Complex span tasks and hippocampal recruitment during working memory

The working memory (WM) system is vital to performing everyday functions that require attentive, non-automatic processing of information. However, its interaction with long term memory (LTM) is highly debated. Here, we used fMRI to examine whether a popular complex WM span task, thought to force the displacement of to-be-remembered items in the focus of attention to LTM, recruited medial temporal regions typically associated with LTM functioning to a greater extent and in a different manner than traditional neuroimaging WM tasks during WM encoding and maintenance. fMRI scans were acquired while participants performed the operation span (OSPAN) task and an arithmetic task. Results indicated that performance of both tasks resulted in significant activation in regions typically associated with WM function. More importantly, significant bilateral activation was observed in the hippocampus, suggesting it is recruited during WM encoding and maintenance. Right posterior hippocampus activation was greater during OSPAN than arithmetic. Persitimulus graphs indicate a possible specialization of function for bilateral posterior hippocampus and greater involvement of the left for WM performance. Recall time-course activity within this region hints at LTM involvement during complex span.     

fMRI-acoustic noise alters brain activation during working memory tasks

Scanner noise during functional magnetic resonance imaging (fMRI) may interfere with brain function and change blood oxygenation level dependent (BOLD) signals, a problem that generally worsens at the higher field strengths. Therefore, we studied the effect of increased acoustic noise on fMRI during verbal working memory (WM) processing. The sound pressure level of scanner noise was increased by 12 dBA from "Quiet" to "Loud" echo planar imaging (EPI) scans by utilizing resonant vibration modes of the gradient coil. A WM paradigm with graded levels of task difficulty was used to further access WM load. Increased scanner noise produced increased BOLD responses (percent signal change) bilaterally in the cerebellum, inferior (IFG), medial (medFG), and superior (SFG) frontal, fusiform (FusG), and the lingual (LG) gyri, and decreased BOLD responses bilaterally in the anterior cingulate gyrus (ACG) and the putamen. This finding suggests greater recruitment of attention resources in these brain regions, probably to compensate for interference due to louder scanner noise. Increased working memory load increased the BOLD signals in IFG and the cerebellum, but decreased the BOLD signals in the putamen and the LG. These findings also support the idea that brain function requires additional attention resources under noisier conditions. Load- and acoustic-noise-related changes in BOLD responses correlated negatively in the WM network. This study demonstrates that MR noise affects brain activation pattern. Future comparisons between studies performed under different acoustic conditions (due to differing magnetic field strengths, pulse sequences, or scanner manufacturers) might require knowledge of the sound pressure level of acoustic noise during fMRI.     

Temporal dynamics of basal ganglia response and connectivity during verbal working memory

Research on the neural basis of working memory (WM) has generally focused on neocortical regions; comparatively little is known about the role of subcortical structures. There is growing evidence that the basal ganglia are involved in WM, but their contribution to different component processes of WM is poorly understood. We examined the temporal dynamics of basal ganglia response and connectivity during the encoding, maintenance and response phases of a Sternberg WM task. During the encoding and maintenance phases, WM-load-dependent activation was observed in the left anterior caudate, anterior putamen and globus pallidus; activation in the right anterior caudate was observed only during the maintenance phase. During the response phase, the basal ganglia were equally active in both the high-load and low-load WM conditions. Caudate and putamen activations were primarily localized to the (rostral) associative parts of the basal ganglia, consistent with the putative role of these regions in cognitive processing. Effective connectivity analyses revealed increased WM-load-dependent interaction of the left anterior caudate with the left posterior parietal cortex during all three phases of the task; with the visual association cortex, including the fusiform gyrus and inferior temporal gyrus, only during the encoding phase; with the ventrolateral prefrontal cortex during the encoding and maintenance phases; with the pre-supplementary motor area during the maintenance and response phases; and with the dorsolateral prefrontal and anterior cingulate cortices only during the response phase. Taken together with known neuroanatomy of the basal ganglia, these results suggest that the anterior caudate helps to link signals in distinct functional networks during different phases of the WM task. Our study offers new insight into the integrative and adaptive role of the basal ganglia in higher cognitive function.     

Hippocampal activations during encoding and retrieval in a verbal working memory paradigm

Though the hippocampus has been associated with encoding and retrieval processes in episodic memory, the precise nature of its involvement in working memory has yet to be determined. This functional magnetic resonance imaging (fMRI) study employed a verbal working memory paradigm that allows for the within-subject comparison of functional activations during encoding, maintenance, and retrieval. In each trial, participants were shown 5 target words and, after an 8 s delay, a series of probe words. Probe words consisted of target matches, phonetically or semantically related foils, or foils unrelated to the target words. Both the left and right hippocampi showed higher mean activation amplitudes during encoding than maintenance. In contrast, the right dorsolateral prefrontal cortex (DLPFC) showed greater activation during maintenance than encoding. Both hippocampal and DLPFC regions were more active during retrieval than maintenance. Furthermore, an analysis of retrieval activation separated by probe type showed a trend toward greater bilateral hippocampal activation for probes related (both semantically and phonetically) to the target than for unrelated probes and still greater activation for target matches. This pattern suggests that there may be roles for the hippocampus and DLPFC in working memory that change as function of information processing stage. Additionally, the trend towards increased involvement of the hippocampus with the increase in relatedness of the probe stimuli to the information maintained is interpreted to be consistent with the role of the hippocampus in recollection-based retrieval in long-term memory and may indicate that this role extends to working memory processes.     

Differences between auditory evoked responses recorded during spatial and nonspatial working memory tasks

Results from several recent studies suggest that neuronal processing of sound content and its spatial location may be dissociated. The use of modern neuroimaging techniques has allowed for the determination that different brain structures may be specifically activated during working memory processing of pitch and location of sound. The time course of these task-related differences, however, remains uncertain. In the present study, we performed simultaneous whole-head electroencephalogram and magnetoencephalogram recordings, using a new behavioral paradigm, to investigate the dynamics of differences between "what" and "where" evoked responses in the auditory system as a function of memory load. In the location task the latency of the N1m was shorter and its generator was situated more inferiorly than in the pitch task. Working memory processing of the tonal frequency enhanced the amplitude of the N2 component, as well as the negative-going deflection at a latency around 400 ms. A memory-load-dependent task-related difference was found in the positive slow wave which was higher during the location than pitch task at the low load. Late slow waves were affected by memory load but not type of task. These results suggest that separate neuronal networks are involved in the attribute-specific analysis of auditory stimuli and their encoding into working memory, whereas the maintenance of auditory information is accomplished by a common, nonspecific neuronal network.     

Increased frontocerebellar activation in alcoholics during verbal working memory: an fMRI study

Although there is clear evidence of alcoholism-related damage to the frontal lobes and cerebellum from neuroimaging, neuropathological, and neuropsychological studies, the functional role of the cerebellum and cerebrocerebellar circuits related to verbal working memory deficits of alcoholics have not been well studied. Alcoholic and nonalcoholic subjects performed a Sternberg verbal working memory task while receiving an fMRI scan in a 3T magnet. This task has been found in previous studies to reliably activate the articulatory control and phonological storage components of the phonological loop (left frontal, left temporal/parietal structures, right superior cerebellar regions) in young healthy controls. We hypothesized that the alcoholics would show a different pattern of activation from the controls, based on the regions of interest (ROIs) identified from a previous study of healthy subjects. Behavioral results showed the alcoholics to be performing at a comparable level to the matched controls in terms of accuracy and median reaction time, with no statistically significant differences. However, analysis of the functional data revealed that the alcoholics exhibited greater activation in the left frontal (BA44/45) and right superior cerebellum (HVI) regions relative to the matched controls. These findings suggest that brain activation in left frontal and right cerebellar regions that support the articulatory control system of verbal working memory may require a compensatory increase in alcoholics in order to maintain the same level of performance as controls.     

Changes in brain network activity during working memory tasks: a magnetoencephalography study

In this study, we elucidate the changes in neural oscillatory processes that are induced by simple working memory tasks. A group of eight subjects took part in modified versions of the N-back and Sternberg working memory paradigms. Magnetoencephalography (MEG) data were recorded, and subsequently processed using beamformer based source imaging methodology. Our study shows statistically significant increases in θ oscillations during both N-back and Sternberg tasks. These oscillations were shown to originate in the medial frontal cortex, and further to scale with memory load. We have also shown that increases in θ oscillations are accompanied by decreases in β and γ band oscillations at the same spatial coordinate. These decreases were most prominent in the 20-40 Hz frequency range, although spectral analysis showed that γ band power decrease extends up to at least 80 Hz. β/γ Power decrease also scales with memory load. Whilst θ increases were predominately observed in the medial frontal cortex, β/γ decreases were associated with other brain areas, including nodes of the default mode network (for the N-back task) and areas associated with language processing (for the Sternberg task). These observations are in agreement with intracranial EEG and fMRI studies. Finally, we have shown an intimate relationship between changes in β/γ band oscillatory power at spatially separate network nodes, implying that activity in these nodes is not reflective of uni-modal task driven changes in spatially separate brain regions, but rather represents correlated network activity. The utility of MEG as a non-invasive means to measure neural oscillatory modulation has been demonstrated and future studies employing this technology have the potential to gain a better understanding of neural oscillatory processes, their relationship to functional and effective connectivity, and their correspondence to BOLD fMRI.     

Graph properties of synchronized cortical networks during visual working memory maintenance

Oscillatory synchronization facilitates communication in neuronal networks and is intimately associated with human cognition. Neuronal activity in the human brain can be non-invasively imaged with magneto- (MEG) and electroencephalography (EEG), but the large-scale structure of synchronized cortical networks supporting cognitive processing has remained uncharacterized. We combined simultaneous MEG and EEG (MEEG) recordings with minimum-norm-estimate-based inverse modeling to investigate the structure of oscillatory phase synchronized networks that were active during visual working memory (VWM) maintenance. Inter-areal phase-synchrony was quantified as a function of time and frequency by single-trial phase-difference estimates of cortical patches covering the entire cortical surfaces. The resulting networks were characterized with a number of network metrics that were then compared between delta/theta- (3–6 Hz), alpha- (7–13 Hz), beta- (16–25 Hz), and gamma- (30–80 Hz) frequency bands. We found several salient differences between frequency bands. Alpha- and beta-band networks were more clustered and small-world like but had smaller global efficiency than the networks in the delta/theta and gamma bands. Alpha- and beta-band networks also had truncated-power-law degree distributions and high k-core numbers. The data converge on showing that during the VWM-retention period, human cortical alpha- and beta-band networks have a memory-load dependent, scale-free small-world structure with densely connected core-like structures. These data further show that synchronized dynamic networks underlying a specific cognitive state can exhibit distinct frequency-dependent network structures that could support distinct functional roles.     

Spatiotemporal characteristics of hemodynamic changes in the human lateral prefrontal cortex during working memory tasks

The prefrontal cortex (PFC) is widely believed to subserve mental manipulation and monitoring processes ascribed to the central executive (CE) of working memory (WM). We attempted to examine and localize the CE by functional imaging of the frontal cortex during tasks designed to require the CE. Using near-infrared spectroscopy, we studied the spatiotemporal dynamics of oxygenated hemoglobin (oxy-Hb), an indicator of changes in regional cerebral blood flow, in both sides of lateral PFC during WM intensive tasks. In most participants, increases in oxy-Hb were localized within one subdivison during performance of the n-back task, whereas oxy-Hb increased more diffusely during the random number generation (RNG) task. Activation of the ventrolateral PFC (VLPFC) was prominent in the n-back task; both sustained and transient dynamics were observed. Transient dynamics means that oxy-Hb first increases but then decreases to less than 50% of the peak value or below the baseline level before the end of the task. For the RNG task sustained activity was also observed in the dorsolateral PFC (DLPFC), especially in the right hemisphere. However, details of patterns of activation varied across participants: subdivisions commonly activated during performance of the two tasks were the bilateral VLPFCs, either side of the VLPFC, and either side of the DLPFC in 4, 2, and 4 of the 12 participants, respectively. The remaining 2 of the 12 participants had no regions commonly activated by these tasks. These results suggest that although the PFC is implicated in the CE, there is no stereotyped anatomical PFC substrate for the CE.     

Interaction between a verbal working memory network and the medial temporal lobe

The irrelevant speech effect illustrates that sounds that are irrelevant to a visually presented short-term memory task still interfere with neuronal function. In the present study we explore the functional and effective connectivity of such interference. The functional connectivity analysis suggested an interaction between the level of irrelevant speech and the correlation between in particular the left superior temporal region, associated with verbal working memory, and the left medial temporal lobe. Based on this psycho-physiological interaction, and to broaden the understanding of this result, we performed a network analysis, using a simple network model for verbal working memory, to analyze its interaction with the medial temporal lobe memory system. The results showed dissociations in terms of network interactions between frontal as well as parietal and temporal areas in relation to the medial temporal lobe. The results of the present study suggest that a transition from phonological loop processing towards an engagement of episodic processing might take place during the processing of interfering irrelevant sounds. We speculate that, in response to the irrelevant sounds, this reflects a dynamic shift in processing as suggested by a closer interaction between a verbal working memory system and the medial temporal lobe memory system.     

Functional MRI of auditory verbal working memory: long-term reproducibility analysis

Although functional MRI (fMRI) has shown to be a tool with great potential to study the normal and diseased human brain, the large variability in the detected hemodynamic responses across sessions and across subjects hinders a wider application. To investigate the long-term reproducibility of fMRI activation of verbal working memory (WM), eight normal subjects performed an auditory version of the 2-back verbal WM task while fMRI images were acquired. The experiment was repeated nine times with the same settings for image acquisition and fMRI task. Data were analyzed using SPM99 program. Single-session activation maps and multi-subject session-specific activation maps were generated. Regions of interest (ROIs) associated to specific components of verbal WM were defined based on the voxels' coordinates in Talairach space. Visual observation of the multi-subject activation maps showed similar activation patterns, and quantitative analysis showed small coefficients of variance of activation within ROIs over time, suggesting small longitudinal variability of activation. Visual observation of the activation maps of individual sessions demonstrated striking variation of activation across sessions and across subjects, and quantitative analysis demonstrated larger contribution from between-subject variation to overall variation than that from within-subject variation. We concluded that by multi-subject analysis of data from a relatively small number of subjects, reasonably reproducible activation for the 2-back verbal WM paradigm can be achieved. The level of reproducibility encourages the application of this fMRI paradigm to the evaluation of cognitive changes in future investigations. The quantitative estimation of the proportions of within-subject and between-subject variabilities in the overall variability may be helpful for the design of future studies.     

Functional dissociations within the inferior parietal cortex in verbal working memory

Neuroimaging studies of working memory have revealed two sites in the left supramarginal gyrus that may support the short-term storage of phonological information. Activation in the left dorsal aspect of the inferior parietal cortex (DIPC) has been observed in contrasts of working memory load, whereas activation in the ventral aspect of the inferior parietal cortex (VIPC) has been found primarily in contrast of information type (verbal vs. nonverbal). Our goal was to determine whether these two areas are functionally distinct or if instead they are part of a homogeneous region with large variations in the focus of peak activity. Toward this end, we used fMRI to assess the neural response in two working memory tasks (N-back and item recognition) in which we also manipulated memory load and the type of information to be recalled (verbal vs. nonverbal). We found both DIPC and VIPC activation in the same group of subjects and further demonstrated that they have differential sensitivity to our experimental factors. Only the DIPC showed robust load effects, whereas only the VIPC showed reliable effects of information type. These results help to account for the differences observed in between-subject comparisons, and they indicate that the two regions are functionally dissociable. In contrast to the DIPC, activity of the VIPC was also recruited in the fixation and low-load conditions, a surprising result that has not been fully explored in prior studies. Despite their distinctive patterns of performance, neither of these regions displayed a pattern of activity that entirely corresponds to common assumptions of a dedicated phonological short-term store (STS). Instead, we hypothesize that the DIPC may support domain-general executive processes, while the VIPC may support phonological encoding-recoding processes central to a variety of language tasks.     

Neurodevelopmental changes in verbal working memory load-dependency: an fMRI investigation

Development of working memory (WM) aptitude parallels structural changes in the frontal–parietal association cortices important for performance within this cognitive domain. The cerebellum has been proposed to function in support of the postulated phonological loop component of verbal WM, and along with frontal and parietal cortices, has been shown to exhibit linear WM load-dependent activation in adults. It is not known if these kinds of WM load-dependent relationships exist for cerebro-cerebellar networks in developmental populations, and whether there are age-related changes in the nature of load-dependency between childhood, adolescence, and adulthood. The present study used fMRI and a verbal Sternberg WM task with three load levels to investigate developmental changes in WM load-dependent cerebro-cerebellar activation in a sample of 30 children, adolescents, and young adults between the ages of 7 and 28. The neural substrates of linear load-dependency were found to change with age. Among adolescents and adults, frontal, parietal and cerebellar regions showed linear load-dependency, or increasing activation under conditions of increasing WM load. In contrast, children recruited only left ventral prefrontal cortex in response to increasing WM load. These results demonstrate that, while children, adolescents, and young adults activate similar cerebro-cerebellar verbal working memory networks, the extent to which they rely on parietal and cerebellar regions in response to increasing task difficulty changes significantly between childhood and adolescence.     

Behavioral and functional MRI study of attention shift in human verbal working memory

The tripartite model of memory proposed the requirement of attentional switching when accessing different items in working memory [J. Exp. Psychol. Learn. Mem. Cogn. 27 (2001) 817]. This internal focus of attention is limited to just one item and the switching process is time-consuming [Mem. Cogn. 26 (1998) 263].In the current study, given a three-digit list stored in working memory, we found that it took longer to shift attention in the direction of "Upstream" than "Downstream", and that each shift was a "single step" process. To investigate the neural basis of this type of attention switching, we performed a functional MRI study. The results revealed that at least three important brain areas are involved, including the left dorsal lateral prefrontal cortex, the cingulate gyrus, and the medial occipital cortex. These areas all showed greater activation in the attention shift condition compared to control conditions of no (or decreased) attention shift requirements. In addition, the hemodynamic activities in these areas are highly correlated, suggesting a strong functional connectivity between them. Taken together with evidence from several recent investigations, our results suggest that these areas each play an important and specific role in collaboratively supporting the function of attention shift in working memory.     

Mapping of verbal working memory in nonfluent Chinese-English bilinguals with functional MRI

Existing cognitive and neural imaging studies have suggested a frontoparietal network of multiple, cooperative components for verbal working memory (WM). We used functional MRI to investigate whether this neural network is also involved in the processing of second language by nonfluent bilinguals. Twelve (five males, seven females) native Chinese speakers who had limited English proficiency were scanned while performing working memory tasks in Chinese and English. They were asked to make judgment continuously whether the word presented on the screen was semantically related to (i.e., the semantic tasks) another word presented two words earlier. On a different task (i.e., the phonological tasks), they were asked to make judgment whether the target word rhymed with the other word. A naming and judgment task in each language was adopted to control for the visual process, initial lexical process, and motor responses. Behavioral data showed that subjects performed better at tasks in their native language (Chinese, L1) than in English (L2). Imaging results showed that all working memory tasks in both L1 and L2 elicited a very similar pattern of left-hemisphere-dominated activation in the dorsolateral prefrontal cortex, pars opercularis region, pars triangularis region, precentral cortex, and parietal lobule. Consistent with the behavioral data, the volume of activation in the left opercularis region, left parietal lobule, and right precentral region was greater for L2 than for L1. These results suggest that working memory in L1 and L2 is mediated by a unitary neural system (i.e., frontoparietal region), which is capable of recruiting surrounding cortical resources to meet the increased computational demand caused by low L2 proficiency.     

Maintenance versus manipulation in verbal working memory revisited: an fMRI study

Working memory (WM) is the ability to keep a limited amount of information "on line" for immediate use during short intervals. Verbal WM has been hypothesized to consist of neuroanatomically segregated components, i.e., maintenance (storage, rehearsal, and matching) and manipulation (reordering or updating), corresponding to ventrolateral and dorsolateral prefrontal cortex. Previous imaging studies of maintenance vs manipulation processes in WM have produced inconsistent results, which may have been due to methodological issues such as low statistical power and the use of insertion (subtraction) designs. In the present functional magnetic resonance imaging study we used parametric versions of both a prototypical maintenance task (Sternberg) and a prototypical manipulation task (n-letter back task) in 21 healthy subjects. Increased signal correlated with load common for both tasks was found in bilateral dorsolateral and anterior prefrontal, left ventrolateral prefrontal, and bilateral parietal regions. Workload x task interactions were found in bilateral dorsolateral prefrontal cortex for manipulation vs maintenance, but also for responding vs encoding (storage) in the maintenance task. Therefore, our data support a functional rather than a neuroanatomical distinction between maintenance and manipulation, given our finding that these tasks differentially activate virtually identical systems.     

Left TPJ activity in verbal working memory: implications for storage- and sensory-specific models of short term memory

Patients with damage to the left temporoparietal junction (TPJ) have a low verbal span without concomitant deficits in speech perception. This pattern of cognitive impairment is taken as evidence for a dedicated phonological buffer that plays little role in perception (storage-specific account). In contrast, other research suggests that items are maintained and perceived in the same regions (sensory-specific account). In an fMRI study, we demonstrate that the left TPJ does not respond in a way predicted of a phonological buffer; that is, activity in this region is not sustained during encoding or maintenance. Instead, a region in the superior temporal gyrus that has been associated with both speech perception and production demonstrated the expected profile of a store: it was more active in the verbal condition than the object condition and was active during both encoding and maintenance. These results support the sensory-specific account of short term memory rather than the storage-specific account. Based on the pattern of activity in the left TPJ, we suggest that the impairment of verbal working memory observed in patients with TPJ damage may be due to diminished attentional processes rather than reduced storage capacity.     

Frontal midline EEG dynamics during working memory

We show that during visual working memory, the electroencephalographic (EEG) process producing 5-7 Hz frontal midline theta (fmtheta) activity exhibits multiple spectral modes involving at least three frequency bands and a wide range of amplitudes. The process accounting for the fmtheta increase during working memory was separated from 71-channel data by clustering on time/frequency transforms of components returned by independent component analysis (ICA). Dipole models of fmtheta component scalp maps were consistent with their generation in or near dorsal anterior cingulate cortex. From trial to trial, theta power of fmtheta components varied widely but correlated moderately with theta power in other frontal and left temporal processes. The weak mean increase in frontal midline theta power with increasing memory load, produced entirely by the fmtheta components, largely reflected progressively stronger theta activity in a relatively small proportion of trials. During presentations of letter series to be memorized or ignored, fmtheta components also exhibited 12-15 Hz low-beta activity that was stronger during memorized than during ignored letter trials, independent of letter duration. The same components produced a brief 3-Hz burst 500 ms after onset of the Probe letter following each letter sequence. A new decomposition method, log spectral ICA, applied to normalized log time/frequency transforms of fmtheta component Memorize-letter trials, showed that their low-beta activity reflected harmonic energy in continuous, sharp-peaked theta wave trains as well as independent low-beta bursts. Possibly, the observed fmtheta process variability may index dynamic adjustments in medial frontal cortex to trial-specific behavioral context and task demands.     

Coherence between fMRI time-series distinguishes two spatial working memory networks

Widespread and distributed brain regions are thought to form networks that together support working memory. We recently demonstrated that different cortical areas maintain relatively different codes across a memory delay (Curtis et. al., J Neurosci, 2004; 24:3944-3952). The frontal eye fields (FEF), for example, were more active during the delay when the direction of the memory-guided saccade was known compared to when it was not known throughout the delay. Other areas showed the opposite pattern. Despite these task-dependent differences in regional activity, we could only assume but not address the functional interactions between the identified nodes of the putative network. Here, we use a bivariate technique, coherence, to formally characterize functional interactions between a seed region and other brain areas. We find that the type of representational codes that are being maintained in working memory biases frontal-parietal interactions. For example, coherence between FEF and other oculomotor areas was greater when a motor representation was an efficient strategy to bridge the delay period. However, coherence between the FEF and higher-order heteromodal areas, e.g., dorsolateral prefrontal cortex, was greater when a sensory representation must be maintained in working memory.     

Load- and practice-dependent increases in cerebro-cerebellar activation in verbal working memory: an fMRI study

Load-dependent and practice-related changes in neocortical and cerebellar structures involved in verbal working memory (VWM) were investigated using functional MRI (fMRI) and a two alternative forced choice Sternberg paradigm. Using working memory loads ranging from 2 to 6 letters, regions exhibiting linear and quadratic trends in load-dependent activations were identified. Behaviorally, reaction time measurements revealed significant linear increases with increasing memory load, and significant decreases with increased task practice. Brain activations indicated a preponderance of linear load-dependent responses in both superior (lobule VI/Crus I) and inferior (lobule VIIB/VIIIA) cerebellar hemispheres, as well as in areas of neocortex including left precentral (BA 6), inferior frontal (BA 47), parahippocampal (BA 35), inferior parietal (BA 40), cingulate (BA 32), and right inferior and middle frontal (BA 46/47) regions. Fewer voxels exhibited quadratic without linear trends with the most prominent of these activations located in left inferior parietal (BA 40), precuneus, and parahippocampal regions. Analysis of load x session interactions revealed that right inferior cerebellar and left inferior parietal activations increased with practice, as did the correlations between activation in each region with reaction time, suggesting that changes in this cerebro-cerebellar network underlie practice-related increases in efficiency of VWM performance. These results replicate and extend our previous findings of fMRI activation in the cerebellum during VWM, and demonstrate predominately linear increases in cerebro-cerebellar activation with increasing memory load as well as changes in network function with increased task proficiency.