Abstract: A novel research paradigm in cognitive neuroscience is the use of movies as stimulus material to investigate the neural mechanisms of cognitive and mental activities in subjects during watching movies. Previous studies have suggested that there are sex-related differences in the relationship between the visual and auditory design of movies and the cognitive and mental activities of viewers. Sex differences determine aesthetic differences, and aesthetic differences have different effects on different links in the whole process of film communication. Although it is well documented that there are obvious differences in the cognitive and mental activities between males and females when watching the same movie, the underlying neural mechanisms remain largely unknown. It is important to investigate the sex-related differences in functional brain activity during movie watching.
With the development of functional brain imaging technology, brain age prediction is considered to be an important indicator of an individual's level of brain aging. In addition, a large number of studies have shown that the gap between predicted and actual age is closely related to the deterioration of human brain function. Predicted brain age has been shown to be greater than actual age in patients with Alzheimer's disease, patients with traumatic brain injury, and patients with psychiatric disorders, such as schizophrenia and major depressive disorder. Comparing the similarities and differences in brain age prediction between males and females is expected to reveal different brain ageing trajectories. In particular, comparing male and female brain age prediction models based on movie viewing would be expected to reveal similarities and differences in the patterns of brain functional networks reflected by males and females in response to audio-visual stimuli from films.
Accordingly, this study used functional magnetic resonance imaging (fMRI) data from 528 subjects (261 males, aged 18~87 years) to investigate sex differences in the functional brain network while watching the classic psychological suspense short film 'Bang! You're Dead', directed by Alfred Hitchcock. Using these data, we obtained functional brain connectivity information during movie viewing, performed individualized brain age prediction for males and females separately, and explored the similarities and differences in brain connectivity between males and females during movie viewing. Specifically, a whole-brain functional connectivity matrix was constructed based on the functional activity of the subjects while watching the film. A machine learning technique was then used to predict the brain age of male and female viewers, and the differences in the functional connectivity driving the predictions were analyzed. We examined the similarities and differences in the overall process of brain aging between males and females through the correlation coefficients between the predicted age and the true age, and examined the similarities and differences in the functional brain networks driving the prediction of brain age between males and females.
Our results show that brain connectivity during film viewing can predict the brain age of both men and women with high accuracy. Specifically, the model prediction accuracy for brain age based on male functional brain imaging data was = .821 (± .007, .866), MAE = 8.240 (±.123) years, whereas the model prediction accuracy for brain age prediction for female data was = .848 (± .004), MAE = 7.978 (±.101) years. When analyzing the connections that drive the predictions, there are clear sex differences in the connections associated with the default mode network, reflecting differences in the integration of endogenous and exogenous information between the males and the females. Specifically, the contribution of functional connectivity in prediction models based on male data only explains 2.7% of functional connectivity contributing in prediction models based on female data. Overall, the study demonstrates significant differences in functional brain connectivity between male and female viewers during film viewing, providing neuroscientific evidence for the study of the diverse cognitive psychological activities of film viewers.

Key words: brain age, sex-related difference, movie viewer, alfred hitchcock, movie neuroscience

摘要： 男女受众在观看同一部电影时具有认知心理活动的差异,但是其潜在的神经机制并不清楚。研究者采用脑功能成像技术,探索了男性和女性（262名男性和266名女性,18~87岁）在观看希区柯克经典心理悬疑短片《砰！你死了》时的脑功能网络差别。研究基于被试在观影时的脑功能活动建立全脑功能连接矩阵,结合机器学习技术预测了男女受众的脑龄,进而研究了驱动脑龄预测的脑功能连接性别差异。结果显示观影时的脑连接虽能显著预测男性和女性脑龄,但驱动预测模型的特征在默认网络存在性别差异,反映了不同性别在内源和外源信息整合方面的差异。综上,研究表明男女受众在观影时存在明显脑功能网络模式的差异,为研究电影受众的认知心理活动的神经机制提供了实证依据。

关键词: 脑龄, 性别差异, 电影受众, 希区柯克, 神经电影学