CNS 2014 Abstract: Nikolaidis, A., Reggente, N. et al
Using the the task-dependent functional connectivity across nodes within established cortical network from both before and after videogame training, we were able to asses the “plasticity” of these networks as a function of training. The plasticity of these networks, combined with graph theoretical metrics were used as features in a leave-one-out Ridge Regression that was able to account for upwards of 80% of the variance in individual difference scores in learning. Each network contributed varying levels of accuracy to classification depending on its involvement in the subject’s instructed priorities within the task. For example, plasticity in the Cingulo-Opercular network preferentially predicted(upwards of 55% of variance accounted for) learning in a training strategy that relied more heavily on executive control of attention and goal directed behavior.
Predicting individual differences in cognitive gains from videogame training using machine learning analyses of fMRI functional connectivity patterns
Aki Nikolaidis1, Nicco Reggente2, Drew Goatz3, Kathryn Hurley3, Andrew Westphal2, Arthur F. Kramer1; 1University of Illinois, Urbana Champaign, Beckman Institute, 2University of California, Los Angeles, Psychology Department, 3University of Illinois, Bioengineering Department
One of the important questions in cognitive training, and learning and memory more broadly, is how pre-existing individual differences in brain connectivity influence the effect of training. In this study, we use the fMRI functional connectivity of multiple networks, including the frontal-parietal and motor networks, to predict individual differences in learning over the course of 30 hours of cognitive training with the Space Fortress videogame. We used various metrics of functional connectivity and graph theory-derived parameters from 45 young adult participants as features to train adaptive multivariate regression models. Using a leave-one-participant-out cross-validation procedure, we find that we can predict a significant percentage of the variance in learning performance (defined as pre-post differences in Space Fortress score). By analyzing the performance of different regression models, we find that distinct brain networks contain different types of information regarding individual differences in learning rate. Furthermore, using both support vector regression and ridge regression we demonstrate how different feature and model parameters have important effects on model performance, and we consider how these parameters may have limited previous research using such techniques. We discuss implications of our results for cognitive training, as well as the continued use of machine learning and graph theoretical analyses in cognitive neuroscience.
CNS 2014 Poster: Zheng, Z., Reggente, N. et al. CNS 2014
For this poster we used Diffusion Tensor Imaging (DTI) techniques to compute thalamo-cortical probabilistic tractography maps. That is, for every voxel in the Thalamus, we determined every other voxel in the brain’s probabilist structural connectivity with that voxel. We did this for a total of 23 patients diagnosed with disorders of consciousness that fell into one of three possible clinical stratification (Vegetative State(VS), Minimally Conscious State+(MCS+), Minimally Conscious State-(MCS-)). We utilized a support vector machine to observe the patterns local anatomical connectivity of regions of cortex with the thalamus unique to subjects within each of the three groupings. We then used this classifier to “diagnose” unseen subjects in an extensive cross-validation. At times, particularly when using the patterns of activity local to the left prefrontal cortex, we were able to achieve 100% accuracy when distinguishing between VS and MCS+. We used a searchlight-mapping technique to determine which regions in the brain were most informative to classification. We are hoping that this technology could be found useful as a diagnostic aid that relies on the underlying pathology of patients with disorders of consciousness as opposed to the purely behavioral diagnoses currently utilized.
CNS 2014 Abstract:
Disentangling Disorders of Consciousness: Insights from DTI and MVPA
Zhong A. Zheng1, Nicco Reggente1, Evan S. Lutkenhoff1, Adrian Owen2, Martin M. Monti1; 1University of California, Los Angeles, 2University of Western Ontario
The stratification of individuals surviving severe brain injury in Minimally Conscious State (MCS) and Vegetative State (VS) patients is, currently, entirely based on behavioral criteria. This approach is problematic for at least two reasons: (i) behavioral assessments are known to be susceptible to sizeable misdiagnosis (~40%); (ii) this stratification of patients is entirely blind to the underlying pathology. To address both issues, we employed diffusion probabilistic tractography to assess projections from thalamic nuclei in 8 MCS plus (+) patients, who exhibit high-level behavioral responses, 8 MCS minus (-) patients, who only show low-level responses, and 8 VS patients. Evaluation of thalamo-cortical connectivity revealed more connections from the lateral-group nuclei to prefrontal, motor, and sensory regions in MCS+, as compared to VS. Additionally, tractography maps from thalamic nuclei were used as patterns in a logistic regression classification scheme. Using the ventral lateral nucleus’ whole-brain tractography maps as patterns, a leave-two-patients-out cross-validation correctly classified 6/8 VS patients and 7/8 MCS+ patients. This classification relied mostly on increased thalamo-frontal connections in MCS+ patients, as compared to VS. These results suggest that DTI combined with machine learning classification may facilitate the diagnostic distinction between VS and subcategories of MCS by uncovering the neural markers and pathological changes underlying disorders of consciousness.
This work was a collaboration with Zhong (Amy) Sheng Zheng.