aCLAS

This project contains data and scripts for the paper:

A closed-loop auditory stimulation approach selectively modulates alpha oscillations and sleep onset dynamics in humans

Data to produce figures 1-6 is contained in \Data

Outlined below is the organisation of the data structure for each figure

Figure 1

clusters
The channel numbers for the clusters of interest in experiment 1 and 2
[experiment x channel]

num_participants
number of participants for experiments 1 and 2

ISI
The average inter-stimulus interval for each participant in experiments 1 and 2

power_off
The average power spectrum, taken from the phase-locking electrode, for each participants in experiments 1 and 2
[participant x frequency]

ecHT_angle
The average phase angle (radians) at stimulus onset for each participant and condition in experiments 1 and 2. Taken from the phase-locking electrode [participant x condition]

fz_angle
The average phase angle (radians) at stimulus onset for each high-density EEG channel, participant, and condition in experiments 1 and 2. Taken from the high-density EEG net [channel x participant x condition]

fz_R
The resultant of stimulus onset phases for each high-density EEG channel, participant, and condition in experiments 1 and 2. Taken from the high-density EEG net [channel x participant x condition]

ecHT_R
The resultant of stimulus onset phases for each participant, and condition in experiments 1 and 2. Taken from the phase-locking electrode [participant x condition]

frequency_off
The average instantaneous alpha frequency (Hz) taken from the 'off' periods for each participant, for experiments 1 and 2. Taken from the phase-locking electrode.

f
The frequencies to which the aforementioned power spectra correspond

layout
High density EEG channel layout for topoplots

neighbours High density EEG channel neighbours

Figure 2

f
The frequencies to which the power spectra correspond

frequency_change
The frequency change (Hz) from 'off' to 'on' periods (i.e., stimulus induced changes) for each channel, participant, and condition for experiments 1 and 2 [channel x participant x condition]

num_participants
participant numbers for experiments 1 and 2

cluster_power_change
The change in power (relative to 'off' periods, log ratio)for each condition, frequency, second of stimulation, and participant. Values averaged over channels in clusters of interest for experiments 1 and 2.

average_power_change
The average change in power (9.8-10.2 Hz) (relative to 'off' periods, log ratio) for each channel, condition, frequency, and participant. Values averaged over the stimulation period, for experiments 1 and 2.

cluster_spectrum The average change in power (relative to 'off' periods, log ratio) for each condition, frequency, and participant. Values averaged over channels in clusters of interest, and across the stimulation period, for experiments 1 and 2.

Figure 3_

ERP
Broadband ERP amplitudes for each participant, timepoint, and condition, in experiments 3 and 4. Taken from high-density EEG channel of interest in each experiment, Fz and Pz respectively. Timepoints are -1 second to +1 second, sampling rate is 500 Hz. [participant x time x condition]

ecHT_real Alpha band ERP amplitudes for each participant, timepoint, and condition, in experiments 3 and 4. Taken from high-density EEG channel of interest in each experiment, Fz and Pz respectively. Timepoints are -1 second to +1 second, sampling rate is 500 Hz. [participant x time x condition]

ecHT_phase
Alpha band phase for each participant, timepoint, and condition, in experiments 3 and 4. Taken from high-density EEG channel of interest in each experiment, Fz and Pz respectively. Timepoints are -1 second to +1 second, sampling rate is 500 Hz. [participant x time x condition]

ecHT_r
Alpha band phase resultant for each participant, timepoint, and condition, in experiments 3 and 4. Taken from high-density EEG channel of interest in each experiment, Fz and Pz respectively. Timepoints are -1 second to +1 second, sampling rate is 500 Hz. [participant x time x condition]

octile_z
Maximum z-scores (i.e., measure of non-uniformity) evoked by stimuli in 8 octiles of alpha power, going from low to high. For experiments 3 and 4. [octile x experiment]

num_participants
Participant numbers for experiments 3 and 4.

Figure 4

phase_after
Alpha band phase (radians) for each participant, timepoint (0-1004 ms), and phase bin, in experiments 3 and 4. Taken from high-density EEG channel of interest in each experiment, Fz and Pz respectively. Sampling rate is 500 Hz. Phase bins are 36 degrees in width, starting at 0 and proceeding around the circle. [participant x time x phase bin]

phase_change
Same as above, but for difference between actual phase and expected phase, assuming a 10 Hz oscillation.

R_after
Same as above, but for phase resultant.

num_participants
Participant numbers for experiments 3 and 4.

Figure 5

power
The change in power (relative to 'off' periods, log ratio)for each stimulation condition (conditions are pre-peak stimulation, pre-trough stimulation) , frequency, second of the nap (up to 30.6 minutes), and participant for the Fz channel for experiment 5. Frequencies are 2:.1:30 Hz. [frequency x second x participant]

frequency_line
Instantaneous alpha frequency at Fz for each stimulation condition (conditions are pre-peak stimulation, pre-trough stimulation) in 10-second bins across the nap. Data are sham-subtracted. [participant x time bin x condition]

frequency_violins
Instantaneous alpha frequency at Fz for each stimulation condition (conditions are pre-peak stimulation, pre-trough stimulation) averaged across two periods of the nap (stimulation and post-stimulation, respectively). Data are sham-subtracted. [participant x condition x period]

Figure 6

The data are contained herein are for plotting either line plots or violin plots of various sleep staging and EEG features for experiment 5

The line plots are organised as follows: [condition x 30-second time bin x feature], (conditions are pre-peak stimulation, pre-trough stimulation, sham) and correspond to the following features:
(1) Number of participants in any sleep stage (i.e., not wake) (%)
(2) Average sleep stage (i.e., wake is assigned 0 and N1-N3 are assigned 1-3). Averaged over participants.
(3) Cumulative sum of minutes of N2+ sleep (i,e., epochs scored as N2 or N3), averaged over participants
(4) Participants having had at least 30 seconds of sleep (any stage) (%)
(5) Participants having had at least 120 seconds of sleep (any stage) (%)
(6) Participants having had at least 30 seconds of sleep (N2 and greater) (%)
(7) Participants having had at least 120 seconds of sleep (N2 and greater) (%)
(8) Abundance of alpha oscillations, averaged over frontal channels (F3 and F4) (9) Abundance of sigma oscillations, averaged over frontal channels (F3 and F4) (10) Offset of aperiodic component of power spectrum, averaged over frontal channels (F3 and F4) (11) Exponent of aperiodic component of power spectrum, averaged over frontal channels (F3 and F4). More negative values indicate a steeper slope.

The violin plot data are organised as follows: [participant x condition x feature], (conditions are pre-peak stimulation, pre-trough stimulation, sham) and correspond to the following features:
(1) Number of epochs of any sleep stage (i.e., not wake) during stimulation period. (2) Number of epochs of any sleep stage (i.e., not wake) during post-stimulation period. (3) Average sleep stage (i.e., wake is assigned 0 and N1-N3 are assigned 1-3) for stimulation period. (4) Average sleep stage (i.e., wake is assigned 0 and N1-N3 are assigned 1-3) for post-stimulation period. (5) Number of epochs of N2+ sleep (i,e., epochs scored as N2 or N3) for stimulation perod. (6) Number of epochs of N2+ sleep (i,e., epochs scored as N2 or N3) for post-stimulation perod. (7) Latency to begining of first 30 seconds of sleep (any stage). Data are log-transformed minutes (8) Latency to begining of first 120 seconds of sleep (any stage). Data are log-transformed minutes (9) Latency to begining of first 30 seconds of sleep (N2 or greater). Data are log-transformed minutes (10) Latency to begining of first 120 seconds of sleep (N2 or greater). Data are log-transformed minutes
(11) Abundance of alpha oscillations, averaged over frontal channels (F3 and F4) and stimulation period (12) Abundance of alpha oscillations, averaged over frontal channels (F3 and F4) and post-stimulation period (13) Abundance of sigma oscillations, averaged over frontal channels (F3 and F4) and stimulation period (14) Abundance of sigma oscillations, averaged over frontal channels (F3 and F4) and post-stimulation period (15) Offset of aperiodic component of power spectrum, averaged over frontal channels (F3 and F4), averaged over stimulation period
(16) Offset of aperiodic component of power spectrum, averaged over frontal channels (F3 and F4), averaged over post-stimulation period
(17) Exponent of aperiodic component of power spectrum, averaged over frontal channels (F3 and F4), averaged over stimulation period
(18) Exponent of aperiodic component of power spectrum, averaged over frontal channels (F3 and F4), averaged over post-stimulation period

line_plot_names
Labels for the y axis of line plots

num_participants
participant numbers for experiment 5