Fig. 9.

Time-frequency analysis reveals significant changes in spectrogram after injury. (a) Representative time-frequency spectrograms from a single animal showing the alterations in spectrotemporal content following CCI. The white arrows indicate the time point at which forelimb stimulation was applied. (b) Spectrotemporal changes over time, averaged over all animals. Each image depicts, in grayscale, the difference in spectrotemporal content via subtraction of the preinjury spectrogram (darker shades indicate decreases, brighter shades indicate increases). Superimposed in color on the grayscale difference spectrogram are points that, when analyzed over all experiments, demonstrated a difference from the preinjury values with $P<0.05$ using a two-tailed student’s $t$-test. The colors are defined as in the colorbar of (a), and note that nearly all of the spectrotemporal power changes are focal increases. (c) Comparison of a representative postinjury $\mathrm{\Delta }\mathrm{CBF}$ waveform (here, at a latency of 25 min following injury) with a plot showing the sum, over all experiments, of pixels showing alterations from baseline with $P<0.05$. Note that the significantly changed spectrotemporal features lined up with the cycles of apparent “oscillations” following injury. The gray shading projected onto the postinjury $\mathrm{\Delta }\mathrm{CBF}$ waveform (a representative plot) highlights temporal periods that contained statistically significant alterations in spectral power, assessed over all experiments ($n=10$).