Using EEG for recording Sleep Elaine Tham
1. What is sleep polysomnography or sleep EEG? 2. Sleep stages/scoring 3. Relating sleep EEG with behavioural data
1. Sleep Polysomnography
THE REST (Head) EYES = EOG CHIN = EMG
30 seconds = 1 epoch 1 second Electrodes are bilateral Odd = left side (e.g. C3), Even = right side (e.g. C4)
FREQUENCY AMPLITUDE Hz = Cycles Per second μv = Microvolts SLOW WAVE SLEEP N3
2. Sleep stages
Sleep occurs in natural cycles of about 90 minutes This is based on a normal overnight participant More SWS early in the night. More REM later. Nap studies may differ more REM if the nap is early morning More SWS if later in the day
WAKEFULNESS Mixed frequency Alpha particularly in the Occipital (O1/O2) regions but NOT in all participants May have movement and blink artefacts. Natural, brief awakenings throughout the night.
Stage 1 (shallow sleep) EEG: EOG: EMG: - the absence of alpha activity - vertex sharp waves - slow eye movement - relatively low amplitude
Stage 2 EEG: EOG: - sleep spindles (oscillating with frequency between 12-15 Hz) strongest in central electrodes - K-complexes (high voltage, sharp rising and falling wave) usually in frontal electrodes - relatively low voltage mixed frequency - no eye movements
Stages 3-4 (Slow Wave Sleep) EEG: EOG: EMG: - slow waves: high-voltage (>=75uV) slow delta activity (<=2 Hz) - no eye movement - low tonic activities
REM sleep EEG: EOG: EMG: - relatively low voltage, mixed frequency - contains rapid eye movements - tonically suppressed (Sleep Paralysis)
3. Relating sleep EEG with behavioural data
Focus on methods using REMlogic Duration (e.g Total Sleep Time (TST)) Individual Sleep stages Other software/programme EEGLAB (Matlab) Frequency analysis Spindle analysis
Hypogram Total SLEEP Time Total SCORED Time
Study 1: Overnight Sleep Wake Sleep Wake Wake Group (n=16) Training AM Same day 10- hours Test PM Sleep Group (n=15) Training PM 10-hours Test AM Next day
Autonomous: Size Congruity Effect Select the (semantically) larger item out of each pair: A) ANT COW congruent B) ANT COW incongruent Size Congruity Effect: Reaction Time for A < B Paivio, 1975; Rubinsten & Henik, 2002
Intentional: Semantic Distance Effect Select the larger animal out of each pair: A) CAT COW Small distance B) ANT COW Large distance Semantic Distance Effect: Reaction Time for B<A Rubinsten & Henik, 2002
Study 1 Task: Select the SEMANTICALLY larger item Malay Presentation of NEW word-pairs differing in both size congruity and semantic distance English Presentation of similar word-pairs in ENGLISH (first language)
Strength of SCE (ms) Strength of SCE (ms) Size Congruity Effect (Most Salient Items: large dist & font diff) 120 Malay New Words 120 English Existing Words 100 80 * 100 80 60 60 40 40 20 20 0-20 -40 WAKE SLEEP 0-20 -40 WAKE SLEEP
Strength of SCE (ms) Strength of SCE (ms) Sleep Spindles (stage 2) and Size Congruity 350 Malay New Words English Existing Words 350 250 250 150 150 50 50-500.00 1.00 2.00-500.00 1.00 2.00-150 Spindle density (num of Spindles per min) -150 Spindle density (num of Spindles per min) r=.650, p<.05; uncorrected p=.009 r=.225, p>.05
Strength of SDE (ms) Strength of SDE (ms) Semantic Distance Effect (Congruent trials) Malay New Words English Existing Words 180 160 * 180 160 140 140 120 120 100 100 80 80 60 WAKE SLEEP 60 WAKE SLEEP
Strength of SDE (ms) Strength of SDE (ms) Slow-Wave Sleep and Semantic Distance (congruent trials) 250 Malay New Words 250 English Existing Words 200 150 100 50 0 10% 20% 30% Time in SWS (% total sleep time) r=.661, p<.05; uncorrected p=.007 200 150 100 50 0 10% 15% 20% 25% 30% Time in SWS (% total sleep time) r=-.053, p>.05
SWS % Neural Reorganisation (PPI) Seed placed in the right hippocampus (27, -10, -17) Remote > Recent Recollection*SWS x = 36 35.00 30.00 25.00 20.00 y = 53 15.00 10.00 5.00 0.00-4.00-2.00 0.00 2.00 4.00 Parameter Estimates p = 0.001 (uncorrected), minimum cluster extent threshold: k = 10 voxels
Nap Design Second order constraints built into syllable list Wake group Syllable production (x48) Video Syllable production (x96) 12 pm 1 pm 2 pm 3 pm 4 pm 5 pm Sleep group Syllable production (x48) Sleep Syllable production (x96)
Implicit learning of second order constraints (Warker & Dell, 2006; Warker et al, 2008; Warker, 2010) Speeded repetition of 4-syllable tongue twisters in time with metronome: Restricted Consonants (s/f) mas fak hag nang kin sing hif gim s after a, before i f before a, after i can appear freely in onset or coda position Unrestricted Consonants (k/m/g/n)
Change in Constraint Strength (Nap) Correlation with SWS Proportion 60% 50% 40% 30% 20% 10% 0% -10% -20% -30% -40% r =.54 *** 0% 10% 20% 30% 40% 50% 60% 70% Slow Wave Sleep Percentage
Elaine Tham (PhD student): Ekht500@york.ac.uk Special thanks to Alex Reid for previous version of this talk Thanks to Gareth (nap data) and Scott (fmri)