28 September 2018 in IEEE Transactions on Biomedical Engineering
It is largely unknown whether there is functional role difference between cortical gyral and sulcal regions. Recent advancements in neuroimaging studies demonstrate clear difference of structural connection profiles in gyral and sulcal areas, suggesting possible functional role difference in these convex and concave cortical regions. To explore and confirm such possible functional difference, we design and apply a powerful deep learning model of convolutional neural networks (CNN) that has been proven to be superior in learning discriminative and meaningful patterns on fMRI. By using the CNN model, gyral and sulcal fMRI signals are learned and predicted, and the prediction performance is adopted to demonstrate the functional difference between gyri and sulci. By using the Human Connectome Project (HCP) fMRI data and macaque brain fMRI data, an average of 83% and 90% classification accuracy has been achieved to separate gyral/sulcal HCP task fMRI signals at the population and individual subject level, respectively; 81% and 86% classification accuracy for resting state fMRI signals at the group and individual subject level, respectively. In addition, 78% classification accuracy has been achieved to separate gyral/sulcal resting state fMRI signals in macaque brains. Importantly, further analysis reveals that the discriminative features that are learned by CNNs to differentiate gyral/sulcal fMRI signals can be meaningfully interpreted, thus unveiling the fundamental functional difference between cortical gyri and sulci. That is, gyri are more global functional integration centers with simpler lower frequency signal components, while sulci are more local processing units with more complex higher frequency signal components.
22 September 2020 in IEEE Transactions on Cognitive and Developmental Systems
With selective auditory attention (SAA), the human brain can cope with dynamic natural environments efficiently by prioritizing relevant auditory information while filtering out the rest. Auditory saliency describes how bottom-up auditory attention is attracted by local acoustic properties and plays an important role in SAA. In existing neuroimaging studies, the neural processing of auditory saliency has usually been explored via strictly controlled laboratory paradigms with abstract stimuli, overlooking the intrinsic functional interactions among spatially distributed brain networks underlying auditory saliency. In addition, it is unclear how auditory saliency related neural responses vary in auditory information in different semantic categories. In this study, we explored functional brain networks underlying auditory saliency via a multivariate brain decoding approach, in which naturalistic auditory excerpts in three semantic categories (pop music, classic music and speech) were used as stimuli in functional magnetic resonance imaging (fMRI). Our experimental results demonstrate that: 1) Dynamic auditory saliency can be decoded with relatively high accuracy from fMRI brain activities; 2) The neural processing of auditory saliency in different types of auditory information on the one hand shares some common brain regions (e.g., the primary auditory cortex), on the other hand recruits distinct brain regions/networks, e.g., the sensorimotor network and working memory network for classic music and the auditory-language network for speech. Our study provides complementary evidence to the neural processing of auditory saliency under naturalistic experience.
16 June 2020 in IEEE Transactions on Cognitive and Developmental Systems
Dynamic visual scenes lead to varying consistency of observers’ eye movements, reflecting one intrinsic characteristic of visual scenes, termed inter-observer visual congruency (IOVC). However, the neural correlates underlying IOVC is largely unknown especially in free-viewing condition. In this study, we explored the neural correlates of IOVC using functional magnetic resonance imaging (fMRI) and eye-tracking data acquired in a naturalistic paradigm. Specifically, we estimated IOVC of movie shots from eye-tracking data and then conducted two statistical analysis for functional inference, including a hypothesis-driven analyses (general linear model, GLM) and a data-driven approach (inter-subject correlation, ISC). The GLM analysis demonstrated that IOVC recruited two distinctive streams of neural systems. Specifically, neural activities in superior temporal gyrus, default mode network and hippocampus were positively correlated with IOVC, whereas those in the primary and secondary visual cortices, as well as dorsal attention network exhibited negative correlations. Further ISC analysis revealed that movie shots with higher IOVC evoked more synchronous brain activities in the primary auditory cortex, primary visual cortex and superior parietal lobule compared to those shots with lower IOVC. This study provides some novel evidences of visual processing in human brain in free-viewing condition.