Cerebral Cortex Advance Access published online on May 22, 2008
Cerebral Cortex, doi:10.1093/cercor/bhn084
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Error-Related Negativity is Mediated by Fractional Anisotropy in the Posterior Cingulate Gyrus—A Study Combining Diffusion Tensor Imaging and Electrophysiology in Healthy Adults
1 Center for the Study of Human Cognition, Department of Psychology, University of Oslo, Oslo 0317, Norway, 2 Department of Neuropsychology, Ullevaal University Hospital, Oslo 0317, Norway, 3 Department of Medical Physics and the Interventional Centre, Rikshospitalet University Hospital, Oslo 0317, Norway, 4 Department of Physics, University of Oslo, Oslo 0317, Norway, 5 Department of Radiology, Rikshospitalet University Hospital, Oslo 0317, Norway
Address correspondence to Lars T. Westlye, Cand Psychol, Center for the Study of Human Cognition, Department of Psychology, University of Oslo, P.O. Box 1094 Blindern, 0317 Oslo, Norway. Email: l.t.westlye{at}psykologi.uio.no.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
White matter (WM) is critical to cognitive function and brain activity. The objective of the present study was to test whether diffusion tensor imaging (DTI) derived WM measures are related to the cognitive event-related potential error-related negativity (ERN). Eighty-seven healthy middle-aged adults underwent DTI scanning and electrophysiological recordings while doing a version of the Eriksen flanker task. ERN was elicited in error trials. Fractional anisotropy (FA) was calculated based on the DTI scans. FA indexes degree of anisotropic diffusion in every voxel, and is assumed related to the integrity of myelinated fiber bundles. The principal neuronal generator for ERN is located in the anterior cingulate cortex (ACC). Hence, the relationship between FA in the cingulum bundle and ERN amplitude was tested. It was found that FA in the left posterior cingulate correlated with ERN. Eigenvalue analyses revealed that radial diffusivity was responsible for the FA effect. ERN amplitude predicted response accuracy in the Flanker task, suggesting that electrophysiological measures are intermediate explanatory variables connecting DTI indices of WM organization, synchronization of large cell assemblies, and behavior.
Key Words: event-related potentials • error processing • MRI • tract-based spatial statistics • white matter integrity
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Diffusion tensor imaging (DTI) provides an in vivo quantitative measure of white matter (WM) characteristics by utilizing the restricted water diffusion in biological brain tissues. Because even simple cognitive tasks involve a complex interplay of multiple brain areas, it is likely that the organization of fiber connections between different areas is related to cognitive function. DTI measures may possibly be used to study this relation. Fractional anisotropy (FA) is an intravoxel index of the degree of diffusion anisotropy. Several neurobiological features contribute to FA, among others the degree of myelination (Le Bihan 2003
; Mori and Zhang 2006; Wozniak and Lim 2006). It should be noted, though, that sources other than myelin are sufficient for creating anisotropy in the brain (Beaulieu 2002). For instance, less than 20% reduction in anisotropy has been found in myelin-deficient rats (Gulani et al. 2001). Still, degree of myelination remains one important factor in FA, and this is a reason why DTI and FA have received much recent research interest.
FA has been shown to correlate negatively with age (Moseley 2002; Head et al. 2004; Salat, Tuch, Greve, et al. 2005; Salat, Tuch, Hevelone, et al. 2005; Hugenschmidt et al. 2008), and various neurological diseases (Fellgiebel et al. 2004, 2006; Smith et al. 2006). Also, a recent post mortem study indicates that WM volume is more sensitive to healthy aging than gray matter (GM) volume (Piguet et al. forthcoming), suggesting that WM properties may be a better predictor of age-related cognitive decrements than cortical GM. Several recent studies have found correlations between DTI measures and cognitive performance (Madden et al. 2004; O'Sullivan et al. 2005; Tuch et al. 2005; Charlton et al. 2006, forthcoming; Persson et al. 2006; Grieve et al. 2007) and the blood oxygenation level–dependent (BOLD) response (Olesen et al. 2003; Seghier et al. 2004; Schlösser et al. 2007). However, no studies to date have addressed whether a relationship between more direct measures of synchronized neuronal activity and DTI exists. This pertains to the neurocognitive significance of DTI measures, and the aim of the present article is to shed light on the relationship between neuronal brain activity and DTI indices of regional WM. This is done by relating the amplitude of the electrophysiological event-related potentials (ERP) component error-related negativity (ERN) to the FA index in the WM of the cingulate gyrus.
Even though it has not been tested, there is reason to expect a relationship between FA and the amplitude of specific ERP components in healthy adults. Scalp recorded electrophysiological activity is assumed to reflect synchronous action potentials in large cell assemblies (Varela et al. 2001), and related to degree of axonal myelinization (Bartzokis 2003). Increased myelinization would result in faster nerve conduction velocity, facilitating swift synchronization of neuronal activity in response to both internal and external stimuli. Neuronal synchronization is an important factor contributing to scalp recorded electrophysiological potentials (Varela et al. 2001). Assuming that FA to some extent indexes variability in degree of axonal myelinization, this would suggest a correlation between ERP amplitude and FA. Further, a correlation between ERP latency and regional WM volume has been found, and taken to indicate that WM connectivity between neural generators is as important for the ERPs as the generators themselves (Cardenas et al. 2005).
A direct test of the relationship between WM fiber organization and synchronous neuronal activity would be to correlate ERP amplitude with FA in targeted regions. In the present study, ERN was chosen as the ERP component of interest. ERN is a response locked negative ERP often evoked by commission errors in speeded response tasks (Gehring et al. 1993), typically peaking about 40–70 ms after the erroneous response (Taylor et al. 2007). There is a general consensus that ERN reflects cognitive control mechanisms involved in monitoring processes (Bush et al. 2000; Falkenstein et al. 2000) and in the detection and processing of conflict and errors (Yeung et al. 2004). ERN is also thought to reflect behavioral reinforcement processes initially executed in striatal areas (Holroyd and Coles 2002; Holroyd et al. 2003, 2006; Nieuwenhuis et al. 2004). Several studies have pointed to cingulate involvement in error processing and in the generation of ERN (van Veen and Carter 2002, 2006; Mathalon et al. 2003; Herrmann et al. 2004; Debener et al. 2005). Using source localization, Herrmann et al. (2004) located the neuronal source of the ERN to the anterior cingulate cortex (ACC). Applying concurrent functional magnetic resonance imaging (fMRI) and ERP, Debener et al. (2005) reported converging evidence of ACC involvement in error processing in subjects performing a version of Eriksen flanker task (Eriksen and Eriksen 1974).
In sum, ERN was chosen in the present study because it is tightly connected to complex cognitive processing, it is easily identifiable in most participants, and the anatomical localization of its principal neuronal generator is known. In addition to the neuronal generator in the cingulate gyrus, evidence from lesion studies point to distributed cortical and subcortical involvement in ERN generation and indicate that WM integrity contributes to ERN (Gehring and Knight 2000; Stemmer et al. 2004; Hogan et al. 2006; Ullsperger and von Cramon 2006). Based on assumptions about the role of WM myelin for FA as well as for swift and synchronous neuronal transmission, a positive correlation between ERN and FA was hypothesized. To test for regionally specific effects, voxel by voxel analyses were performed in a priori selected regions of interest (ROIs) in the cingulum bundles. This was done because the principal generator for ERN is situated in the cingulate, alleviating the multiple comparison problem.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Subjects
The sample was drawn from a large longitudinal research project in Oslo, called Cognition and Plasticity through the Life-Span. The study has been approved by the Regional Ethical Committee of South Norway (REK-Sør), and written informed consent was obtained from all participants prior to the examinations. Volunteers were recruited by newspaper advertisements. Participants were required to be right handed native Norwegian speakers in the age range 40 and 60 years, feel healthy, not use medicines known to affect central nervous system (CNS) functioning, including psychoactive drugs, not to be under psychiatric treatment, be free from worries for own memory abilities and injury or diseases known to affect CNS function, including neurological or psychiatric illness, serious head injury, or history of stroke. One hundred and 5 participants satisfied these criteria. Four participants were excluded due to lacking MRI data. All MR scans were subjected to a radiological evaluation by a specialist in neuroradiology, and were required to be deemed free of significant injuries or conditions. This led to the exclusion of 1 additional participant, reducing the n to 100 (58 F/42 M). All scored 
16 on Beck Depression Inventory (BDI; Beck 1987) and 
27 on Mini Mental State Examination (MMSE) (Folstein et al. 1975).
Of the remaining 100 participants, 13 were excluded due to behavioral criteria defined in the behavioral task (see below). The remaining 87 study participants included 50 female (mean age 51.5, SD = 5.0) and 37 male (mean age 51.8, SD = 4.7) healthy adults, with mean MMSE score of 29.3 (range 27–30) and BDI score of 4.1 (range 0–16). To further evaluate the participants’ cognitive functioning, we assessed general intellectual ability by Wechsler Abbreviated Scale of Intelligence (Wechsler 1999). Mean IQ was 114.0 (96–128). Demographic, screening and intellectual variables for the 87 included participants are reported in Table 1.
|
Eriksen Flanker Task
We administered a modified version of the Eriksen flanker task (Eriksen and Eriksen 1974), similar to the task used by Debener et al. (2005). A sketch of the task is presented in Figure 1. The stimuli were horizontal arrows of length 1 cm (approx 1°), pointing either to the right or the left displayed in a vertical stack 2.5° high. Subjects were to respond as accurately and quickly as possible by button presses indicating which direction the middle arrow was pointing. Each trial consisted of the following stimuli; first, a fixation cross was presented for a random interval ranging between 1200 and 1800 ms. Then the 4 flanker arrows were presented for 80 ms before the target arrow was presented for 30 ms along with the flanker arrows. The flanker arrows were presented prior to the target to increase prepotent responding and make the task more difficult. At presentation of the target, the task was to push 1 button if the target was pointing to the left and another button if the target was pointing to the right. Based on the mean reaction time (RT) for the 20 first consecutive trials, an individually adjusted RT criterion was set. After every subsequent third trial with RT exceeding this criterion, a message occurred on the screen for 1 second telling the participant to respond faster. This was also implemented to increase the demand to respond swiftly and thus increase task difficulty.
|
Responses were obtained on a PST Serial Response Box, and the experimental procedures and responses were collected using E-prime (Psychological Software Tools, Pittsburgh, PA) software. There were 2 experimental task conditions; congruent and incongruent. In the congruent condition all arrows were pointing in the same direction. In the incongruent condition, the middle arrow was pointing toward the opposite side as the flanker arrows. The task included 416 trials with a short break half way. The probability of an incongruent trial was 50% in a randomized fashion, with a rule of no more than 3 consecutive incongruent trials.
This and similar tasks have been shown to produce a reliable number of commission errors in the incongruent condition (Debener et al. 2005), and also to produce the well described congruence effect on RT with higher RT in the incongruent compared with the congruent condition (Eriksen and Eriksen 1974). The congruence effect is probably reflecting response conflict induced in the incongruent condition (Botvinick et al. 2001). In this study, we were mainly interested in the immediate electrophysiological potential after erroneous responses compared with that of correct responses. To exclude any condition/conflict related interactions with the measured potentials (van Veen and Carter 2002), only data from the incongruent condition will be reported here.
Three behavioral exclusion criteria were employed in the Flanker task. First, no participants with less than 80% accuracy in congruent trials were included. Second, participants with a non significant congruence effect on RT in correct trials were excluded [paired samples t-tests, incongruent RT > congruent RT, P > 0.05]. These criteria were used to exclude participants with suboptimal motivation. The first criterion led to the exclusion of 7 participants, and 2 more were excluded on the basis of the second criterion. Six of the 7 subjects excluded due to the accuracy criterion would also have been excluded due to the RT criterion alone. The third criterion employed was number of accepted incongruent commission error ERP trials, and participants with less than 10 trials were excluded. This criterion led to 4 more excluded subjects. However, subjects with less than 15 error trials were only accepted upon manual inspection of the average curves. In sum, 13 participants were excluded on the basis of behavioral criteria in the flanker task. All data reported are based on the remaining 87 participants.
Electrophysiological Recordings
The electrophysiological recordings were done with a 128 Channel EasyCap Montage No. 15 (http://www.easycap.de/easycap/) with a sampling rate of 1000 Hz. During recording, participants were seated in a shielded Faraday chamber in a comfortable chair at approximately 60 cm distance from a 19-inch computer monitor. The signals were amplified with Neuroscan SynAmps2 and filtered online with a 30-Hz low-pass and 0.15-Hz high-pass analog filter prior to digitalizing and saving of the continuous data set. All electrodes were referenced to a common electrode placed on the left mastoid. Vertical eye blinks were recorded with bipolar electrodes above and below the left canthi. Impedances were kept below 10 kOhm.
Electrophysiological Processing
The continuous data set was segmented into 1000-ms response locked epochs starting 600 ms prior to response and lasting until 400 ms after response. The entire epochs were linearly detrended and baseline corrected relative to a 100-ms time window –600 to –500 ms prior to response. This time window was chosen in order to avoid stimulus related activity potentially contaminating the response locked epochs (Debener et al. 2005). Due to systematic biases toward higher RTs in correct compared with erroneous responses, it is likely that stimulus evoked potentials (e.g., P300) would differently influence the correct and erroneous responses, thus inviting biased inferences of the response locked activity if a time window more proximate to response was chosen as baseline. For the same reason, all trials with RT > 600 ms were excluded from further analyses.
To minimize noise, epochs containing signals ±100 µV were excluded. The epochs were further corrected for eye blinks (Semlitsch et al. 1986) and digitally filtered with a 30-Hz low-pass filter. Remaining epochs containing response locked correct and erroneous responses from incongruent trials were extracted for further analysis. Mean number of accepted error trials used in the analyses were 41.6 (SD = 26.9, range: 10–135). Correct and error trials were averaged independently.
Quantification of the ERN
Figure 2 shows the grand average ERN for all 87 participants. Figure 3 shows the topographic distribution of the electrophysiological potential after correct and erroneous responses, clearly indicating a medial frontocentral maximum of the ERN, probably reflecting cingulate neuronal activity in response to commission errors (Taylor et al. 2007). Visual inspection of the mean grand average curves and topographical voltage plots for error trials for all participants revealed a maximum peak on channel FCz. This corresponds well to previous literature (Taylor et al. 2007). Peak amplitude was defined as the most negative point in the time window 0- to 130-ms post response at FCz (mean latency = 63.14 ms, SD = 22.06). A principal component analysis (PCA) with amplitude measured at this time point at 22 selected electrodes (AFz, Fz, FCz, Cz, CPz, Pz, POz, Oz, FP1, F3, F7, F9, FT9, T7, O1, FP2, F4, F8, F10, FT8, T8, O2) yielded 4 components with an eigenvalue > 1. The eigenvalues obtained from PCA express the amount of variance in the original variables accounted for by each component. The original variables do each have a variance of 1, because they are standardized for PCA. Thus, any component with an eigenvalue of at least 1 explains more of the variance than any original variable. One component comprising medial frontal channels explained 51.7% of the variance in the measured activity, representing a valid candidate for ERN. The electrodes with the highest corrected item total correlation with this component were FCz (r = 0.96), Fz (r = 0.90), and Cz (r = 0.92). To minimize single electrode noise and possible signal distortions, the mean of these 3 channels were chosen to represent the ERN in further analyses. Peak amplitude may be influenced by fast oscillating waves, and may therefore not be an accurate estimate of signal intensity (Luck 2005). Thus, individually adjusted average amplitude signal intensity was estimated by averaging the measured potential in the 20 ms preceding and 20 ms after peak amplitude at FCz for each participant. Mean amplitude for Fz, FCz, and Cz in the individually adjusted time window was –5.77 µV (SD = 5.35). The measured electrophysiological activity after correct responses did not yield any well defined negative peak. Thus, we did not compute the mean signal after correct responses. The amplitude was converted to Z scores based on the mean of all participants, inverted, and submitted to statistical analysis. Note that a high Z score denotes more negative ERN amplitude, that is, a strong ERN deflection.
|
|
DTI Acquisition and Processing
Imaging data were collected on a 1.5-Tesla Siemens Avanto scanner (Siemens Medical Solutions, Erlangen, Germany) with a 12 channel head coil. The pulse sequence used for diffusion weighted imaging was a single-shot echo planar imaging (EPI) with a twice-refocused spin echo pulse with 30 directions and the following parameters: repetition time/echo time = 8200 ms/82 ms, b-value = 700 s/mm2, voxel size = 2.0 x 2.0 x 2.0 mm. This sequence is optimized to minimize eddy current-induced image distortions (Reese et al. 2003). The sequence was repeated in 2 successive runs with 10 non-diffusion-weighted (b = 0) images in addition to 30 diffusion weighted images collected per acquisition. Each volume consisted of 64 axial slices. Total scanning time was 11 min 21 s.
The volumes were anonymized and transferred offline to a Linux workstation for processing. First, the volumes were eddy current and motion corrected relative to the first b = 0 volume in the series using FMRIB's Diffusion Toolbox (FDT), part of FMRIB Software Library (FSL) (http://www.fmrib.ox.ac.uk/fsl/) (Smith et al. 2004). Then, the 2 acquisitions were averaged before extracting nonbrain tissue (skull, cerebrospinal fluid, etc.) using Brain Extraction Tool (Smith 2002). FDT was used to fit a diffusion tensor model to the data at each voxel. Voxel-wise values of FA as well as diffusivity parallel (
1) and perpendicular (2 and 3) to the principal diffusion direction were calculated.
DTI Data Analysis
We used Tract-Based Spatial Statistics (TBSS, also part of FSL), described in depth elsewhere (Smith et al. 2006), to test for local correlations between electrophysiological measures and FA across the whole brain WM. All subjects’ FA volumes were nonlinearly aligned into a common space by means of spline-based free-form deformation using the nonlinear registration tool Image Registration Toolkit (IRTK; Rueckert et al. 1999) implemented in FSL. Then, a mean FA volume of all subjects was generated and thinned to create a mean FA "skeleton" representing the centers of all tracts common to the participants. The skeleton was thresholded at FA > 0.2 and included 136 392 1 x 1 x 1 mm WM voxels. Individual participants’ FA values were warped onto this mean skeleton mask by searching perpendicular from the skeleton for maximum FA values. Using maximum FA values from the centers of the tracts minimizes confounding effects from partial voluming in the borders between tissue types (Smith et al. 2007). The resulting tract invariant skeleton for each participant was then fed into voxel-wise cross-subject statistics.
To test for regional specific effects we confined the statistical tests to the cingulum bundles. The ROIs used in the subsequent statistical analyses are depicted in Figure 4. These ROIs were chosen a priori because the principle neuronal generators of the ERN are located in anterior regions of the cingulate gyrus (Taylor et al. 2007). A probabilistically defined neuroanatomical atlas (Wakana et al. 2004; Mori et al. 2005) provided in FSL was used to localize the cingulum bundle in the mean FA volume. This area was masked by the mean TBSS skeleton in order to include only voxels included in the skeleton. Further, in addition to the probabilistic labeling, we also manually labeled voxels in the cingulum bundle not included by the probabilistically defined atlas. After this extension of the ROIs, they were thinned across space and had the anterior most (to the genu of the corpus callosum; CC) and posterior most (to the splenium of the CC) parts removed. Restricting the ROIs to regions earlier reported (Taylor et al. 2007) to be crucial in ERN generation eases the multiple comparison burden, reducing the probability of type I errors. The left and right cingulum ROI included 490 and 519 voxels, respectively.
|
To test for regional nonspecific effects of FA on ERN, linear regression analysis with mean FA in the TBSS skeleton and ERN amplitude was performed. To test for local correlations and anatomical specificity, a general linear model (GLM) of the effect of FA on ERN amplitude was fitted in every voxel included in the skeleton, and the number of significant voxels (P < 0.05, uncorrected) within different probabilistically defined WM tracts were counted. The difference between number of negative and positive correlations within each WM tract was tested with paired sampled t-test.
Statistical Inference
For statistical analyses the randomize tool within FSL was used to carry out permutation-based (Nichols and Holmes 2002) cluster size thresholding (Worsley 2001) within the ROIs. Clusters were defined by thresholding the raw t-statistics map of the skeleton at t > 2.5, and then searching for contiguous clusters of suprathreshold voxels by the means of 26-neighbor connectivity (Giorgio et al. 2008). The null distribution of the cluster size statistic was built up over 5000 random permutations of the effect of ERN on FA, recording the maximum size across space at every permutation. The clusters were then thresholded at P < 0.05 corrected for multiple comparisons. Age (demeaned) and sex were used as covariates in the permutation-based analyses to control for possible confounding age or sex related effects. Next, mean FA within significant clusters was calculated for each participant. These values were fed to a linear regression analysis and plotted against ERN amplitude to visualize the between subject distribution of FA. We also tested the relationship between principal and radial diffusivity, respectively, and ERN in order to further explore the neurobiological underpinnings of the effects of DTI indices on ERN amplitude. This was done by thresholding the statistical cluster size maps at P < 0.05 and select significant clusters from the skeleton based GLM analyses. Significant clusters were nonlinearly deprojected back to native space of each participant by means of trilinear interpolation. The individually back projected clusters were used as masks on the 1, 2 and 3 volume of each participant calculating principal (1) and radial ((2 + 3)/2) diffusivity in this region for each subject. The same overall strategy was employed for the whole skeleton analysis. The resulting voxel-based P statistic map was thresholded at P < 0.05 (uncorrected) and displayed for visual validation purposes.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Behavioral Results (Flanker Task)
Median RT and mean accuracy (% correct) data from the flanker task are given in Table 2. As a significant congruence-incongruence effect was used as an inclusion criteria, RT was naturally significantly slower in the incongruent compared with the congruent correct trials (t = 54.2, P < 0.0001). Not surprisingly, the accuracy was significantly (t = 15.6, P < 0.0001) higher for the congruent compared with the incongruent condition.
|
Electrophysiology and Behavioral Measures
Stronger ERN amplitude correlated with behavioral accuracy in both the incongruent (r = 0.29, P < 0.007) and the congruent condition (r = 0.39, P < 0.0002), predicting fewer errors in both conditions. Partialling out effects of age and sex did not affect the relationships (incongruent: r = 0.33, P < 0.002, congruent: r = 0.40, P < 0.0002).
DTI and Behavioral Measures
There was a significant correlation between number of correct responses in the congruent condition and mean FA in the reported effect site located in posterior cingulate cortex (PCC) (see below) (r = 0.27, P = 0.012). However, this effect was no longer significant after removal of the 2 subjects with the highest number of commission errors (Z score < –3.5) in the congruent condition (r = 0.202, P = 0.64). There were no other significant correlations between behavior (accuracy in the incongruent condition, RT measures) and any of the reported DTI indices (mean total FA, mean FA in skeleton, local ROI analyses, see below).
DTI and ERN
The result of the permutation and cluster size inference based voxel by voxel statistics in each of the 2 a priori selected cingulum ROIs is displayed in Figure 5. The analysis revealed that a supra threshold cluster of 13 voxels in the posterior part of the left cingulum bundle reached significance at P < 0.05 corrected for multiple comparisons across space. Mean FA over all subjects in the cluster was 0.55 (SD = 0.05, range: 0.43–0.68). There was a significant positive correlation (more negative ERN 
higher FA) between mean FA in the cluster and ERN amplitude (t = 3.04, F = 9.24, df = 86, β =. 313, P = 0.003). The localization of the cluster and the regression plot are given in Figure 5.
|
The individually interpolated and back projected cluster is exhibited overlaid on FA volumes in Figure 6 for 40 participants comprising subjects showing the strongest (n = 20) and weakest (n = 20) ERN amplitude. As can be seen in the figure, TBSS provides an excellent alignment between the skeleton and native space of each participant, further validating the regional localization of the FA effect on ERN amplitude.
|
The relation between ERN and principal (
1) and radial (2 + 3/2) diffusivity is given in the plots in Figure 7. The plots show a significant negative correlation between mean radial diffusivity in the cluster and ERN amplitude Z scores (t = –2.78, F = 7.71, df = 86, β = –0.288, P < 0.007), indicating less radial diffusion in participants with more negative ERN deflection. No relation was found between ERN and the mean principal eigenvalue in the same cluster. Thus, the significant relationship between FA and ERN is primarily driven by diffusivity perpendicular to the axonal direction.
|
Mean FA in the individual subjects’ skeleton was not related to ERN amplitude (t = 1.09, F = 1.19, df = 86, β = 0.117, P = 0.28). As shown in Figure 8, voxel-based GLM analysis yielded several regions with significant (P < 0.05, uncorrected) correlations between FA and ERN. The number of positive correlations in the whole skeleton was substantially larger than negative correlations at P < 0.05 (9050 vs. 3724 voxels). At P < 0.01, the ratio was very similar (1711 vs. 740 voxels). The significant cluster found in the left PCC revealed by the permutation-based ROI analyses is clearly visible in the left hemisphere at x = –10 mm. Interestingly, clustering analyses by the means of a 26-neighbor connectivity search of contiguous voxels exceeding the P < 0.05 threshold from the whole skeleton TBSS revealed 4 positive clusters with more than 150 voxels included. These clusters roughly corresponds to (xyz center of gravity [COG] according to the MNI template) 1) left posterior cingulate gyrus extending into the splenium of CC and precuneal regions (COG: 110, 80.6, 99.8, 681 voxels included, apparent in Fig. 8 at x = –20 and x = –15), 2) the anterior limb of the right internal capsule/anterior thalamic radiation (COG: 69.6, 142, 78.3, 217 voxels included, apparent in Figure 8 at x = 20 and x = 15), 3) left calossal body extending into the left cingulum bundle (COG: 99, 139, 97, 181 voxels included, apparent in Figure 8 at x = –15 and x = –10), 4) right posterior cingulate gyrus extending into the splenium of CC and precuneal areas (COG: 71, 89, 104, 177 voxels included, apparent in Figure 8 at x = 20). No clusters of negative correlations exceeded 150 voxels, but 1 cluster roughly corresponding to the left uncinate fasciculus (COG: 109, 174, 79.7) comprised 148 voxels. This cluster is clearly apparent in blue in Figure 8 at x = –20. Note that none of these clusters survived permutation-based cluster size inference correction for multiple comparisons across the whole skeleton. Hence, interpretations of the functional significance of these effect sites should be done with caution.
|
To further explore the regional distribution and test for anatomical specificity of the effects revealed by the whole skeleton analyses, we counted all skeleton voxels within 16 different probabilistically defined WM tracts exceeding the P < 0.05 threshold. These tracts are shown in Figure 9 and the percentage of all voxels showing positive and negative correlations in each tract is displayed in Figure 10 and summarized in Table 3. As seen in Figure 10, the effects were especially pronounced in the left cingulum bundle and the left superior longitudinal fasciculus. The number of positive correlations was significantly larger than the number of negative correlations (P < 0.05 as revealed by paired sampled t-tests, Bonferroni corrected for multiple tests) in 11 WM tracts (left/right cingulum, left superior longitudinal fasciculus (also when temporal projections included), forceps major, left inferior longitudinal fasciculus, right uncinate fasciculus, right corticospinal tract and right inferior occipitofrontal fasciculus). The opposite relation was found in 2 WM tracts (right superior longitudinal fasciculus and left uncinate fasciculus). There was no significant difference in the remaining 4 WM tracts (right superior longitudinal fasciculus, right inferior longitudinal fasciculus, left corticospinal tract and left inferior occipitofrontal fasciculus).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
The present results indicate that WM properties as indexed by FA correlate with the amplitude of the ERP component ERN. Specifically, a cluster of voxels in the posterior part of the a priori selected ROI in the left cingulum bundle was related to ERN amplitude. To our knowledge, no previous study has tested whether fast synchronization of neuronal processes in the brain as indexed by scalp measured electrophysiology is related to DTI measures of brain WM in healthy adults. However, the results are in accordance with theoretical assumptions about the importance of WM connectivity for cognitive processes. The lack of significant relations between DTI and behavioral accuracy, and the significant relationship between ERN amplitude and accuracy, indicates that ERPs may constitute an intermediate level linking DTI derived brain WM indices and behavior. Further studies are needed in order to confirm this hypothesis.
The relatively restricted localization of the cortical generator for ERN makes it a suitable candidate for testing the prediction that WM integrity in localized fiber tracts is predictive of the amplitude of the component. Our results suggest that such a hypothesis is warranted. Under the null distribution, one would expect a random and anatomically uniform distribution of the negative and positive correlations throughout the brain. However, sampling from 16 different neuroanatomical WM regions revealed a different pattern, confirming regionally specific distribution of significant voxels. 11 of 16 WM tracts tested showed a significant positive bias in respect to number of correlations and 2 WM tracts showed the opposite pattern. As expected, the most pronounced effects were found in the cingulum bundles, but also in the left superior longitudinal fasciculus. More surprising was the negative pattern found in the left uncinate fasciculus, an important WM tract connecting the left temporal lobe and the left prefrontal cortex. This finding could be incidental and further replications in different samples and by different methods are needed in order to validate the results.
The regional specificity of the effects found in the present study is intriguing, as it suggests that DTI derived measures of WM properties are intrinsically regionally informative, and not merely a measure of the global condition of the brain WM. Methodologically, this further illustrates the importance of including several WM regions in DTI analyses, extending the traditional employment of a small sample of ROIs. Voxel-based analyses as employed in TBSS may thus be a suitable approach in the study of the effects of individual differences in WM properties as revealed by DTI on aspects of neurocognition.
Functional Neuroanatomy of Posterior Cingulate
The present results point to WM properties in the posterior part of the cingulum bundle as important for scalp measured ERN amplitude. Source localization and lesion studies have typically reported a more anterior distribution of the neuronal generators for ERN. However, fibers in the posterior effect site in the present data probably project toward the anterior parts of the cingulate gyrus (Schmahmann and Pandya 2006), though the ventral and dorsal posterior cingulate and retrosplenial cortex have been found to exhibit anatomically and functionally distinct projections to different areas of the brain (Vogt et al. 2006). Thus, the ROIs used in the present study may represent fiber bundles with functional overlap and/or interdependence, although the voxel resolution in the present study naturally does not allow for detailed projection analyses on a neuronal level. Still, it is likely that reduced FA anywhere along the fibers in the cingulate gyrus will have effects related to cingulate function, for example, as measured by ERN amplitude.
The posterior cingulate is involved in different aspects of visuospatial orientation. It has been suggested that the ventral parts of PCC evaluates information arriving through the ventral visual stream, and interacts with the subgenual ACC in order to code for emotional content (Vogt et al. 2006). Vogt et al. (2006) propose that the ventral PCC both anatomically and functionally is at an intermediate stage of information processing between visual recognition in sensory areas and emotion-related regions in ACC. ERN may therefore be an index of early information processing partly subserved by the posterior cingulate gyrus prior to emotional flavoring by the ACC. Recently, the posterior cingulate gyrus has been tightly coupled to the default-mode network of brain function (Raichle et al. 2001; Fox and Raichle 2007; Mason et al. 2007), shown to be predictive of attentional lapses in cognitive control tasks (Weissman et al. 2006), and sensitive to normal aging, WM connectivity and cognitive functioning across a range of cognitive domains (Andrews-Hanna et al. 2007).
WM Integrity and Neuronal Activity
DTI measures have, with the restrictions mentioned above, been coupled to WM integrity, and are assumed related to cognitive function (Moseley et al. 2002; Sullivan and Pfefferbaum 2006). A few recent studies have reported relationships between WM FA and scores on cognitive and neuropsychological tests (Grieve et al. 2007; Charlton et al. forthcoming). These reports indicate that DTI indices are meaningful predictors of cognitive function. This view is also inherent in theories postulating that disconnection of cortical structures by reduced WM integrity is a major factor behind cognitive decrements with advancing age (Bartzokis et al. 2004). The present results suggest that a direct relationship between WM integrity and cognitive electrophysiological responses exists. Such a relationship may mediate correlations between DTI indices like FA and cognitive performance. A previous study using voxel-based morphometry found regional WM volume to be more closely related to ERP components than GM volume, which was interpreted by the authors suggesting that WM connectivity between neural generators influences ERPs more than the generators themselves (Cardenas et al. 2005). A central question regards whether DTI indices of WM properties are relevant to neuronal activity and cognitive function as a global measure of WM integrity, or whether more specific effects can be found. The present results indicate that FA in specific fiber tracts indeed is relevant for a specified type of neuronal activity.
ERN and WM Neurobiology
FA is an index computed by the principal (1) and radial ((2 + 3)/2) diffusivity (mm2/s) in every voxel (Pierpaoli and Basser 1996). Elevated FA may therefore reflect both reduced radial and/or increased principal diffusivity. Principal diffusivity is measured along the fiber orientation. Radial diffusivity is the relative degree of Brownian movement of water molecules perpendicular to the fiber orientation. The exact neurobiological causes of radial and principal diffusivity in brain WM are not known. Still, radial diffusivity has been related to the integrity and thickness of myelin sheaths covering the axons (Song et al. 2002) and axonal membranes (Beaulieu 2002), whereas principal diffusion may index axonal integrity (Mori and Zhang 2006). Our results indicate that diffusivity perpendicular to the axonal direction is responsible for the effects seen in the FA analyses. This may suggest that degree of myelination in the effect area is correlated with the scalp recorded electrophysiological potential. The participants were sampled from a healthy age homogenous middle-aged group (40–60 years). Hence, the results could be indexing normal individual differences in degree of myelination rather than indices of neurological illness or aging. The present study sampled the age range in which a breaking point in WM volume has been established (Walhovd et al. 2005), so one may reason that especially large individual differences in degree of myelination and demyelination could be present in this group compared with other age cohorts. Patient studies on samples exhibiting pathological myelin degradation would help further ameliorate the possible link between myelination and scalp recorded electrophysiological activity. Inclusion of younger cohorts is also needed to establish degree of stability over different age samples.
One should, however, show great caution making strong neurobiological inferences based on DTI derived measures. For example, FA is highly sensitive to crossing fibers, a phenomenon widely applicable to brain WM, especially as one gets closer to the cortical GM. Given both the relatively high mean FA in the significant cluster (0.55), the FA thresholding applied prior to TBSS as well as the non linear warping employed in TBSS itself, it is not very likely that the cluster is severely suffering from partial voluming effects due to sampling non WM brain tissue like cortical GM etc. However, it could still be argued that an alternative explanation of the present findings could be that number of fibers leaving the cingulate bundle and projecting into the cortex influences both the ERN and the radial diffusivity. In accordance with this, Tuch et al. (2005) argues that FA is not exclusively an index of degree of myelination, but is, as earlier mentioned, also highly influenced by factors like axon diameter and crossing fibers. Larger axon diameter would increase the radial diffusivity, and thus decrease FA. In regions with crossing fibers, the FA is highly influenced by the number and direction of these fiber tracts, complicating the biological interpretations. Even though some methods to estimate and differentiate multiple fibers in each voxel have been developed (Tuch et al. 2003; Tuch 2004; Behrens et al. 2007), the challenge of crossing fibers in DTI studies has not been resolved. More research on the biological underpinnings of diffusion anisotropy in the human brain is needed.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
FA in the posterior cingulum bundle was found to correlate positively with ERN amplitude in a sample of 87 healthy adults. The results indicate that less diffusivity perpendicular to the fiber orientation in persons with stronger ERN amplitude was the primary cause of the observed correlation between FA and ERN. Sampling different probabilistically defined functional WM fiber networks suggested regional specificity of the relationship between FA and ERN amplitude, validating WM cingulate involvement in ERN generation. It is encouraging for the potential use of DTI measures in studies of neurocognitive processes that a direct measure of cognitively significant neuronal activity is mediated by DTI measures of specific WM fiber tracts. Our results also suggest that ERN may be an intermediate explanatory variable between DTI and behavior. However, more studies combining similar methods are needed in order to strengthen this link. Also, further neurobiological validation of DTI measures will be crucial before one is able to make stronger neurobiological interpretations of the present results.
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Norwegian Research Council grants (177404/W50) to K.B.W., (175066/D15) to A.M.F.; a research student fellowship to L.T.W.; Institute of Psychology, University of Oslo (A.M.F.); and University of Oslo, Norway (K.B.W. and A.M.F.).
|
Acknowledgments |
|---|
Conflicts of Interest: None declared.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionFundingReferences |
|---|
Andrews-Hanna JR, Snyder AZ, Vincent JL, Lustig C, Head D, Raichle ME, Buckner RL. Disruption of large-scale brain systems in advanced aging. Neuron. (2007) 56:924–935.[CrossRef][Web of Science][Medline]
Bartzokis G. Myelination and brain electrophysiology in healthy and schizophrenic individuals. Neuropsychopharmacology. (2003) 28:1217–1218.[Web of Science][Medline]
Bartzokis G, Sultzer D, Lu PH, Nuechterlein KH, Mintz J, Cummings JL. Heterogeneous age-related breakdown of white matter structural integrity: implications for cortical "disconnection" in aging and Alzheimer's disease. Neurobiol Aging. (2004) 25:843–851.[CrossRef][Web of Science][Medline]
Beaulieu C. The basis of anisotropic water diffusion in the nervous system—a technical review. NMR Biomed. (2002) 15:435–455.[CrossRef][Web of Science][Medline]
Beck AaSR. Beck depression inventory scoring manual (1987) New York: The Psychological Corporation.
Behrens TE, Berg HJ, Jbabdi S, Rushworth MF, Woolrich MW. Probabilistic diffusion tractography with multiple fiber orientations: what can we gain? Neuroimage. (2007) 34:144–155.[CrossRef][Web of Science][Medline]
Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen JD. Conflict monitoring and cognitive control. Psychol Rev. (2001) 108:624–652.[CrossRef][Web of Science][Medline]
Bush G, Luu P, Posner MI. Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci. (2000) 4:215–222.[CrossRef][Web of Science][Medline]
Cardenas VA, Chao LL, Blumenfeld R, Song E, Meyerhoff DJ, Weiner MW, Studholme C. Using automated morphometry to detect associations between ERP latency and structural brain MRI in normal adults. Hum Brain Mapp. (2005) 25:317–327.[CrossRef][Web of Science][Medline]
Charlton RA, Barrick TR, McIntyre DJ, Shen Y, O'Sullivan M, Howe FA, Clark CA, Morris RG, Markus HS. White matter damage on diffusion tensor imaging correlates with age-related cognitive decline. Neurology. (2006) 66:217–222.
Charlton RA, Landau S, Schiavone F, Barrick TR, Clark CA, Markus HS, Morris RG. A structural equation modeling investigation of age-related variance in executive function and DTI measured white matter damage. Neurobiol Aging (2007) [Epub ahead of print].
Debener S, Ullsperger M, Siegel M, Fiehler K, von Cramon DY, Engel AK. Trial-by-trial coupling of concurrent electroencephalogram and functional magnetic resonance imaging identifies the dynamics of performance monitoring. J Neurosci. (2005) 25:11730–11737.
Eriksen BA, Eriksen CW. Effects of noise letters upon the identification of a target letter in a nonsearch task. Percept Psychophys. (1974) 16:143–149.[Web of Science]
Falkenstein M, Hoormann J, Christ S, Hohnsbein J. ERP components on reaction errors and their functional significance: a tutorial. Biol Psychol. (2000) 51:87–107.[CrossRef][Web of Science][Medline]
Fellgiebel A, Dellani PR, Greverus D, Scheurich A, Stoeter P, Muller MJ. Predicting conversion to dementia in mild cognitive impairment by volumetric and diffusivity measurements of the hippocampus. Psychiatry Res. (2006) 146:283–287.[CrossRef][Web of Science][Medline]
Fellgiebel A, Wille P, Muller MJ, Winterer G, Scheurich A, Vucurevic G, Schmidt LG, Stoeter P. Ultrastructural hippocampal and white matter alterations in mild cognitive impairment: a diffusion tensor imaging study. Dement Geriatr Cogn Disord. (2004) 18:101–108.[CrossRef][Web of Science][Medline]
Folstein MF, Folstein SE, McHugh PR. Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. (1975) 12:189–198.[CrossRef][Web of Science][Medline]
Fox MD, Raichle ME. Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging. Nat Rev Neurosci. (2007) 8:700–711.[CrossRef][Web of Science][Medline]
Gehring WJ, Goss B, Coles MGH, Meyer DE, Donchin E. A neural system for error detection and compensation. Psychol Sci. (1993) 4:385–390.[CrossRef][Web of Science]
Gehring WJ, Knight RT. Prefrontal-cingulate interactions in action monitoring. Nat Neurosci. (2000) 3:516–520.[CrossRef][Web of Science][Medline]
Giorgio A, Watkins KE, Douaud G, James AC, James S, De Stefano N, Matthews PM, Smith SM, Johansen-Berg H. Changes in white matter microstructure during adolescence. Neuroimage. (2008) 39:52–61.[CrossRef][Web of Science][Medline]
Grieve SM, Williams LM, Paul RH, Clark CR, Gordon E. Cognitive aging, executive function, and fractional anisotropy: a diffusion tensor MR imaging study. Am J Neuroradiol. (2007) 28:226–235.
Gulani V, Webb AG, Duncan ID, Lauterbur PC. Apparent diffusion tensor measurements in myelin-deficient rat spinal cords. Magn Reson Med. (2001) 45:191–195.[CrossRef][Web of Science][Medline]
Head D, Buckner RL, Shimony JS, Williams LE, Akbudak E, Conturo TE, McAvoy M, Morris JC, Snyder AZ. Differential vulnerability of anterior white matter in nondemented aging with minimal acceleration in dementia of the Alzheimer type: evidence from diffusion tensor imaging. Cereb Cortex. (2004) 14:410–423.
Herrmann MJ, Rommler J, Ehlis AC, Heidrich A, Fallgatter AJ. Source localization (LORETA) of the error-related-negativity (ERN/Ne) and positivity (Pe). Brain Res Cogn Brain Res. (2004) 20:294–299.[CrossRef][Medline]
Hogan AM, Vargha-Khadem F, Saunders DE, Kirkham FJ, Baldeweg T. Impact of frontal white matter lesions on performance monitoring: ERP evidence for cortical disconnection. Brain. (2006) 129:2177–2188.
Holroyd CB, Coles MG. The neural basis of human error processing: reinforcement learning, dopamine, and the error-related negativity. Psychol Rev. (2002) 109:679–709.[CrossRef][Web of Science][Medline]
Holroyd CB, Hajcak G, Larsen JT. The good, the bad and the neutral: electrophysiological responses to feedback stimuli. Brain Res. (2006) 1105:93–101.[CrossRef][Web of Science][Medline]
Holroyd CB, Nieuwenhuis S, Yeung N, Cohen JD. Errors in reward prediction are reflected in the event-related brain potential. Neuroreport. (2003) 14:2481–2484.[CrossRef][Web of Science][Medline]
Hugenschmidt CE, Peiffer AM, Kraft RA, Casanova R, Deibler AR, Burdette JH, Maldjian JA, Laurienti PJ. Relating imaging indices of white matter integrity and volume in healthy older adults. Cereb Cortex. (2008) 18:433–442.
Le Bihan D. Looking into the functional architecture of the brain with diffusion MRI. Nat Neurosci. (2003) 4:469–480.[CrossRef][Web of Science]
Luck SJ. An introduction to the event-related potential technique (2005) Cambridge: Massachusetts Institute of Technology Press.
Madden DJ, Whiting WL, Huettel SA, White LE, MacFall JR, Provenzale JM. Diffusion tensor imaging of adult age differences in cerebral white matter: relation to response time. Neuroimage. (2004) 21:1174–1181.[CrossRef][Web of Science][Medline]
Mason MF, Norton MI, Van Horn JD, Wegner DM, Grafton ST, Macrae CN. Wandering minds: the default network and stimulus-independent thought. Science. (2007) 315:393–395.
Mathalon DH, Whitfield SL, Ford JM. Anatomy of an error: ERP and fMRI. Biol Psychol. (2003) 64:119–141.[CrossRef][Web of Science][Medline]
Mori S, Wakana S, van Zijl PCM, Nagae-Poetscher LM. MRI atlas of human white matter (2005) Amsterdam: Elsevier.
Mori S, Zhang J. Principles of diffusion tensor imaging and its applications to basic neuroscience research. Neuron. (2006) 51:527–539.[CrossRef][Web of Science][Medline]
Moseley M. Diffusion tensor imaging and aging—a review. NMR Biomed. (2002) 15:553–560.[CrossRef][Web of Science][Medline]
Moseley M, Bammer R, Illes J. Diffusion-tensor imaging of cognitive performance. Brain Cogn. (2002) 50:396–413.[CrossRef][Web of Science][Medline]
Nichols TE, Holmes AP. Nonparametric permutation tests for functional neuroimaging: a primer with examples. Hum Brain Mapp. (2002) 15:1–25.[CrossRef][Web of Science][Medline]
Nieuwenhuis S, Holroyd CB, Mol N, Coles MG. Reinforcement-related brain potentials from medial frontal cortex: origins and functional significance. Neurosci Biobehav Rev. (2004) 28:441–448.[CrossRef][Web of Science][Medline]
O'Sullivan M, Barrick TR, Morris RG, Clark CA, Markus HS. Damage within a network of white matter regions underlies executive dysfunction in CADASIL. Neurology. (2005) 65:1584–1590.
Olesen PJ, Nagy Z, Westerberg H, Klingberg T. Combined analysis of DTI and fMRI data reveals a joint maturation of white and grey matter in a fronto-parietal network. Brain Res Cogn Brain Res. (2003) 18:48–57.[CrossRef][Medline]
Persson J, Nyberg L, Lind J, Larsson A, Nilsson LG, Ingvar M, Buckner RL. Structure-function correlates of cognitive decline in aging. Cereb Cortex. (2006) 16:907–915.
Pierpaoli C, Basser PJ. Toward a quantitative assessment of diffusion anisotropy. Magn Reson Med. (1996) 36:893–906.[Web of Science][Medline]
Piguet O, Double KL, Kril JJ, Harasty J, Macdonald V, McRitchie DA, Halliday GM. White matter loss in healthy ageing: a postmortem analysis. Neurobiol Aging (2007) [Epub ahead of print].
Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. A default mode of brain function. Proc Natl Acad Sci USA. (2001) 98:676–682.
Reese TG, Heid O, Weisskoff RM, Wedeen VJ. Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echo. Magn Reson Med. (2003) 49:177–182.[CrossRef][Web of Science][Medline]
Rueckert D, Sonoda LI, Hayes C, Hill DL, Leach MO, Hawkes DJ. Nonrigid registration using free-form deformations: application to breast MR images. IEEE Trans Med Imaging. (1999) 18:712–721.[CrossRef][Web of Science][Medline]
Salat DH, Tuch DS, Greve DN, van der Kouwe AJ, Hevelone ND, Zaleta AK, Rosen BR, Fischl B, Corkin S, Rosas HD, et al. Age-related alterations in white matter microstructure measured by diffusion tensor imaging. Neurobiol Aging. (2005) 26:1215–1227.[CrossRef][Web of Science][Medline]
Salat DH, Tuch DS, Hevelone ND, Fischl B, Corkin S, Rosas HD, Dale AM. Age-related changes in prefrontal white matter measured by diffusion tensor imaging. Ann N Y Acad Sci. (2005) 1064:37–49.[CrossRef][Web of Science][Medline]
Schlösser RG, Nenadic I, Wagner G, Gullmar D, von Consbruch K, Kohler S, Schultz CC, Koch K, Fitzek C, Matthews PM, et al. White matter abnormalities and brain activation in schizophrenia: a combined DTI and fMRI study. Schizophr Res. (2007) 89:1–11.[CrossRef][Web of Science][Medline]
Schmahmann JD, Pandya DN. Fiber pathways of the brain (2006) New York: Oxford University Press.
Seghier ML, Lazeyras F, Zimine S, Maier SE, Hanquinet S, Delavelle J, Volpe JJ, Huppi PS. Combination of event-related fMRI and diffusion tensor imaging in an infant with perinatal stroke. Neuroimage. (2004) 21:463–472.[CrossRef][Web of Science][Medline]
Semlitsch HV, Anderer P, Schuster P, Presslich O. A solution for reliable and valid reduction of ocular artifacts, applied to the P300 ERP. Psychophysiology. (1986) 23:695–703.[Web of Science][Medline]
Smith SM. Fast robust automated brain extraction. Hum Brain Mapp. (2002) 17:143–155.[CrossRef][Web of Science][Medline]
Smith SM, Jenkinson M, Johansen-Berg H, Rueckert D, Nichols TE, Mackay CE, Watkins KE, Ciccarelli O, Cader MZ, Matthews PM, et al. Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data. Neuroimage. (2006) 31:1487–1505.[CrossRef][Web of Science][Medline]
Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TE, Johansen-Berg H, Bannister PR, De Luca M, Drobnjak I, Flitney DE, et al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage. (2004) 23(Suppl. 1):S208–S219.[CrossRef][Web of Science][Medline]
Smith SM, Johansen-Berg H, Jenkinson M, Rueckert D, Nichols TE, Miller KL, Robson MD, Jones DK, Klein JC, Bartsch AJ, et al. Acquisition and voxelwise analysis of multi-subject diffusion data with tract-based spatial statistics. Nat Protoc. (2007) 2:499–503.[CrossRef][Medline]
Song SK, Sun SW, Ramsbottom MJ, Chang C, Russell J, Cross AH. Dysmyelination revealed through MRI as increased radial (but unchanged principal) diffusion of water. Neuroimage. (2002) 17:1429–1436.[CrossRef][Web of Science][Medline]
Stemmer B, Segalowitz SJ, Witzke W, Schonle PW. Error detection in patients with lesions to the medial prefrontal cortex: an ERP study. Neuropsychologia. (2004) 42:118–130.[CrossRef][Web of Science][Medline]
Sullivan EV, Pfefferbaum A. Diffusion tensor imaging and aging. Neurosci Biobehav Rev. (2006) 30:749–761.[CrossRef][Web of Science][Medline]
Taylor SF, Stern ER, Gehring WJ. Neural systems for error monitoring: recent findings and theoretical perspectives. Neuroscientist. (2007) 13:160–172.
Tuch DS. Q-ball imaging. Magn Reson Med. (2004) 52:1358–1372.[CrossRef][Web of Science][Medline]
Tuch DS, Reese TG, Wiegell MR, Wedeen VJ. Diffusion MRI of complex neural architecture. Neuron. (2003) 40:885–895.[CrossRef][Web of Science][Medline]
Tuch DS, Salat DH, Wisco JJ, Zaleta AK, Hevelone ND, Rosas HD. Choice reaction time performance correlates with diffusion anisotropy in white matter pathways supporting visuospatial attention. Proc Natl Acad Sci USA. (2005) 102:12212–12217.
Ullsperger M, von Cramon DY. The role of intact frontostriatal circuits in error processing. J Cogn Neurosci. (2006) 18:651–664.[CrossRef][Web of Science][Medline]
van Veen V, Carter CS. The anterior cingulate as a conflict monitor: fMRI and ERP studies. Physiol Behav. (2002) 77:477–482.[CrossRef][Medline]
van Veen V, Carter CS. Error detection, correction, and prevention in the brain: a brief review of data and theories. Clin Electroencephalogr Neurosci. (2006) 37:330–335.
Varela F, Lachaux JP, Rodriguez E, Martinerie J. The brainweb: phase synchronization and large-scale integration. Nat Rev Neurosci. (2001) 2:229–239.[CrossRef][Web of Science][Medline]
Vogt BA, Vogt L, Laureys S. Cytology and functionally correlated circuits of human posterior cingulate areas. Neuroimage. (2006) 29:452–466.[CrossRef][Web of Science][Medline]
Wakana S, Jiang H, Nagae-Poetscher LM, van Zijl PC, Mori S. Fiber tract-based atlas of human white matter anatomy. Radiology. (2004) 230:77–87.
Walhovd KB, Fjell AM, Reinvang I, Lundervold A, Dale AM, Eilertsen DE, Quinn BT, Salat D, Makris N, Fischl B. Effects of age on volumes of cortex, white matter and subcortical structures. Neurobiol Aging. (2005) 26:1261–1270.[CrossRef][Web of Science][Medline]
Wechsler D. Wechsler Abbreviated Scale of Intelligence (1999) San Antonio (TX): The Psychological Corporation.
Weissman DH, Roberts KC, Visscher KM, Woldorff MG. The neural bases of momentary lapses in attention. Nat Neurosci. (2006) 9:971–978.[CrossRef][Web of Science][Medline]
Worsley KJ. Statistical analysis of activation images. In: Functional MRI: an introduction to methods—Jezzard PM, Matthews PM, Smith SM, eds. (2001) New York: Oxford University Press. 251–270.
Wozniak JR, Lim KO. Advances in white matter imaging: a review of in vivo magnetic resonance methodologies and their applicability to the study of development and aging. Neurosci Biobehav Rev. (2006) 30:762–774.[CrossRef][Web of Science][Medline]
Yeung N, Cohen JD, Botvinick MM. The neural basis of error detection: conflict monitoring and the error-related negativity. Psychol Rev. (2004) 111:931–959.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. K. Tamnes, Y. Ostby, A. M. Fjell, L. T. Westlye, P. Due-Tonnessen, and K. B. Walhovd Brain Maturation in Adolescence and Young Adulthood: Regional Age-Related Changes in Cortical Thickness and White Matter Volume and Microstructure Cereb Cortex, June 11, 2009; (2009) bhp118v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||