Fig. 2 - Supplemental 4
- ID
- ZDB-FIG-250722-66
- Publication
- Wang et al., 2025 - The geometry and dimensionality of brain-wide activity
- Other Figures
-
- Fig. 1
- Fig. 2
- Fig. 2 - Supplemental 1
- Fig. 2 - Supplemental 2
- Fig. 2 - Supplemental 3
- Fig. 2 - Supplemental 4
- Fig. 3
- Fig. 3 - Supplemental 1
- Fig. 3 - Supplemental 2
- Fig. 4
- Fig. 4 - Supplemental 1
- Fig. 4 - Supplemental 2
- Fig. 5
- Fig. 5 - Supplemental 1
- Fig. 5 - Supplemental 2
- Fig. 5 - Supplemental 3
- Fig. 5 - Supplemental 4
- Fig. 5 - Supplemental 5
- Fig. 5 - Supplemental 6
- Fig. 5 - Supplemental 7
- Fig. Appendix Figure 1
- Fig. Appendix Figure 1 (2)
- Fig. Appendix Figure 2
- Fig. Appendix Figure 2 (2)
- Fig. Appendix Figure 3
- Fig. Appendix Figure 3 (2)
- Fig. Appendix Figure 4
- Fig. Appendix Figure 5
- Fig. Appendix Figure 6
- Fig. Appendix Figure 7
- Fig. Appendix Figure 8
- Fig. Appendix Figure 9
- All Figure Page
- Back to All Figure Page
Scale-invariant properties persist across different temporal sampling rates in neural recordings. Analysis of multi-area Neuropixels recordings (Stringer et al., 2019b) from 1024 neurons, downsampled to different rates resulting in 7200 time frames per condition (6, 12, 18, and 24 Hz; columns 1–4, respectively). (A–D) Distribution of pairwise covariances after normalization to unit variance (E(σ2i)=1 , see Methods). (E–H) Eigenvalue spectra of the covariance matrices, showing similar power-law scaling across sampling rates. (I–L) Probability density functions (PDFs) of the eigenvalues, demonstrating that the characteristic shape of the distribution is preserved across different temporal resolutions. |