while older ages (i.e., one standard deviation above age) are depicted by the black dashed line. At younger ages, right PFC activity was significantly predictive of accuracy (controlling for age and the interaction term) such that, as activity decreased, accuracy was expected to increase. At older ages, right PFC activity and accuracy were not significantly related (controlling for age and the interaction term).

DISCUSSION

In this study, we utilized high‐density MEG to investigate age‐related differences in the neural oscillatory dynamics of the encoding and maintenance operations of working memory processing. In each age group, our data indicated a sharp decrease in alpha/beta activity that began in bilateral occipital regions immediately during encoding. This decrease in alpha/beta activity spread anterior to include left temporal and frontal cortices during the latter half of encoding, narrowed to the alpha band and grew in magnitude during early maintenance, and dissipated throughout the latter half of maintenance. This pattern of decreased left hemispheric oscillatory activity was remarkably similar in younger and older adults. This work corroborates and extends previous imaging work, which has shown that verbal working memory activates a network of left hemispheric language regions. As for age‐related differences, the neural dynamics distinguished younger and older participants in the left hemispheric regions noted above, as well as parieto‐occipital areas, but our most important findings involved homologous regions in the right hemisphere. Essentially, differences emerged in the right STG and supramarginal gyrus (i.e., homologue of Wernicke's area) during early encoding, and later older adults exhibited a sharp, sustained decrease in the right IFG (i.e., homologue of Broca's area). Finally, our results indicated that oscillatory amplitude in the right PFC was a strong predictor of accuracy in younger, but not older adults, which provides new supporting data for theories of how aging modulates functional brain networks. Below, we discuss the implications of these findings for understanding the dynamics of working memory processing, and the modulatory effects that healthy aging has on the inherent function and dynamics in these networks.

The extensive oscillatory responses observed in the left hemisphere were not surprising, especially when considering Baddeley's neurocognitive model of working memory, which divides the working memory construct into four subcomponents [Baddeley, 1992; Baddeley, 2000]. Previous neuroimaging work has suggested that certain neuroanatomical locations underlie the functioning of these four subsystems. The phonological loop is believed to be tied to language areas in the left hemisphere, specifically Wernicke's area which encompasses left inferior parietal and posterior temporal regions. This area is thought to be involved in the temporary storage of verbal information (i.e., the phonological store), and has been widely implicated in language comprehension [Cabeza and Nyberg, 2000; D'Esposito, 2007; Wilson et al., 2005a, 2005b, 2007]. Left prefrontal regions (e.g., Broca's area) are also believed to be critical for language processing and the maintenance of verbal information (i.e., the articulatory loop) during working memory performance [Cabeza and Nyberg, 2000; D'Esposito, 2007; D'Esposito and Postle, 2015; Reuter‐Lorenz et al., 2000]. Considering our task involved language‐related stimuli (i.e., letters), oscillatory responses in both the left PFC and Wernicke's area were expected throughout much of encoding and maintenance, and we observed such in both age groups, although these responses were slightly delayed in the older group. The involvement of these brain areas is consistent with another MEG study, which reported decreased activity in Wernicke's area, as well as the medial frontal cortex and bilateral frontal operculum during a Sternberg task [Brookes et al., 2011]. Additionally, our results complement two fMRI meta‐analyses that suggested greater activity in left hemispheric language areas during performance on verbal‐based memory tasks [Cabeza and Nyberg, 2000; Rottschy et al., 2012]. Importantly, multiple EEG‐fMRI studies have demonstrated negative correlations between alpha oscillatory activity and fMRI activation during cognitive tasks in general [Murta et al., 2015; Scheeringa et al., 2011], and specifically between decreased alpha/low‐beta activity and increased fMRI activation in language areas during a working memory task (Michels et al., 2010). Thus, our results further support the role of these areas in verbal working memory performance across age groups.

This study's most important findings were likely the age‐related differences in alpha/beta (9 − 16 Hz) oscillatory activity observed in the right IFG and STG during encoding and maintenance. Essentially, older adults exhibited a sustained decrease in alpha/beta activity during the second half of encoding in the right IFG, which narrowed to the alpha band (9 − 12 Hz) and spread to encompass the right STG during late maintenance; such oscillatory responses were not observed in younger adults during these spatiotemporal windows. Thus, during the latter half of encoding and through most of maintenance, both groups exhibited decreased power in Broca's area, but only older adults also had decreased activity in its right‐hemisphere homologue. These results are exciting in that they are largely congruent with an influential hypothesis concerning neurocognitive aging: the compensation hypothesis. This well‐supported hypothesis [Cabeza, 2002; Cabeza et al., 2002, 2004, 1997; Davis et al., 2008; Eyler et al., 2011; Fakhri et al., 2012; Grady, 2012; Grady et al., 2005; Greenwood, 2007; Mattay et al., 2006; Park and Reuter‐Lorenz, 2009; Reuter‐Lorenz et al., 2000, 2008; Saliasi et al., 2014; Spreng et al., 2010; Turner and Spreng, 2012] posits that when older adults engage greater volumes of brain tissue relative to younger adults, particularly in regards to the PFC, the increase in neural activity serves as a compensatory mechanism to aid cognitive performance and counteract age‐related declines [Cabeza, 2002; Reuter‐Lorenz et al., 2000]. Oftentimes, such age‐related differences are strongest in the hemisphere that is nondominant for the specific cognitive function (e.g., right hemisphere in language tasks), and thus activation appears more bilateral in the older adults [Cabeza, 2002; Cabeza et al., 2002; Emery et al., 2008; Reuter‐Lorenz et al., 2000; Spreng et al., 2010]. The functional significance of the decreased alpha activity we observed in these critical brain regions also aligns with the compensation hypothesis, as decreased alpha activity has been theorized to reflect a release from inhibition and/or engagement of the specific area during a task [Klimesch, 2012; Klimesch et al., 2007]. Thus, the persistent and strong decrease in alpha activity we observed in older adults in not only Broca's area (i.e., an area associated with the articulatory loop), but also its right hemisphere homologue, likely reflected the engagement of compensatory resources for the rehearsal and maintenance of the information to be remembered.

Given the influential role of the PFC in theories of neurocognitive aging, we aimed to more closely examine its relationship with behavioral performance and age, as the compensation hypothesis is most strongly supported when behavioral performance remains equal between groups, despite older adults engaging more neural resources than younger adults [Grady, 2012], which occurred in the present study. Thus, we performed a hierarchical regression analysis to determine if age differentially moderated the relationship between oscillatory activity in the right IFG (i.e., loci of largest group difference) and behavioral performance. Note that these results should be interpreted with caution, as we did not test all brain regions where group differences were observed, and consequently there is some potential for bias in the interaction analysis by selecting the area which demonstrated the greatest age‐related difference. Nevertheless, our results were somewhat surprising in that a significant relationship was observed for only the younger group, wherein greater decreases in alpha activity within the right IFG were predictive of better accuracy. We propose that these data are consistent with the compensation‐related utilization of neural circuits hypothesis (CRUNCH; [Reuter‐Lorenz, 2008]. CRUNCH extends upon the compensation hypothesis by clarifying that, under lower cognitive demands, older adults engage greater volumes of cortical tissue during task performance relative to younger adults, which aids in successful performance. However, under higher demands, older adults have already exhausted their compensatory circuits and reached a resource ceiling, resulting in poorer task performance. Meanwhile, younger adults are able to engage these compensatory circuits to meet the increased cognitive demands. Support for CRUNCH comes from neuroimaging studies that showed that age‐related differences in PFC activation varied with task demands [Cappell et al., 2010; Mattay et al., 2006]. For example, Cappell et al. [2010] found that at a moderate memory load (i.e., 5 letters), older adults displayed an over‐activation within the right dorsolateral PFC relative to younger adults, but at a higher memory load (i.e., 7 letters) older adults exhibited an under‐activation within the same region relative to younger adults. Our task used a 6‐letter memory load, which is at the more demanding end for a Sternberg task. Thus, a neural resources ceiling may have been reached by the older adults, accounting for the plateau that we observed between neural activity and accuracy in this group. However, additional neural resources were likely still available in the younger group, resulting in the relationship we observed between neural activity and accuracy in this group.

In addition to the age‐related differences observed in anterior regions, we hypothesized differences in parieto‐occipital cortices, which were also supported by our results. Specifically, age‐related differences in oscillatory activity were found in bilateral parieto‐occipital areas during early maintenance. Older adults exhibited a sharp increase in parieto‐occipital alpha activity during early maintenance, which continued throughout maintenance, dissipating to only right‐lateralized regions towards retrieval. Younger adults exhibited a similar increase in parieto‐occipital alpha during maintenance, but the onset was delayed and the activity was more lateralized. Thus, parieto‐occipital alpha activity began earlier and engaged more neural tissue during maintenance in older relative to younger adults. Previous neurophysiological studies have reported similar parieto‐occipital alpha oscillations during working memory maintenance in healthy young adults [Bennett et al., 2013; Bonnefond and Jensen, 2012, 2013; Heinrichs‐Graham and Wilson, 2015; Jensen et al., 2002; Jiang et al., 2015; McDermott et al., 2016; Tuladhar et al., 2007; Wilson et al., 2016]. An influential theory posits that such alpha activity reflects the inhibition of the dorsal visual stream, which functions to protect items (e.g., letters) that are being retained in more anterior regions (e.g. Broca's area) from being disturbed by incoming visual information [Bonnefond and Jensen, 2012; Jensen et al., 2002; Jensen and Mazaheri, 2010; Klimesch et al., 2007; Tuladhar et al., 2007]. In line with this, we propose that the earlier and more widespread parieto‐occipital alpha activity that we observed in older relative to younger adults reflects an additional compensatory mechanism, which further aids working memory performance in older age.

Before closing, it is important to acknowledge the limitations of this study. First, several electrophysiological studies of working memory have reported theta activity [Brookes et al., 2011; Jensen and Tesche, 2002; Onton et al., 2005], whereas we did not find significant theta. Interestingly, the previous studies that reported frontal theta used sequentially presented stimuli, whereas our stimuli were presented simultaneously and this may account for the divergence in theta results. Additionally, the simultaneous presentation of our stimuli may have been more susceptible to eye movements during the encoding period, relative to sequential designs. However, working memory meta‐analyses and reviews, which incorporated studies that used sequential designs, reported similar patterns of activity regarding left hemispheric language regions [Cabeza and Nyberg, 2000; Hill et al., 2014; Nee et al., 2013; Owen et al., 2005; Rottschy et al., 2012] and bilateral activity in older adults [Grady, 2012; Spreng et al., 2010]. This converging evidence, coupled with the fact that we did not observe effects in the frontal eye fields nor the orbitofrontal cortices, suggests that our results were not contaminated by eye movement artifacts. Lastly, our study was limited in its restriction to only adult males, although this was by design. A recent meta‐analysis highlighted sex‐related differences in working memory processing and we aimed to avoid such confounding variables in the current study [Hill et al., 2014]. Future MEG studies of working memory should focus on female participants.

CONCLUSIONS

The present study utilized advanced MEG imaging of the neural oscillatory activity that underlies working memory processing to provide novel insight into age‐related differences in the dynamics of encoding and maintenance. The results of this study were the first to identify the neural dynamics evolving during successful working memory encoding and maintenance in an older adult sample, and to demonstrate that aging modulates the relationship between oscillatory activity and behavioral working memory performance. These results provide invaluable insight into the spatiotemporal dynamics that serve working memory functions in older adults, as well as how healthy aging modulates functional brain networks. Additionally, our results provided partial support for both the compensation hypothesis and the CRUNCH.

Supporting information

Supporting Information

Click here for additional data file. (123.4KB, pdf)

ACKNOWLEDGMENTS

The Center for Magnetoencephalography at the University of Nebraska Medical Center was founded through an endowment from an anonymous donor. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Financial disclosures: No conflicts of interest, financial or otherwise, are declared by the authors.

REFERENCES

  1. Aiken LS, West SG (1991): Interactions Between Continuous Predictors in Multiple Regression. Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA: Sage Publications; pp 9–27. [Google Scholar]
  2. Baddeley A (1992): Working memory. Science 255:556–559. [DOI] [PubMed] [Google Scholar]
  3. Baddeley A (2000): The episodic buffer: A new component of working memory? Trends Cogn Sci 4:417–423. [DOI] [PubMed] [Google Scholar]
  4. Bennett IJ, Rivera HG, Rypma B (2013): Isolating age‐group differences in working memory load‐related neural activity: Assessing the contribution of working memory capacity using a partial‐trial fMRI method. NeuroImage 72:20–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bonnefond M, Jensen O (2012): Alpha oscillations serve to protect working memory maintenance against anticipated distracters. Current Biol 22:1969–1974. [DOI] [PubMed] [Google Scholar]
  6. Bonnefond M, Jensen O (2013): The role of gamma and alpha oscillations for blocking out distraction. Commun Integrat Biol 6:e22702. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brookes MJ, Wood JR, Stevenson CM, Zumer JM, White TP, Liddle PF, Morris PG (2011): Changes in brain network activity during working memory tasks: A magnetoencephalography study. NeuroImage 55:1804–1815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cabeza R (2002): Hemispheric asymmetry reduction in older adults: The HAROLD model. Psychol Aging 17:85–100. [DOI] [PubMed] [Google Scholar]
  9. Cabeza R, Anderson ND, Locantore JK, McIntosh AR (2002): Aging gracefully: Compensatory brain activity in high‐performing older adults. NeuroImage 17:1394–1402. [DOI] [PubMed] [Google Scholar]
  10. Cabeza R, Daselaar SM, Dolcos F, Prince SE, Budde M, Nyberg L (2004): Task‐independent and task‐specific age effects on brain activity during working memory, visual attention and episodic retrieval. Cerebral Cortex 14:364–375. [DOI] [PubMed] [Google Scholar]
  11. Cabeza R, Grady CL, Nyberg L, McIntosh AR, Tulving E, Kapur S, Jennings JM, Houle S, Craik FI (1997): Age‐related differences in neural activity during memory encoding and retrieval: A positron emission tomography study. J Neurosci 17:391–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cabeza R, Nyberg L (2000): Imaging cognition II: An empirical review of 275 PET and fMRI studies. J Cogn Neurosci 12:1–47. [DOI] [PubMed] [Google Scholar]
  13. Cappell KA, Gmeindl L, Reuter‐Lorenz PA (2010): Age differences in prefontal recruitment during verbal working memory maintenance depend on memory load. Cortex 46:462–473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. D'Esposito M (2007): From cognitive to neural models of working memory. Phil Trans Roy Soc Lond 362:761–772. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. D'Esposito M, Postle BR. 2015. The cognitive neuroscience of working memory. Ann Rev Psychol 66:115–142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Davis SW, Dennis NA, Daselaar SM, Fleck MS, Cabeza R (2008): Que PASA? The posterior‐anterior shift in aging. Cerebral Cortex 18:1201–1209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Emery L, Heaven TJ, Paxton JL, Braver TS (2008): Age‐related changes in neural activity during performance matched working memory manipulation. NeuroImage 42:1577–1586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Ernst MD (2004): Permutation methods: A basis for exact inference. Stat Sci 19:676–685. [Google Scholar]
  19. Eyler LT, Sherzai A, Kaup AR, Jeste DV (2011): A review of functional brain imaging correlates of successful cognitive aging. Biol Psychiatry 70:115–122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Fakhri M, Sikaroodi H, Maleki F, Ali Oghabian M, Ghanaati H (2012): Age‐related frontal hyperactivation observed across different working memory tasks: An fMRI study. Behav Neurol 25:351–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Grady C (2012): The cognitive neuroscience of ageing. Nat Rev 13:491–505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Grady CL, McIntosh AR, Craik FI (2005): Task‐related activity in prefrontal cortex and its relation to recognition memory performance in young and old adults. Neuropsychologia 43:1466–1481. [DOI] [PubMed] [Google Scholar]
  23. J Gross, J Kujala, M Hamalainen, L Timmermann, A Schnitzler, R Salmelin (2001): Dynamic imaging of coherent sources: Studying neural interactions in the human brain. Proc Natl Acad Sci U S A. 98(2):694–699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Greenwood PM (2007): Functional plasticity in cognitive aging: Review and hypothesis. Neuropsychology 21:657–673. [DOI] [PubMed] [Google Scholar]
  25. Heinrichs‐Graham E, Wilson TW (2015): Spatiotemporal oscillatory dynamics during the encoding and maintenance phases of a visual working memory task. Cortex 69:121–130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hill AC, Laird AR, Robinson JL (2014): Gender differences in working memory networks: A BrainMap meta‐analysis. Biol Psychol 102:18–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hillebrand A, Singh KD, Holliday IE, Furlong PL, Barnes GR (2005): A new approach to neuroimaging with magnetoencephalography. Hum Brain Mapp 25:199–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Jensen O, Gelfand J, Kounios J, Lisman JE (2002): Oscillations in the alpha band (9‐12 Hz) increase with memory load during retention in a short‐term memory task. Cerebral Cortex 12:877–882. [DOI] [PubMed] [Google Scholar]
  29. Jensen O, Mazaheri A (2010): Shaping functional architecture by oscillatory alpha activity: Gating by inhibition. Front Hum Neurosci 4:186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Jensen O, Tesche CD (2002): Frontal theta activity in humans increases with memory load in a working memory task. Eur J Neurosci 15:1395–1399. [DOI] [PubMed] [Google Scholar]
  31. Jiang H, van Gerven MA, Jensen O (2015): Modality‐specific alpha modulations facilitate long‐term memory encoding in the presence of distracters. J Cogn Neurosci 27:583–592. [DOI] [PubMed] [Google Scholar]
  32. Karrasch M, Laine M, Rapinoja P, Krause CM (2004): Effects of normal aging on event‐related desynchronization/synchronization during a memory task in humans. Neurosci Lett 366:18–23. [DOI] [PubMed] [Google Scholar]
  33. Klimesch W (2012): Alpha‐band oscillations, attention, and controlled access to stored information. Trends Cogn Sci 16:606–617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Klimesch W, Sauseng P, Hanslmayr S (2007): EEG alpha oscillations: The inhibition‐timing hypothesis. Brain Res Revi 53:63–88. [DOI] [PubMed] [Google Scholar]
  35. Liljestrom M, Kujala J, Jensen O, Salmelin R (2005): Neuromagnetic localization of rhythmic activity in the human brain: A comparison of three methods. NeuroImage 25:734–745. [DOI] [PubMed] [Google Scholar]
  36. Maris E, Oostenveld R (2007): Nonparametric statistical testing of EEG‐ and MEG‐data. J Neurosci Methods 164:177–190. [DOI] [PubMed] [Google Scholar]
  37. Mattay VS, Fera F, Tessitore A, Hariri AR, Berman KF, Das S, Meyer‐Lindenberg A, Goldberg TE, Callicott JH, Weinberger DR (2006): Neurophysiological correlates of age‐related changes in working memory capacity. Neurosci Lett 392:32–37. [DOI] [PubMed] [Google Scholar]
  38. TJ McDermott, AS Badura‐Brack, KM Becker, TJ Ryan, MM Khanna, E Heinrichs‐Graham, TW Wilson (2016): Male veterans with PTSD exhibit aberrant neural dynamics during working memory processing: an MEG study. J Psychiatry Neurosci. 2015 Dec 7;41(1):150058. doi: 10.1503/jpn.150058. [Epub ahead of print]. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Michels L, Bucher K, Luchinger R, Klaver P, Martin E, Jeanmonod D, Brandeis D (2010): Simultaneous EEG‐fMRI during a working memory task: Modulations in low and high frequency bands. PloS One 5:e10298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Murta T, Leite M, Carmichael DW, Figueiredo P, Lemieux L (2015): Electrophysiological correlates of the BOLD signal for EEG‐informed fMRI. Hum Brain Mapp 36:391–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Nee DE, Brown JW, Askren MK, Berman MG, Demiralp E, Krawitz A, Jonides J (2013): A meta‐analysis of executive components of working memory. Cerebral Cortex 23:264–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Onton J, Delorme A, Makeig S (2005): Frontal midline EEG dynamics during working memory. NeuroImage 27:341–356. [DOI] [PubMed] [Google Scholar]
  43. Owen AM, McMillan KM, Laird AR, Bullmore E (2005): N‐back working memory paradigm: A meta‐analysis of normative functional neuroimaging studies. Hum Brain Mapp 25:46–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Papp N, Ktonas P (1977): Critical evaluation of complex demodulation techniques for the quantification of bioelectrical activity. Biomed Sci Instr 13:135–145. [PubMed] [Google Scholar]
  45. Park DC, Lautenschlager G, Hedden T, Davidson NS, Smith AD, Smith PK (2002): Models of visuospatial and verbal memory across the adult life span. Psychol Aging 17:299–320. [PubMed] [Google Scholar]
  46. Park DC, Reuter‐Lorenz P (2009): The adaptive brain: Aging and neurocognitive scaffolding. Annual Rev Psychol 60:173–196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Poline JB, Worsley KJ, Holmes AP, Frackowiak RS, Friston KJ (1995): Estimating smoothness in statistical parametric maps: variability of p values. J Comp Assisted Tomography 19:788–796. [DOI] [PubMed] [Google Scholar]
  48. Reuter‐Lorenz PA, Jonides J, Smith EE, Hartley A, Miller A, Marshuetz C, Koeppe RA (2000): Age differences in the frontal lateralization of verbal and spatial working memory revealed by PET. J Cogn Neurosci 12:174–187. [DOI] [PubMed] [Google Scholar]
  49. Reuter‐Lorenz PACKA (2008): Neurocognitive aging and the compensation hypothesis. Current Directions Psychol Sci 17:177–182. [Google Scholar]
  50. Rottschy C, Langner R, Dogan I, Reetz K, Laird AR, Schulz JB, Fox PT, Eickhoff SB (2012): Modelling neural correlates of working memory: A coordinate‐based meta‐analysis. NeuroImage 60:830–846. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Rypma B, D'Esposito M (2000): Isolating the neural mechanisms of age‐related changes in human working memory. Nat Neurosci 3:509–515. [DOI] [PubMed] [Google Scholar]
  52. Rypma B, Eldreth DA, Rebbechi D (2007): Age‐related differences in activation‐performance relations in delayed‐response tasks: A multiple component analysis. Cortex 43:65–76. [DOI] [PubMed] [Google Scholar]
  53. Saliasi E, Geerligs L, Lorist MM, Maurits NM (2014): Neural correlates associated with successful working memory performance in older adults as revealed by spatial ICA. PloS One 9:e99250. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Salthouse TA, Babcock RL, Shaw RJ (1991): Effects of adult age on structural and operational capacities in working memory. Psychol Aging 6:118–127. [DOI] [PubMed] [Google Scholar]
  55. Scheeringa R, Fries P, Petersson KM, Oostenveld R, Grothe I, Norris DG, Hagoort P, Bastiaansen MC (2011): Neuronal dynamics underlying high‐ and low‐frequency EEG oscillations contribute independently to the human BOLD signal. Neuron 69:572–583. [DOI] [PubMed] [Google Scholar]
  56. Solesio‐Jofre E, Lorenzo‐Lopez L, Gutierrez R, Lopez‐Frutos JM, Ruiz‐Vargas JM, Maestu F (2011): Age effects on retroactive interference during working memory maintenance. Biol Psychol 88:72–82. [DOI] [PubMed] [Google Scholar]
  57. Spreng RN, Wojtowicz M, Grady CL (2010): Reliable differences in brain activity between young and old adults: A quantitative meta‐analysis across multiple cognitive domains. Neurosci Biobehav Rev 34:1178–1194. [DOI] [PubMed] [Google Scholar]
  58. Taulu S, Simola J (2006): Spatiotemporal signal space separation method for rejecting nearby interference in MEG measurements. Phys Med Biol 51:1759–1768. [DOI] [PubMed] [Google Scholar]
  59. Taulu S, Simola J, Kajola M (2005): Applications of the signal space separation method (SSS). IEEE Trans Signal Process 53:3359–3372. [Google Scholar]
  60. Tuladhar AM, ter Huurne N, Schoffelen JM, Maris E, Oostenveld R, Jensen O (2007): Parieto‐occipital sources account for the increase in alpha activity with working memory load. Hum Brain Mapp 28:785–792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Turner GR, Spreng RN (2012): Executive functions and neurocognitive aging: Dissociable patterns of brain activity. Neurobiol Aging 33:826 e1‐13. [DOI] [PubMed] [Google Scholar]
  62. Uusitalo MA, Ilmoniemi RJ (1997): Signal‐space projection method for separating MEG or EEG into components. Med Biol Eng Comp 35:135–140. [DOI] [PubMed] [Google Scholar]
  63. Van Veen BD, van Drongelen W, Yuchtman M, Suzuki A (1997): Localization of brain electrical activity via linearly constrained minimum variance spatial filtering. IEEE Trans Bio‐Med Eng 44:867–880. [DOI] [PubMed] [Google Scholar]
  64. TW Wilson, E Heinrichs‐Graham, AL Proskovec, TJ McDermott (2016): Neuroimaging with magnetoencephalography: A dynamic view of brain pathophysiology. Transl Res 2016 Jan 25. pii: S1931‐5244(16)00031‐1. doi: 10.1016/j.trsl.2016.01.007. [Epub ahead of print] Review. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. TW Wilson, AC Leuthold, SM Lewis, AP Georgopoulos, PJ Pardo (2005a): Cognitive dimensions of orthographic stimuli affect occipitotemporal dynamics. Exp Brain Res 167(2):141–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. TW Wilson, AC Leuthold, SM Lewis, AP Georgopoulos, PJ Pardo (2005b): The time and space of lexicality: A neuromagnetic view. Exp Brain Res 162(1):1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. TW Wilson, AC Leuthold, JE Moran, PJ Pardo, SM Lewis, AP Georgopoulos (2007): Reading in a deep orthography: Neuromagnetic evidence for dual‐mechanisms. Exp Brain Res 180(2):247–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Worsley KJ, Andermann M, Koulis T, MacDonald D, Evans AC (1999): Detecting changes in nonisotropic images. Hum Brain Mapp 8:98–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Worsley KJ, Marrett S, Neelin P, Vandal AC, Friston KJ, Evans AC (1996): A unified statistical approach for determining significant signals in images of cerebral activation. Hum Brain Mapp 4:58–73. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information

Click here for additional data file. (123.4KB, pdf)

Articles from Human Brain Mapping are provided here courtesy of Wiley

RESOURCES