Figure 5.

Distribution of θ- and δ-burst durations in the dark period are independent of the specific threshold Th used to identify bursts and can be described by unique scaling functions. Distribution of θ- and δ-burst durations in the dark period for different threshold values Th on the ratio R_θδ and window size w = 4 s. a, Distribution of θ burst durations for control rats over a 12 h dark period (pooled data). Distributions evaluated using different Th values consistently follow the same power law behavior, with a cutoff that is controlled by Th. Inset, The data collapse onto a universal function f_θ when we plot P(d)d^α versus Th^ϵd, with α = 2.65 and ϵ = 0.8. b, Rescaled distribution of δ-burst durations for control rats over a 12 h dark period (pooled data). Distributions are rescaled by 〈d_δ〉^η, with 〈d_δ〉 mean δ-burst duration and η = 1.3. After rescaling, distributions collapse onto a unique function f_δ that is well described by a Weibull distribution f(d; λ, β) (thick line), with λ = 0.01 and β = 0.72. Inset, Distributions P_δ for different thresholds (not rescaled). c, Distribution of θ burst durations for VLPO-lesioned rats over a 12 h period (pooled data). Distributions evaluated using different Th values consistently follow the same power law behavior, with a cutoff that is controlled by Th. Data collapse onto a universal function f_θ by plotting P(d)d^α versus Th^ϵd with α = 3.1 and ϵ = 0.8. d, Rescaled distribution of δ burst durations for VLPO-lesioned rats over a 12 h dark period (pooled). Distributions are rescaled by 〈d_δ〉^η, with η = 1.3. After rescaling, the distributions collapse onto a unique function f_δ that is close to a Weibull distribution f(x; λ, β) (thick black line), with λ = 0.05 and β = 0.75. Inset, Distributions P_δ for different thresholds (not rescaled).