Fig. 4: In silico manipulations of sensory-evoked cortical output.
From: Thalamus enables active dendritic coupling of inputs arriving at different cortical layers
a Simulation examples for the example model in Fig. 3a–c show how in silico manipulations affect sensory-evoked responses with 1 AP (top row) and bursts of 2 (center) or 3 APs (bottom). From left to right: no manipulation (i.e., control), we deprived PTs from any indirect sensory input via TC-driven neurons in vS1 (i.e., responses are evoked solely by direct sensory input via the TC→PT pathway; manipulation 1), we deprived PTs from any direct sensory input via the TC→PT pathway (i.e., responses are evoked solely indirect sensory input via TC-driven neurons in vS1; manipulation 2), we deprived PTs from the active properties of their apical dendrites (i.e., PTs could not generate Ca2+ APs; manipulation 9). b Model consensus from all 20,359 model configurations which exhibit burst responses: both types of bursts require direct sensory input from the thalamus (left), whereas bursts of 3 APs, but not bursts of 2 APs, require the active properties of the apical dendrites (right). Box plots represent medians and 25th to 75th percentiles. Whiskers extend to the full range of the data. c Model consensus from manipulation 10 (i.e., we deprived PT models from direct sensory input at different distances from the soma): both types of bursts require direct sensory input to basal dendrites, whereas bursts of 3 APs, but not bursts of 2 APs, also require direct sensory input to apical dendrites – in particular around the primary BP. Shadings represent the 25th to 75th percentiles, bold lines represent the medians. Source data for panels (b, c) are provided in the Source Data file.
