Reviewer #1 (Public review):
Dwulet et al. combined experimental and modeling approaches to investigate how correlated spontaneous activity in the mouse's primary visual (V1) and primary somatosensory (S1) areas drives the development of multisensory integration in area RL. Notably, they focused on early developmental stages, before sensory experience occurs. Consistent with previous experimental findings, the authors first demonstrated that spontaneous activity becomes more sparse across development in all three areas, as measured by event amplitude, event duration, and participation ratio. Using a linear mixed model analysis to compare the maturation of this spontaneous activity, they found evidence that S1 matured the fastest. The authors then presented experimental evidence suggesting that these spontaneous events were moderately correlated both spatially and temporally.
They hypothesized that activity-dependent mechanisms use these correlations to establish connectivity across these regions. To test this hypothesis, the authors modeled a feedforward network with connections from S1 to RL and from V1 to RL, where the strength of connections depended on a Hebbian term for potentiation and a heterosynaptic term for depression. By investigating different levels of V1-S1 correlations, they found that moderate levels of correlation led to the significant development of topographically organized connectivity while maintaining a mix of bimodal and unimodal cells in RL. Additionally, when simulating a network with a more mature S1, they observed that topographical maps improved not only between S1 and RL but also between V1 and RL. Finally, the authors use linear regression to suggest that the mixture of bimodal and unimodal cells in RL is optimal for encoding the maximum amount of information from both V1 and S1.
Comments on revised version:
The revision closes most of the data-model gaps raised in my original review. The authors have clarified the experimental measures and statistical comparisons, improved the spatial correlation-map analysis, added a temporal-lag analysis that argues against stereotyped traveling waves, expanded the model description, and performed additional simulations examining the effects of differences in spontaneous activity. Taken together, these changes provide solid support for the paper's principal conclusion: structured and moderately correlated activity can, within the proposed model, guide the refinement of an initially coarse connectivity scaffold into aligned multisensory representations.
The remaining limitations primarily concern the more specific claim that the somatosensory pathway matures first and guides refinement of the visual pathway. The experiments support the conclusion that spontaneous activity in the somatosensory cortex matures earlier. However, the proposed consequence of this difference is carried in the model by an assumed stronger initial somatosensory-to-higher-order connectivity bias, motivated by pilot anatomical observations that are not included quantitatively in the manuscript. The new supplementary simulations suggest that differences in activity amplitude and frequency alone are insufficient, but the parameters are varied over ranges considerably smaller than the differences measured experimentally, and event duration is not varied. These simulations therefore do not strongly establish that the measured activity differences are insufficient to produce the effect. In addition, the Figure 4 caption and portions of the Discussion continue to imply that more mature somatosensory activity itself instructs map alignment, whereas the revised Results present the more qualified conclusion that an additional connectivity difference is required.
A few internal inconsistencies also remain. The abstract still describes activity in the three areas as being recorded simultaneously, although the cellular-resolution recordings were acquired sequentially; only the wide-field data were collected simultaneously across areas. The revised model also assigns spontaneous events durations and intervals in milliseconds, while the measured calcium events last several seconds and occur only a few times per minute.