The study team induced focal micro-injury in the frontal cortex of awake rats and recorded cortical electrical activity across homotopic somatosensory and retrosplenial sites. Investigators reported that the injury reliably produced a propagating wave of spreading depolarization (SD) that traveled across the ipsilesional cortex. The SD waves did not invade contralesional cortex in these experiments. Recordings used local field potentials (0–45 Hz) sampled at 1 kHz; SD detection relied on characteristic high-amplitude negative DC potential shifts captured in the slow trend component of the signal.
Mutual information analysis indicated a two-phase response. During the SD event, the cohort showed increased synchrony within the injured hemisphere — a short-lived hypercoupling. After the SD terminated, a longer-lasting decrease in interhemispheric coupling emerged. The study team computed mutual information in sequential 20-second nonoverlapping epochs over 400 seconds following SD onset and compared those values to baseline epochs drawn from an earlier control window (800–1100 s).
Signal preprocessing separated a slow trend (which included the SD-associated DC shift) from the detrended fast component used for connectivity measures. Artifact-contaminated recordings were either corrected when possible or excluded; one animal was removed from analysis for incorrect cannula placement, leaving six artifact-free LFP recordings for connectivity reconstruction.
To assess directionality, the authors used transfer entropy as an information-theory measure of effective connectivity. Data indicated that, during the post-SD hypoconnectivity period, the influence exerted by the uninjured hemisphere onto the injured hemisphere increased for both somatosensory and retrosplenial sites. Transfer entropy was computed in both directions for each channel pair; the analysis thus provided two directional values per pair and showed an asymmetric change after SD.
This directional reweighting — larger driving input from the intact cortex to the injured cortex after SD — is consistent with the concept of connectional diaschisis, where focal damage reorganizes network influence patterns at sites distant from the lesion.
The retrosplenial cortex, described by the authors as a key hub of the default-mode network, presented distinct dynamics compared with the somatosensory cortex. At baseline the retrosplenial recordings showed strong interhemispheric coupling. After SD, the retrosplenial cortex exhibited complex connectivity changes rather than a simple increase or decrease in synchrony. The study team noted that because the retrosplenial area is engaged in transmodal, internally directed processing, SD-driven perturbations here could plausibly influence cognitive network function after focal injury.
Rats (male Wistar) were implanted with guide cannulas for microinjury and with LFP electrodes in homotopic retrosplenial and somatosensory locations. The authors reported habituation, baseline recording, then sequential micro-injury of symmetric frontal sites with 15 minutes between hemispheric injuries. Microinjury was delivered by advancing a needle through the implanted cannula; animals remained awake and mobile during procedures.
For functional connectivity the team used a nearest-neighbor estimator of mutual information (k = 1) applied in moving epochs; for effective connectivity they used a nearest-neighbor implementation of transfer entropy, conceptually aligned with nonlinear Granger causality. Both measures were applied across intrahemispheric and interhemispheric channel pairs to map strength and direction of information flow.
The authors made raw data available on Zenodo and documented funding and ethical approvals. However, the source text contains placeholder gaps for some stereotaxic coordinates (AP/ML values are not fully reported), and it does not provide behavioral or cognitive outcome measures tied to the observed connectivity changes. The analysis is based on six artifact-free recordings (one of seven implanted rats excluded), which limits generalizability. The source did not report pharmacologic interventions to modulate SD, nor did it present longitudinal recovery data linking these connectivity shifts to persistent cognitive deficits. These items remain open for future experimental clarification.