Statistics
As mentioned above, all statistics were performed with Statistica 14. The effect of the trauma on the hearing thresholds obtained with the ABR measurements was assessed using the calculated (Schilling et al. , 2019) individual and frequency specific hearing loss (HL, post threshold - pre threshold) with a two-factorial ANOVA (factor ear andfrequency ).
For the evaluation of the GPIAS results, the means of the effect size of the individual log-normalized gap / no-gap amplitude responses of the different background noise frequencies were analyzed for each animal and each frequency (Schilling et al. , 2017). Significant negative effect sizes (Students t-tests) at a given frequency indicate a possible tinnitus percept and an animal with such indication was classified as a tinnitus animal (T group), if no such behavioral sign of tinnitus was found, the animal was classified as non-tinnitus animal (NT group). All animals that received sham-trauma showed no signs of tinnitus and were assigned to the control group C.
The luminance ratio of the ECM of the cortical areas was analyzed using a two-factorial ANOVA (factor side (ipsi- / contralateral to trauma) and animal group (T, NT, C)). Additionally we investigated the luminance ratios of the two hemispheres separately by one-factorial ANOVAs (with Bonferroni correction for analysis repetition) with the factor group only. This was done to find possible side-specific effects that were obscured due to the variance in the comparably few data points of the overall 12 animals.