Rasmus M. Birn

The impact of global signal regression on resting state correlations: Are anti-correlated networks introduced?

Low-frequency fluctuations in fMRI signal have been used to map several consistent resting state networks in the brain. Using the posterior cingulate cortex as a seed region, functional connectivity analyses have found not only positive correlations in the default mode network but negative correlations in another resting state network related to attentional processes. The interpretation is that the human brain is intrinsically organized into dynamic, anti-correlated functional networks. Global variations of the BOLD signal are often considered nuisance effects and are commonly removed using a general linear model (GLM) technique. This global signal regression method has been shown to introduce negative activation measures in standard fMRI analyses. The topic of this paper is whether such a correction technique could be the cause of anti-correlated resting state networks in functional connectivity analyses. Here we show that, after global signal regression, correlation values to a seed voxel must sum to a negative value. Simulations also show that small phase differences between regions can lead to spurious negative correlation values. A combination breath holding and visual task demonstrates that the relative phase of global and local signals can affect connectivity measures and that, experimentally, global signal regression leads to bell-shaped correlation value distributions, centred on zero. Finally, analyses of negatively correlated networks in resting state data show that global signal regression is most likely the cause of anti-correlations. These results call into question the interpretation of negatively correlated regions in the brain when using global signal regression as an initial processing step.

Disrupted modularity and local connectivity of brain functional networks in childhood-onset schizophrenia.

Modularity is a fundamental concept in systems neuroscience, referring to the formation of local cliques or modules of densely intra-connected nodes that are sparsely inter-connected with nodes in other modules. Topological modularity of brain functional networks can quantify theoretically anticipated abnormality of brain network community structure – so-called dysmodularity – in developmental disorders such as childhood-onset schizophrenia (COS). We used graph theory to investigate topology of networks derived from resting-state fMRI data on 13 COS patients and 19 healthy volunteers. We measured functional connectivity between each pair of 100 regional nodes, focusing on wavelet correlation in the frequency interval 0.05–0.1 Hz, then applied global and local thresholding rules to construct graphs from each individual association matrix over the full range of possible connection densities. We show how local thresholding based on the minimum spanning tree facilitates group comparisons of networks by forcing the connectedness of sparse graphs. Threshold-dependent graph theoretical results are compatible with the results of a k-means unsupervised learning algorithm and a multi-resolution (spin glass) approach to modularity, both of which also find community structure but do not require thresholding of the association matrix. In general modularity of brain functional networks was significantly reduced in COS, due to a relatively reduced density of intra-modular connections between neighboring regions. Other network measures of local organization such as clustering were also decreased, while complementary measures of global efficiency and robustness were increased, in the COS group. The group differences in complex network properties were mirrored by differences in simpler statistical properties of the data, such as the variability of the global time series and the internal homogeneity of the time series within anatomical regions of interest.

Event-related fMRI of tasks involving brief motion

The assessment of brain function by blood oxygenation level dependent (BOLD) functional magnetic resonance imaging (fMRI) for tasks involving motion near the field of view is compromised by artifacts arising from the motion. The aim of this study is to demonstrate that these artifacts can be reduced by acquiring the average response from a brief stimulus (a “single‐trial,” or “event‐related,” paradigm) as opposed to alternating blocks of repeated task with rest (a “block‐trial” paradigm). The basis of this technique is that the NMR signal changes from neuronal activation are delayed relative to the motion due to a slow hemodynamic response. By acquiring the average response from a brief stimulus, motion‐induced signal changes occur prior to neuronal activation‐induced signal changes, and the two can thus be distinguished. This technique is applied to the tasks of speaking out loud, swallowing, jaw clenching, and tongue movement. Functional activation maps derived from the single‐trial paradigm contain significantly less artifact than functional activation maps derived from a more traditional block‐trial paradigm. Hum. Brain Mapping 7:106–114, 1999.

The effect of respiration variations on independent component analysis results of resting state functional connectivity

The analysis of functional connectivity in fMRI can be severely affected by cardiac and respiratory fluctuations. While some of these artifactual signal changes can be reduced by physiological noise correction routines, signal fluctuations induced by slower breath‐to‐breath changes in the depth and rate of breathing are typically not removed. These slower respiration‐induced signal changes occur at low frequencies and spatial locations similar to the fluctuations used to infer functional connectivity, and have been shown to significantly affect seed‐ROI or seed‐voxel based functional connectivity analysis, particularly in the default mode network. In this study, we investigate the effect of respiration variations on functional connectivity maps derived from independent component analysis (ICA) of resting‐state data. Regions of the default mode network were identified by deactivations during a lexical decision task. Variations in respiration were measured independently and correlated with the MRI time series data. ICA appears to separate the default mode network and the respiration‐related changes in most cases. In some cases, however, the component automatically identified as the default mode network was the same as the component identified as respiration‐related. Furthermore, in most cases the time series associated with the default mode network component was still significantly correlated with changes in respiration volume per time, suggesting that current methods of ICA may not completely separate respiration from the default mode network. An independent measure of the respiration provides valuable information to help distinguish the default mode network from respiration‐related signal changes, and to assess the degree of residual respiration related effects. Hum Brain Mapp 2008.

Spatial heterogeneity of the nonlinear dynamics in the FMRI BOLD response

Recent studies of blood oxygenation level dependent (BOLD) signal responses averaged over a region of interest have demonstrated that the response is nonlinear with respect to stimulus duration. Specifically, shorter duration stimuli produce signal changes larger than expected from a linear system. The focus of this study is to characterize the spatial heterogeneity of this nonlinear effect. A series of MR images of the visual and motor cortexes were acquired during visual stimulation and finger tapping, respectively, at five different stimulus durations (SD). The nonlinearity was assessed by fitting ideal linear responses to the responses at each SD. This amplitude, which is constant for different SD in a linear system, was normalized by the amplitude of the response to a blocked design, thus describing the amount by which the stimulus is larger than predicted from a linear extrapolation of the response to the long duration stimulus. The amplitude of the BOLD response showed a nonlinear behavior that varied considerably and consistently over space, ranging from almost linear to 10 times larger than a linear prediction at short SD. In the motor cortex different nonlinear behavior was found in the primary and supplementary motor cortexes.

Neural systems supporting lexical search guided by letter and semantic category cues: a self-paced overt response fMRI study of verbal fluency.

Verbal fluency tasks have been widely used to evaluate language and executive control processes in the human brain. FMRI studies of verbal fluency, however, have used either silent word generation (which provides no behavioral measure) or cued generation of single words in order to contend with speech-related motion artifacts. In this study, we use a recently developed paradigm design to investigate the neural correlates of verbal fluency during overt, free recall, word generation so that performance and brain activity could be evaluated under conditions that more closely mirror standard behavioral test demands. We investigated verbal fluency to both letter and category cues in order to evaluate differential involvement of specific frontal and temporal lobe sites as a function of retrieval cue type, as suggested by previous neuropsychological and neuroimaging investigations. In addition, we incorporated both a task switching manipulation and an automatic speech condition in order to modulate the demand placed on executive functions. We found greater activation in the left hemisphere during category and letter fluency tasks, and greater right hemisphere activation during automatic speech. We also found that letter and category fluency tasks were associated with differential involvement of specific regions of the frontal and temporal lobes. These findings provide converging evidence that letter and category fluency performance is dependent on partially distinct neural circuitry. They also provide strong evidence that verbal fluency can be successfully evaluated in the MR environment using overt, self-paced, responses.

The effect of scan length on the reliability of resting-state fMRI connectivity estimates

There has been an increasing use of functional magnetic resonance imaging (fMRI) by the neuroscience community to examine differences in functional connectivity between normal control groups and populations of interest. Understanding the reliability of these functional connections is essential to the study of neurological development and degenerate neuropathological conditions. To date, most research assessing the reliability with which resting-state functional connectivity characterizes the brains functional networks has been on scans between 3 and 11 min in length. In our present study, we examine the test-retest reliability and similarity of resting-state functional connectivity for scans ranging in length from 3 to 27 min as well as for time series acquired during the same length of time but excluding half the time points via sampling every second image. Our results show that reliability and similarity can be greatly improved by increasing the scan lengths from 5 min up to 13 min, and that both the increase in the number of volumes as well as the increase in the length of time over which these volumes was acquired drove this increase in reliability. This improvement in reliability due to scan length is much greater for scans acquired during the same session. Gains in intersession reliability began to diminish after 9-12 min, while improvements in intrasession reliability plateaued around 12-16 min. Consequently, new techniques that improve reliability across sessions will be important for the interpretation of longitudinal fMRI studies.

Detection versus estimation in event-related fMRI: Choosing the optimal stimulus timing

With the advent of event-related paradigms in functional MRI, there has been interest in finding the optimal stimulus timing, especially when the interstimulus interval is varied during the imaging run. Previous works have proposed stimulus timings to optimize either the estimation of the impulse response function (IRF) or the detection of signal changes. The purpose of this paper is to clarify that estimation and detection are fundamentally different goals and to determine the optimal stimulus timing and distribution with respect to both the accuracy of estimating the IRF and the power of detection assuming a particular hemodynamic model. Simulated stimulus distributions are varied systematically, from traditional blocked designs to rapidly varying event related designs. These simulations indicate that estimation of the hemodynamic impulse response function is optimized when stimuli are frequently alternated between task and control states, with shorter interstimulus intervals and stimulus durations, whereas the detection of activated areas is optimized by blocked designs. The stimulus timing for a given experiment should therefore be generated with the required detectability and estimation accuracy.

The effect of resting condition on resting-state fMRI reliability and consistency: A comparison between resting with eyes open, closed, and fixated

Resting-state fMRI (rs-fMRI) has been demonstrated to have moderate to high reliability and produces consistent patterns of connectivity across a wide variety of subjects, sites, and scanners. However, there is no one agreed upon method to acquire rs-fMRI data. Some sites instruct their subjects, or patients, to lie still with their eyes closed, while other sites instruct their subjects to keep their eyes open or even fixating on a cross during scanning. Several studies have compared those three resting conditions based on connectivity strength. In our study, we assess differences in metrics of test-retest reliability (using an intraclass correlation coefficient), and consistency of the rank-order of connections within a subject and the ranks of subjects for a particular connection from one session to another (using Kendalls W tests). Twenty-five healthy subjects were scanned at three different time points for each resting condition, twice the same day and another time two to three months later. Resting-state functional connectivity measures were evaluated in motor, visual, auditory, attention, and default-mode networks, and compared between the different resting conditions. Of the networks examined, only the auditory network resulted in significantly higher connectivity in the eyes closed condition compared to the other two conditions. No significant between-condition differences in connectivity strength were found in default mode, attention, visual, and motor networks. Overall, the differences in reliability and consistency between different resting conditions were relatively small in effect size but results were found to be significant. Across all within-network connections, and within default-mode, attention, and auditory networks statistically significant greater reliability was found when the subjects were lying with their eyes fixated on a cross. In contrast, primary visual network connectivity was most reliable when subjects had their eyes open (and not fixating on a cross).

Childhood maltreatment is associated with altered fear circuitry and increased internalizing symptoms by late adolescence

Significance Childhood maltreatment is a major risk factor for internalizing disorders including depression and anxiety, which cause significant disability. Altered connectivity of the brain’s fear circuitry represents an important candidate mechanism linking maltreatment and these disorders, but this relationship has not been directly explored. Using resting-state functional brain connectivity in adolescents, we show that maltreatment predicts lower prefrontal–hippocampal connectivity in females and males but lower prefrontal–amygdala connectivity only in females. Altered connectivity, in turn, mediated the development of internalizing symptoms. These results highlight the importance of fronto–hippocampal connectivity for both sexes in internalizing symptoms following maltreatment. The additional impact on fronto–amygdala connectivity in females may help explain their higher risk for anxiety and depression. Maltreatment during childhood is a major risk factor for anxiety and depression, which are major public health problems. However, the underlying brain mechanism linking maltreatment and internalizing disorders remains poorly understood. Maltreatment may alter the activation of fear circuitry, but little is known about its impact on the connectivity of this circuitry in adolescence and whether such brain changes actually lead to internalizing symptoms. We examined the associations between experiences of maltreatment during childhood, resting-state functional brain connectivity (rs-FC) of the amygdala and hippocampus, and internalizing symptoms in 64 adolescents participating in a longitudinal community study. Childhood experiences of maltreatment were associated with lower hippocampus–subgenual cingulate rs-FC in both adolescent females and males and lower amygdala–subgenual cingulate rs-FC in females only. Furthermore, rs-FC mediated the association of maltreatment during childhood with adolescent internalizing symptoms. Thus, maltreatment in childhood, even at the lower severity levels found in a community sample, may alter the regulatory capacity of the brain’s fear circuit, leading to increased internalizing symptoms by late adolescence. These findings highlight the importance of fronto–hippocampal connectivity for both sexes in internalizing symptoms following maltreatment in childhood. Furthermore, the impact of maltreatment during childhood on both fronto–amygdala and –hippocampal connectivity in females may help explain their higher risk for internalizing disorders such as anxiety and depression.