Preserved anatomical bypasses predict variance in language functions after stroke.

The severity of post-stroke aphasia is related to damage to white matter connections. However, neural signaling can route not only through direct connections, but also along multi-step network paths. When brain networks are damaged by stroke, paths can bypass around the damage to restore communication. The shortest network paths between regions could be the most efficient routes for mediating bypasses. We examined how shortest-path bypasses after left hemisphere strokes were related to language performance. Regions within and outside of the canonical language network could be important in aphasia recovery. Therefore, we innovated methods to measure the influence of bypasses in the whole brain. Distinguishing bypasses from all residual shortest paths is difficult without pre-stroke imaging. We identified bypasses by finding shortest paths in subjects with stroke that were longer than the most reliably observed connections in age-matched control networks. We tested whether features of those bypasses predicted scores in four orthogonal dimensions of language performance derived from a principal components analysis of a battery of language tasks. The features were the length of each bypass in steps, and how many bypasses overlapped on each individual direct connection. We related these bypass features to language factors using support vector regression, a technique that extracts robust relationships in high-dimensional data analysis. The support vector regression parameters were tuned using grid-search cross-validation. We discovered that the length of bypasses reliably predicted variance in lexical production (R2 = .576) and auditory comprehension scores (R2 = .164). Bypass overlaps reliably predicted variance in Lexical Production scores (R2 = .247). The predictive elongation features revealed that bypass efficiency along the dorsal stream and ventral stream were most related to Lexical Production and Auditory Comprehension, respectively. Among the predictive bypass overlaps, increased bypass routing through the right hemisphere putamen was negatively related to lexical production ability.

The role of transcranial magnetic stimulation in understanding attention-related networks in single subjects.

Attention is a cognitive mechanism that has been studied through several methodological viewpoints, including animal models, MRI in stroke patients, and fMRI in healthy subjects. Activation-based fMRI research has also pointed to specific networks that activate during attention tasks. Most recently, network neuroscience has been used to study the functional connectivity of large-scale networks for attention to reveal how strongly correlated networks are to each other when engaged in specific behaviors. While neuroimaging has revealed important information about the neural correlates of attention, it is crucial to better understand how these processes are organized and executed in the brain in single subjects to guide theories and treatments for attention. Noninvasive brain stimulation is an effective tool to causally manipulate neural activity to detect the causal roles of circuits in behavior. We describe how combining transcranial magnetic stimulation (TMS) with modern precision network analysis in single-subject neuroimaging could test the roles of regions, circuits, and networks in regulating attention as a pathway to improve treatment effect magnitudes and specificity.

Examining network dynamics after traumatic brain injury using the extended unified SEM approach.

The current study uses effective connectivity modeling to examine how individuals with traumatic brain injury (TBI) learn a new task. We make use of recent advancements in connectivity modeling (extended unified structural equation modeling, euSEM) and a novel iterative grouping procedure (Group Iterative Multiple Model Estimation, GIMME) in order to examine network flexibility after injury. The study enrolled 12 individuals sustaining moderate and severe TBI to examine the influence of task practice on connections between 8 network nodes (bilateral prefrontal cortex, anterior cingulate, inferior parietal lobule, and Crus I in the cerebellum). The data demonstrate alterations in networks from pre to post practice and differences in the models based upon distinct learning trajectories observed within the TBI sample. For example, better learning in the TBI sample was associated with diminished connectivity within frontal systems and increased frontal to parietal connectivity. These findings reveal the potential for using connectivity modeling and the euSEM to examine dynamic networks during task engagement and may ultimately be informative regarding when networks are moving in and out of periods of neural efficiency.

Neural recruitment after mild traumatic brain injury is task dependent: a meta-analysis.

Individuals with mild traumatic brain injury (TBI) often have deficits in processing speed and working memory (WM) and there is a growing literature using functional imaging studies to document these deficits. However, divergent results from these studies revealed both hypoactivation and hyperactivation of neural resources after injury. We hypothesized that at least part of this variance can be explained by distinct demands between WM tasks. Notably, in this literature some WM tasks use discrete periods of encoding, maintenance, and retrieval, whereas others place continuous demands on WM. The purpose of this meta-analysis is to examine the differences in neural recruitment after mTBI to determine if divergent findings can be explained as a function of task demand and cognitive load. A comprehensive literature review revealed 14 studies using functional magnetic resonance imaging to examine brain activity of individuals with mTBI during working memory tasks. Three of the fourteen studies included reported hypoactivity, five reported hyperactivity, and the remaining six reported both hypoactivity and hyperactivity. Studies were grouped according to task type and submitted to GingerALE maximum likelihood meta-analyses to determine the most consistent brain activation patterns. The primary findings from this meta-analysis suggest that the discrepancy in activation patterns is at least partially attributable to the classification of WM task, with hyperactivation being observed in continuous tasks and hypoactivation being observed during discrete tasks. We anticipate that differential task load expressed in continuous and discrete WM tasks contributes to these differences. Implications for the interpretation of fMRI signals in clinical samples are discussed.

Changes in resting connectivity during recovery from severe traumatic brain injury.

In the present study we investigate neural network changes after moderate and severe traumatic brain injury (TBI) through the use of resting state functional connectivity (RSFC) methods. Using blood oxygen level dependent functional MRI, we examined RSFC at 3 and 6 months following resolution of posttraumatic amnesia. The goal of this study was to examine how regional off-task connectivity changes during a critical period of recovery from significant neurological disruption. This was achieved by examining regional changes in the intrinsic, or "resting", BOLD fMRI signal in separate networks: 1) regions linked to goal-directed (or external-state) networks and 2) default mode (or internal-state) networks. Findings here demonstrate significantly increased resting connectivity internal-state networks in the TBI sample during the first 6 months following recovery. The most consistent finding was increased connectivity in both internal and external state networks to the insula and medial temporal regions during recovery. These findings were dissociable from repeat measurements in a matched healthy control sample.