Biology subjects

Ghose, G. M.

Publications and source records attributed to Ghose, G. M..

2 recordsLinked to original sources

bioRxiv2021

Realistic alpha oscillations and transient responses in a cortical microcircuit model

Neural-mass modeling of neural population data (EEG, ECoG, or LFPs) has shown promise both in elucidating the neural processes underlying cortical rhythms and changes in brain state, as well as offering a framework for testing the interplay between these rhythms and information processing. Models of cortical alpha rhythms (8 - 12 Hz) and their impact in visual sensory processing have been at the forefront of this effort, with the Jansen-Rit being one of the more popular models in this domain. The Jansen-Rit model, however, fails in reproducing key physiological observations including the level of inputs that cortical neurons receive and their responses to visual transients. To address these issues we generated a neural mass model that complies better with synaptic mediated dynamics, intrinsic alpha behavior, and produces realistic responses. The model is robust to many changes in parameter values but critically depends on the ratio of excitation to inhibition, producing response transients whose features are dependent on this ratio and alpha phase and power. The model is sufficiently flexible so as to be able to easily replicate the range of low frequency oscillations observed in different studies. Consistent with experimental observations, we find phase-dependent response dynamics to both visual and electrical stimulation using this model. The model suggests that stimulation facilitates alpha at particular phases and suppresses it in others due to a phase dependent lag in inhibitory responses. Hence, the model generates insight into the physiological parameters responsible for intrinsic oscillations and testable hypotheses regarding the interactions between visual and electrical stimulation on those oscillations. Author summaryWhile there is increasing evidence of the fundamental role brain states play in shaping information processing in the cerebral cortex, a mechanistic understanding of how those brain states are manifested and alter the signals underlying sensory processing and decision making has proved challenging. To address this issue we have modeled a well established signature of inattention in visual cortex: synchronized low frequency (8 - 12 Hz) oscillations. To allow for inferences regarding the local generation of these rhythms within a cortical microcircuit we used a neural mass model approach that incorporates physiologically realistic interactions between 3 populations of neurons. Our model is able to explain a variety of experimental observations that previous neural mass models have not, including spontaneous rhythms in the absence of input, the faithful transmission of strong input transients, a range of oscillation frequencies, and phase dependent visual responses. The model is robust to a range of parameters, but critically depends on the balance between local excitation and inhibition.

neuroscience↗

bioRxiv2021

Improving Sensitivity to Functional Responses without a Loss of Spatiotemporal Precision in Human Brain Imaging

As the neuroimaging field moves towards detecting smaller effects at higher spatial resolutions, and faster sampling rates, there is increased attention given to the deleterious contribution of unstructured, thermal noise. Here, we critically evaluate the performance of a recently developed reconstruction method, termed NORDIC, for suppressing thermal noise using datasets acquired with various field strengths, voxel sizes, sampling rates, and task designs. Following minimal preprocessing, statistical activation (t-values) of NORDIC processed data was compared to the results obtained with alternative denoising methods. Additionally, we examined the consistency of the estimates of task responses at the single-voxel, single run level, using a finite impulse response (FIR) model. To examine the potential impact on effective image resolution, the overall smoothness of the data processed with different methods was estimated. Finally, to determine if NORDIC alters or removes important temporal information, we employed an exhaustive leave-p-out cross validation approach, using FIR task responses to predict held out timeseries, quantified using R2. After NORDIC, the t-values are increased, an improvement comparable to what could be achieved by 1.5 voxels smoothing, and task events are clearly visible and have less cross-run error. These advantages are achieved in the absence of large changes in estimates of spatial smoothness. Cross-validated R2s based on the FIR models show that NORDIC is not measurably distorting the temporal structure of the data and is the best predictor of non-denoised time courses. The results demonstrate that analyzing 1 run of data after NORDIC produces results equivalent to using 2 to 3 original runs and that NORDIC performs equally well across a diverse array of functional imaging protocols. Significance StatementFor functional neuroimaging, the increasing availability of higher field strengths and ever higher spatiotemporal resolutions has led to concomitant increase in concerns about the deleterious effects of thermal noise. Historically this noise source was suppressed using methods that reduce spatial precision such as image blurring or averaging over a large number of trials or sessions, which necessitates large data collection efforts. Here, we critically evaluate the performance of a recently developed reconstruction method, termed NORDIC. Across datasets varying in field strength, voxel sizes, sampling rates, and task designs, NORDIC produces substantial gains in data quality. Both conventional t-statistics derived from general linear models and coefficients of determination for predicting unseen data are improved, while avoiding meaningful increases in typical estimates of image smoothness or substantial losses of temporal information.

neuroscience↗