- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper introduces BrainDyn, a framework for modeling brain dynamics that integrates dynamic sheaf neural networks with Neural ODEs. The method employs LSTMs to encode the historical activity of brain regions, models heterogeneous interactions between regions via learnable restriction maps, and combines temporal embeddings with Neural ODEs to facilitate brain activity prediction and autoregressive generation. The authors conducted experiments using fMRI, EEG, and NEST simulation data, and further investigated the model's ability to predict neural activity following perturbations.
Critical strengths
- Combining dynamic sheaf structures with Neural ODEs offers a novel approach to modeling heterogeneous inter-regional brain interactions and continuous-time dynamics.
- Experiments were conducted using fMRI, EEG, and simulated neural activity data, evaluating performance across short-term prediction, long-term autoregressive generation, and perturbation scenarios.
- The authors analyzed variations in restriction maps and their singular value spectra, attempting to link the features learned by the model to information interactions between brain regions.
Critical weaknesses
- The technical contributions of the core method relative to existing Sheaf Neural Networks, graph attention mechanisms, and Neural ODEs have not yet been fully elucidated.
- The performance gains observed in some experiments—particularly regarding long-term prediction and perturbation scenarios—are limited and insufficient to fully demonstrate the model's generalizability.
- The analysis of restriction maps relies primarily on mathematical properties, lacking sufficient evidence to support a biological interpretation.
Questions for authors
- The method computes restriction maps based on node features at the end of the historical window and fixes the representations—post-sheaf message passing—throughout the entire prediction horizon, whereas the Neural ODE evolves solely via time embeddings and the current predicted state. This raises a critical question: to what extent do the dynamic interactions learned by the model genuinely evolve with the neural state, rather than being merely static conditions determined by the initial context?
- The authors present prediction results following a silencing perturbation of a single neuron using NEST-simulated data; however, the training data consists solely of unperturbed activity, and the prediction context already includes the initial response following the perturbation. Consequently, this experiment primarily tests the model's ability to predict subsequent trajectories given the early response to a perturbation, rather than fully demonstrating its capacity to predict the causal effects of unknown perturbations based on intervention information.
- This paper integrates dynamic restriction maps, the Sheaf Laplacian, and Neural ODEs; notably, the restriction maps are generated by an MLP based on neighboring node features, establishing a strong mechanistic link to existing graph attention mechanisms and learnable sheaf networks. Current experiments—primarily involving ablation studies that remove the sheaf structure, LSTM, or temporal embeddings—do not yet conclusively demonstrate whether the performance gains stem from the sheaf's geometric modeling mechanism or from the increased parameter count and more flexible feature transformations.
If the author addresses the questions raised above, I will raise my score.
5 comments
Response to Reviewer fxHA (2/3)
Biological interpretation of restriction maps [W3]. The reviewer noted that our analysis of restriction maps relies primarily on mathematical properties. As pointed out by the reviewer, Figure 4 describes the mathematical properties of the maps. Beyond their mathematical properties, the learned maps are approximately low-rank (as seen in Figure 4), so each map transmits only a few directions of a region’s representation and filters out the rest. This parallels evidence that cortical areas communicate through a low-dimensional subspace (Semedo et al., 2019) as highlighted in the manuscript. Figure 3 shows the map norms vary across edges and over time in EEG, consistent with dynamically modulated communication between regions. We extend this analysis to fMRI (Figures 7 and 8). Each of the 400 Schaefer parcels is assigned the mean Frobenius norm of its outgoing restriction maps. These values are projected onto a 3D glass brain and shown from four orientations. Figure 7 shows this for highlighted windows of the fMRI signal, and Figure 8 stacks the regional norms across 5 timepoints. The fMRI results agree with the EEG results. Map norms are spatially heterogeneous across regions, and the high- and low-norm zones shift between timepoints. The model therefore learns time-varying, region-specific communication strengths in a second modality, at higher spatial resolution than in EEG.
Magnitude of performance gains in long-term prediction and perturbation scenarios [W2]. The reviewer noted that the performance gains in long-term prediction and perturbation scenarios are limited and may be insufficient to demonstrate the model’s generalizability. We agree that some of the performance gains are small in these settings. We attribute these to the difficulty of the underlying problem rather than a lack of generality. Autoregressive generation is a particularly hard problem; we note that several other baselines had an MSE greater than one (worse than predicting the mean). In this setting, even small relative improvements are significant. Additionally, BrainDyn achieves the best MSE and MAE across every modality in the long-horizon setting, and the best DTW on fMRI and NEST. We believe this is evidence of superior generality. Where performance gains are particularly modest, this is a valuable result to inform future work and application of the method. For example, a valuable direction for future work is exploring why performance gains for long-term performance gains are more modest for EEG compared to fMRI. In particular, the interpretability of BrainDyn and its biological inspiration mean that these results, combined with ablation experiments, provide mechanistic insight into the important features for different data modalities.
Response to Reviewer fxHA (3/3)
Contribution of the sheaf vs. parameter count [Q3]. The reviewer asked whether the gains stem from the sheaf mechanism or from increased parameters and more flexible feature transformations. We note two additional sets of experiments:
- Capacity-matched controls. We implemented two capacity-matched controls to better establish the contribution of edge-specific sheaf geometry to both short- and long-term accuracy. First, we implemented a capacity-matched graph convolution network (GCN) layer to replace sheaf message passing. In order to ensure the GCN has the same capacity as sheaf message passing we choose an appropriately-sized hidden dimension W and updated each node’s LSTM embedding as , where and . Ultimately, the capacity-matched GCN had eight more parameters than BrainDyn on fMRI data and 14 more parameters than BrainDyn on EEG data. The capacity-matched GCN performed better than the Graph ablation but worse than BrainDyn across both modalities and every metric (see Table 1). On short-horizon forecasting for the fMRI data, the GCN control had 12x worse MSE. We also note that BrainDyn outperformed an ablation which removed the sheaf structure while maintaining the same number of parameters by adding additional capacity to the neural ODE (see Tables 2 and 3).
- Static-sheaf control. In addition to providing geometric structure, the restriction maps are learned and time-varying. In order to distinguish the impact of time-varying restriction maps we implemented a static sheaf ablation which kept the same sheaf structure and MLP to generate the restriction maps but did not pass node stalks into the MLP. Instead, the MLP was only given the positional embeddings of the nodes, yielding restriction maps which are identical in every window. The static sheaves are competitive with the BrainDyn implementation on the short time horizon; however, they suffer during long-horizon autoregressive generation (see Table 4). On short horizons, the forecast is anchored to observed context which is encoded in the node stalks. Thus, static restriction maps recover most of the benefits of sheaf geometry. However, during autoregressive generation, it is important for regional communication to be able to adapt as the underlying dynamics evolve.
Table 1: Capacity-matched GCN ablation
| Horizon | Method | fMRI MSE ↓ | fMRI MAE ↓ | fMRI DTW ↓ | EEG MSE ↓ | EEG MAE ↓ | EEG DTW ↓ |
|---|---|---|---|---|---|---|---|
| Short | BrainDyn | 0.0531 ± 0.0091 | 0.1238 ± 0.0121 | 0.0599 ± 0.0054 | 0.4001 ± 0.0253 | 0.4209 ± 0.0126 | 0.1905 ± 0.0292 |
| Short | BrainDyn Capacity-matched GCN | 0.6562 ± 0.0420 | 0.5686 ± 0.0156 | 0.2488 ± 0.0076 | 0.4145 ± 0.0658 | 0.4320 ± 0.0337 | 0.1971 ± 0.0146 |
| Long | BrainDyn | 0.6374 ± 0.0465 | 0.5356 ± 0.0157 | 0.2654 ± 0.0084 | 0.7199 ± 0.0292 | 0.6005 ± 0.0161 | 0.3008 ± 0.0080 |
| Long | BrainDyn Capacity-matched GCN | 0.8806 ± 0.0558 | 0.6687 ± 0.0172 | 0.3360 ± 0.0090 | 0.7522 ± 0.1443 | 0.6174 ± 0.0571 | 0.3112 ± 0.0282 |
Table 2: Sheaf ablation (parameter-matched in Neural ODE) on fMRI and EEG data
| Data | Method | MSE ↓ | MAE ↓ | DTW ↓ |
|---|---|---|---|---|
| fMRI | BrainDyn | 0.0531 ± 0.0091 | 0.1238 ± 0.0121 | 0.0599 ± 0.0054 |
| fMRI | BrainDyn No sheaf (parameter-matched) | 0.4860 ± 0.0354 | 0.4889 ± 0.0147 | 0.2121 ± 0.0076 |
| EEG | BrainDyn | 0.4001 ± 0.0253 | 0.4209 ± 0.0126 | 0.1905 ± 0.0292 |
| EEG | BrainDyn No sheaf (parameter-matched) | 0.4357 ± 0.0589 | 0.4446 ± 0.0265 | 0.2029 ± 0.0116 |
Table 3: Static sheaf ablation
| Horizon | Method | fMRI MSE ↓ | fMRI MAE ↓ | fMRI DTW ↓ | EEG MSE ↓ | EEG MAE ↓ | EEG DTW ↓ |
|---|---|---|---|---|---|---|---|
| Short | BrainDyn | 0.0531 ± 0.0091 | 0.1238 ± 0.0121 | 0.0599 ± 0.0054 | 0.4001 ± 0.0253 | 0.4209 ± 0.0126 | 0.1905 ± 0.0292 |
| Short | BrainDyn Static sheaf | 0.0546 ± 0.0091 | 0.1236 ± 0.0129 | 0.0598 ± 0.0057 | 0.3942 ± 0.0548 | 0.4185 ± 0.0294 | 0.1907 ± 0.0131 |
| Long | BrainDyn | 0.6374 ± 0.0465 | 0.5356 ± 0.0157 | 0.2654 ± 0.0084 | 0.7199 ± 0.0292 | 0.6005 ± 0.0161 | 0.3008 ± 0.0080 |
| Long | BrainDyn Static sheaf | 0.7150 ± 0.0740 | 0.5789 ± 0.0335 | 0.2884 ± 0.0161 | 0.7351 ± 0.1444 | 0.6108 ± 0.0593 | 0.3072 ± 0.0293 |
The results have been added to Appendix C in our updated manuscript.
We thank the reviewer again for their feedback. We hope our responses address the reviewer’s concerns, and if so, we would be grateful if the reviewer would consider raising their score. We are happy to clarify anything further.
Response to Reviewer fxHA (1/3)
We thank the reviewer for their careful assessment of our paper. We appreciate the reviewer’s recognition that combining dynamic sheaf structures with neural ODEs offers a novel approach to modeling heterogeneous inter-regional brain interactions and continuous-time dynamics. We also appreciate that the reviewer found our evaluation across fMRI, EEG, and simulated neural activity valuable, spanning short-term prediction, long-term autoregressive generation, and perturbation scenarios, and that they noted our analysis of restriction maps and their singular value spectra in linking learned features to interactions between brain regions. We address the concerns as follows:
Technical contributions [W1]. The reviewer noted that our contributions relative to sheaf neural networks, graph attention, and neural ODEs were not fully elucidated. Relative to graph attention, which scales each neighbor’s message by a single scalar, BrainDyn learns a matrix for each edge that can rotate, stretch, and project a region’s representation. Each edge has asymmetric maps for its two endpoints, and the maps are approximately low-rank (Fig. 4), so each edge transmits a selected subspace of information. Consistent with this, BrainDyn outperforms a capacity-matched GCN on every metric. Relative to sheaf neural networks, whose maps are computed from static node features, BrainDyn computes the maps from each region’s temporal history and learned identity, and recomputes them at every forecast window. A static-sheaf control degrades in long-horizon generation. Relative to neural ODEs, whose node features evolve in a single shared latent space, BrainDyn conditions the vector field on edge-specific sheaf messages and a learned-frequency time embedding. To our knowledge, sheaf message passing has not previously been used to condition a continuous-time ODE.
Dynamic interactions [Q1]. The reviewer asked whether the learned interactions evolve with the neural state or are fixed by the initial context. Within a single forecast window, the restriction maps are fixed, and the ODE evolves through the time embedding and predicted state. Across windows, however, the maps are recomputed from the most recent context. In long-horizon generation (autoregression), that context consists increasingly of the model's own predictions, so the interactions evolve with the predicted neural state rather than staying fixed by the initial context.
NEST perturbation [Q2]. The reviewer pointed out that the prediction context already includes the early response to the perturbation. We agree with this reading of the experiment. To test whether performance depends on observing the early response, we reran the evaluation with the context window ending exactly at perturbation onset (0% post-onset), so the model receives no perturbed signal:
Table 1: NEST perturbation experiment table
BrainDyn remains the best model on all metrics (MSE 0.6655 vs. 0.7316 for ODEBRAIN; PCC 0.653 vs. 0.607), with performance nearly identical to the 10% setting reported in the paper (MSE 0.6649). This shows that BrainDyn’s advantage does not depend on observing the early post-perturbation response. We have added the above results to Appendix J in our updated manuscript.