The Influence of Movie-Induced Brain Dynamics on Brain Age Predictions
Poster No:
894
Submission Type:
Abstract Submission
Authors:
Janice Koi1, Junhong Yu1
Institutions:
1Nanyang Technological University, Singapore, Singapore
First Author:
Co-Author:
Introduction:
Naturalistic movie-fMRI utilize ecologically valid stimuli to evoke a wide range of cognitive and emotional responses in the brain (Jääskeläinen et al., 2021). Brain age, an estimate of an individual's chronological age based on neuroimaging data, is commonly used to study deviations from typical aging trajectories (Cole & Franke, 2017). Despite its utility, the examination of brain age derived from movie-fMRI data remains underexplored (Bi et al., 2024), and it is unknown if dynamic brain states induced by changing movie content influence brain age predictions. Using the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) dataset (N=540, age range:18.5 to 88.9), resting-state fMRI (RS-fMRI) functional connectivity (FC) brain age models were applied to the movie-fMRI FC data to decode brain age during movie watching. To capture dynamic changes, a sliding window approach is applied, allowing for brain age predictions on a window-by-window basis (window size=5TRs, step size=1TR). We further investigate how these dynamic brain age gaps (BAG; chronological age – predicted age) scale with specific movie content and explore the role of participant age in modulating these relationships.
Methods:
Brain age models were trained within RS-fMRI using six machine learning algorithms via 10-fold cross-validation. Model performance was evaluated using Pearson's correlation coefficient (r) and mean absolute error (MAE). These models were then applied to the movie-fMRI data to evaluate its predictive accuracy. The model with the overall highest Pearson's r and lowest MAE was applied to predict brain age dynamically in the movie-fMRI data using the sliding window approach. The obtained BAG values were then z-scored within each participant. Seven participants continuously rated their levels of suspense, emotional arousal, emotional valence, and social interaction among the movie characters on a 1–9 scale while watching the same movie as viewed by the Cam-CAN subjects. Automated movie annotations, including number of faces present onscreen and audio loudness (RMS), were collected. Within-subject correlations between BAG values and each movie annotation were computed across all sliding windows. A one-sample t-test was used to assess whether the correlation coefficients significantly differed from zero. Age effects were examined by correlating participants' correlation coefficients with their chronological age.
Results:
As shown in Figure 1, the elastic net model achieved the highest overall performance (Pearson's r = 0.862, MAE = 8.03) when tested on unseen RS-fMRI data. When applied to movie-fMRI, the model retained strong accuracy (Pearson's r = 0.788, MAE = 10.5). Results revealed significant relationships between BAG values and dynamic states induced by the movie stimulus. Specifically, suspense (t(539) = 4.86, false discovery rate-corrected p-value, FDR-p < 0.0001) and emotional arousal (t(539) = 4.54, FDR-p < 0.0001) showed strong positive associations with BAG, while emotional valence was negatively associated (t(539) = -2.70, FDR-p = 0.0136). Social interaction, number of faces present on-screen and audio loudness did not exhibit significant relationships (FDR-p > 0.05). When exploring age effects, both social interaction (r = 0.149, FDR-p = 0.00153) and the number of faces (r = 0.161, FDR-p < 0.001) exhibited significant positive correlations with age.
Conclusions:
Our findings suggest that emotional states such as suspense, arousal and valence may transiently influence the brain's FC in ways that scale with predicted deviations in brain age. Additionally, social factors, including interactions among the movie characters and the number of faces perceived, exhibited age-dependent effects on the predicted age values, highlighting the role of aging in modulating these relationships. This work underscores the dynamic sensitivity of brain age models to external stimuli and highlights the potential for BAG analyses to provide insights into movie-induced neural processes.
Lifespan Development:
Aging 1
Modeling and Analysis Methods:
Classification and Predictive Modeling 2
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Aging
FUNCTIONAL MRI
Other - Brain Age; Naturalistic movie fMRI
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Provide references using APA citation style.
Cole, J. H., & Franke, K. (2017). Predicting Age Using Neuroimaging: Innovative Brain Ageing Biomarkers. Trends in Neurosciences, 40(12), 681–690. https://doi.org/10.1016/j.tins.2017.10.001
Jääskeläinen, I. P., Sams, M., Glerean, E., & Ahveninen, J. (2021). Movies and narratives as naturalistic stimuli in neuroimaging. NeuroImage, 224, 117445. https://doi.org/10.1016/j.neuroimage.2020.117445