Non-artifactual NREM sleep EEG and thalamogram records were subjected to the Individual Adjustment Method (IAM) of sleep spindle analysis (Bódizs et al. 2009; Ujma et al. 2015). Frontally dominant slow and parietally dominant fast sleep spindles were defined for each patient and each EEG and thalamic channel using individual-specific frequency criteria, as well as associated individual- and derivation-specific amplitude criteria in the 9–16 Hz all-night NREM sleep. Thalamic sleep spindles contaminated with assumed IEDs (see below) were removed from further analyses, as IEDs with increased activity in the spindle frequency range may be falsely detected as spindles. The remained sleep spindles were categorized based on their association with or without ripples. If one or more ripple was detected during the spindle, that spindle was categorized as ripple-associated spindle, SP(ripple). Spindles without any association with ripples were categorized as pure sleep spindles, SP(pure). The spindle densities (number per minutes) were calculated separately for each category. Sleep spindle density data were averaged across appropriate derivations within a nucleus, separately for the ANT and MD, and for the fast and slow spindles. The duration of sleep spindles was measured as the time interval from spindle onset and offset, separately for each spindle category. Spindle duration data from all derivations corresponding to same nucleus were averaged, separately for the ANT and MD, and for the fast and slow spindles.