Figure 5. Baseline rectification of the temporal envelope leads to clean CSD bands in virtual LFPs.

The temporal activation of the epochs analyzed was rectified by a constant amount and polarity, as determined experimentally (A). This process enabled the lost DC components to be restored at each recording site by subsequently reconstructing virtual LFPs. Since the corrected LFPs are proportional to the underlying current sinks, either one can be used for quantitative estimation of the individual wave parameters. CSD analysis of rectified virtual LFPs in the experiment and the model (B) renders clean bands of sinks at the synaptic domain of the Schaffer input and surrounding sources in the st. pyr and l-m (the epoch corresponds to the same segments as in Figure 3). C shows the estimation of the amplitude and duration of individual gamma waves before (1) and after (2) correction in long experimental (600 s) and model epochs (450 s). The distribution of amplitudes in raw experimental and model LFPs revealed absolute positive and negative values for the individual waves due to the ongoing offset in raw LFPs, which turned their absolute values to biphasic or pure positive values. Following correction, both the absolute amplitude became solely negative and they increased. The densitograms below show the distribution of the amplitude/duration pairs accumulated from individual waves in representative experiments and simulations. Note that after correction, the cloud is not a simple transposition in the two axes, as individual dots (waves) may have required different degrees of correction.