Abstract:Recent neuroimaging studies have shown that the visual cortex plays an important role in representing the affective significance of visual input. The origin of these affect-specific visual representations is debated: they are intrinsic to the visual system versus they arise through reentry from frontal emotion processing structures such as the amygdala. We examined this problem by combining convolutional neural network (CNN) models of the human ventral visual cortex pre-trained on ImageNet with two datasets of affective images. Our results show that in all layers of the CNN models, there were artificial neurons that responded consistently and selectively to neutral, pleasant, or unpleasant images and lesioning these neurons by setting their output to zero or enhancing these neurons by increasing their gain led to decreased or increased emotion recognition performance respectively. These results support the idea that the visual system may have the intrinsic ability to represent the affective significance of visual input and suggest that CNNs offer a fruitful platform for testing neuroscientific theories. What is the role played by sensory cortices in assessing the emotional significance of sensory input? This question is attracting increasing research interest. Recent work has found affect-specific neural representations in visual cortex. The origins of these representations are debated. According to the reentry hypothesis, these representations result from reentrant feedback arising from anterior emotion processing structures such as the amygdala. An alternative hypothesis holds that sensory cortex may have the intrinsic capacity to represent the emotional qualities of sensory input. We examined this problem by utilizing the convolutional neural networks (CNNs) trained to recognized visual objects as computational models of the primate ventral visual system. Emotionally charged images were divided into three broad categories (pleasant, neutral and unpleasant) and presented to the CNNs. Responses of artificial neurons to these images were found to exhibit robust emotion selectivity. Importantly, enhancing the neurons that were selective for a given emotion led to the increased ability in recognizing that emotion, whereas lesioning these neurons led to the decrease in that ability. This research lends support to the notion that emotional perception might be an intrinsic property of the visual cortex. It also underscores the CNNs' value in examining neuroscientific theories.

Weakly Supervised Learning of Image Emotion Analysis Based on Cross-spatial Pooling.

A Saliency Detection and Gram Matrix Transform-Based Convolutional Neural Network for Image Emotion Classification.

Joint Weakly Supervised Image Emotion Analysis Based on Interclass Discrimination and Intraclass Correlation

Dependency Exploitation: A Unified CNN-RNN Approach for Visual Emotion Recognition

Weakly Supervised Coupled Networks for Visual Sentiment Analysis

Discovering Affective Regions in Deep Convolutional Neural Networks for Visual Sentiment Prediction.

Learning Multi-level Representations for Image Emotion Recognition in the Deep Convolutional Network

EEG-based Emotion Recognition via Efficient Convolutional Neural Network and Contrastive Learning

An Attention Pooling Based Representation Learning Method for Speech Emotion Recognition

Learning multi-level representations for affective image recognition

Emergence of Emotion Selectivity in Deep Neural Networks Trained to Recognize Visual Objects

Emotion Recognition by Assisted Learning with Convolutional Neural Networks

Chinese Painting Emotion Classification Based on Convolution Neural Network and SVM

Exploring Discriminative Representations for Image Emotion Recognition with CNNs.

Boosting Image Emotion Analysis Via Coupling Dependent Attentive Model.

CNN-Based Broad Learning for Cross-Domain Emotion Classification

Video Emotion Recognition Using Local Enhanced Motion History Image and CNN-RNN Networks.

Multiscale Emotion Representation Learning for Affective Image Recognition

EmoCAM: Toward Understanding What Drives CNN-based Emotion Recognition

Correction To: Deep Transfer Learning for Image Emotion Analysis: Reducing Marginal and Joint Distribution Discrepancies Together