Robust inter-subject audiovisual decoding in functional magnetic resonance imaging using high-dimensional regression

Neocortical substrates of feelings evoked with music in the ACC, insula, and somatosensory cortex

Neurobiological models of emotion focus traditionally on limbic/paralimbic regions as neural substrates of emotion generation, and insular cortex (in conjunction with isocortical anterior cingulate cortex, ACC) as the neural substrate of feelings. An emerging view, however, highlights the importance of isocortical regions beyond insula and ACC for the subjective feeling of emotions. We used music to evoke feelings of joy and fear, and multivariate pattern analysis (MVPA) to decode representations of feeling states in functional magnetic resonance (fMRI) data of n = 24 participants. Most of the brain regions providing information about feeling representations were neocortical regions. These included, in addition to granular insula and cingulate cortex, primary and secondary somatosensory cortex, premotor cortex, frontal operculum, and auditory cortex. The multivoxel activity patterns corresponding to feeling representations emerged within a few seconds, gained in strength with increasing stimulus duration, and replicated results of a hypothesis-generating decoding analysis from an independent experiment. Our results indicate that several neocortical regions (including insula, cingulate, somatosensory and premotor cortices) are important for the generation and modulation of feeling states. We propose that secondary somatosensory cortex, which covers the parietal operculum and encroaches on the posterior insula, is of particular importance for the encoding of emotion percepts, i.e., preverbal representations of subjective feeling.

Deep neural networks and kernel regression achieve comparable accuracies for functional connectivity prediction of behavior and demographics

There is significant interest in the development and application of deep neural networks (DNNs) to neuroimaging data. A growing literature suggests that DNNs outperform their classical counterparts in a variety of neuroimaging applications, yet there are few direct comparisons of relative utility. Here, we compared the performance of three DNN architectures and a classical machine learning algorithm (kernel regression) in predicting individual phenotypes from whole-brain resting-state functional connectivity (RSFC) patterns. One of the DNNs was a generic fully-connected feedforward neural network, while the other two DNNs were recently published approaches specifically designed to exploit the structure of connectome data. By using a combined sample of almost 10,000 participants from the Human Connectome Project (HCP) and UK Biobank, we showed that the three DNNs and kernel regression achieved similar performance across a wide range of behavioral and demographic measures. Furthermore, the generic feedforward neural network exhibited similar performance to the two state-of-the-art connectome-specific DNNs. When predicting fluid intelligence in the UK Biobank, performance of all algorithms dramatically improved when sample size increased from 100 to 1000 subjects. Improvement was smaller, but still significant, when sample size increased from 1000 to 5000 subjects. Importantly, kernel regression was competitive across all sample sizes. Overall, our study suggests that kernel regression is as effective as DNNs for RSFC-based behavioral prediction, while incurring significantly lower computational costs. Therefore, kernel regression might serve as a useful baseline algorithm for future studies.

Transferring and generalizing deep-learning-based neural encoding models across subjects

Recent studies have shown the value of using deep learning models for mapping and characterizing how the brain represents and organizes information for natural vision. However, modeling the relationship between deep learning models and the brain (or encoding models), requires measuring cortical responses to large and diverse sets of natural visual stimuli from single subjects. This requirement limits prior studies to few subjects, making it difficult to generalize findings across subjects or for a population. In this study, we developed new methods to transfer and generalize encoding models across subjects. To train encoding models specific to a target subject, the models trained for other subjects were used as the prior models and were refined efficiently using Bayesian inference with a limited amount of data from the target subject. To train encoding models for a population, the models were progressively trained and updated with incremental data from different subjects. For the proof of principle, we applied these methods to functional magnetic resonance imaging (fMRI) data from three subjects watching tens of hours of naturalistic videos, while a deep residual neural network driven by image recognition was used to model visual cortical processing. Results demonstrate that the methods developed herein provide an efficient and effective strategy to establish both subject-specific and population-wide predictive models of cortical representations of high-dimensional and hierarchical visual features.