Table 2 Human findings regarding theta-gamma coupling (TGC)
Authors & Year of Publication | Subjects | Measure | Region of Brain/Scalp Examined | Task Performed/Behavior during Measurement | WM Subprocess(es) Analyzed | Relevant Findings | Conclusions |
|---|---|---|---|---|---|---|---|
Studies utilizing intracranial EEG (iEEG) | |||||||
i. Fell et al. (2003) [106] | 9 patients with pharmacoresistant temporal lobe epilepsy | iEEG | Hippocampus, entorhinal cortex | Single-trial word list learning paradigm | Encoding | - Phase-phase TGC across the rhinal cortex and hippocampus during WM encoding - Coupling greater for words successfully recalled | TGC crucial to actual memory formation rather than general state of WM encoding |
ii. Mormann et al. (2005) [76] | 12 patients with pharmacoreistant temporal lobe epilepsy (5 women, aged 40.3 ± 10.1 years) | iEEG | Hippocampus, rhinal cortex | Continuous visual word recognition paradigm | Continuous recognition (Combination of encoding and retrieval) | - Modulation of gamma- amplitude by theta- phase in both rhinal cortex hippocampus - Rhinal modulation greatest for correct rejections, hippocampal modulation greatest for hits | Phase-amplitude TGC may underlie encoding processes in rhinal cortex and retrieval processes in hippocampus |
iii. Axmacher et al. (2010) [33] | 14 participants with pharmacoresistant temporal lobe epilepsy (3 female, aged 38.3 ± 11.8 years) | iEEG | Hippocampus | Visual analogue of Sternberg task | Maintenance | - Theta phase to gamma power TGC increased during maintenance compared to baseline - More precise coupling associated with faster reaction times - Increased WM load related to less modulation of gamma | - Medial temporal lobe involved in WM when multiple stimuli are to be maintained - TGC functionally relevant to WM processes |
iv. Canolty et al. (2006) [61] | 5 females (aged 24 to 45 years) | iEEG | Frontal, temporal, parietal lobes | - Passive listening - Active listening/target detection - Picture naming - Verb generation - Finger tapping - Mouth movement & articulation - Auditory/vibrotactile stimulus presentation with unimodal target detection - Auditory N-back task - Visual search task - Visual silent reading task - Facial emotion recognition task - Motor, auditory and tactile imagery | Unclear | - Strong modulation of gamma throughout examined areas by theta phase across all tasks; highest modulation in high gamma (80–150Hz) - Largest gamma amplitudes occur at trough of theta | TGC coordinates various cortical regions for purposeful activity |
v. van der Meij, Kahana, & Maris (2012) [107] | 27 patients (12 female) with pharmacoresistant epilepsy | iEEG | Frontal, temporal, parietal, occipital, limbic | Sternberg task | Encoding, maintenance | Widespread phase-amplitude coupling between multiple frequency bands across distances greater than 10Â cm between sensors spread throughout cortex and other measured areas | - Phase-amplitude coupling is pervasive, has a wide spatial distribution, and varies greatly in terms of frequencies involved and phases preferred - As such, phase-amplitude coupling likely involved in various forms of selective communication between brain areas |
vi. Maris, van Vugt, & Kahana (2011) [108] | 26 patients with implanted electrodes for purpose of presurgical diagnosis | iEEG | Varied from patient-to-patient, but included frontal, temporal and parietal sites | Sternberg task | Encoding, maintenance | - Phase to amplitude coupling widely distributed across frequency as well as location; included frontal, temporal and parietal sites -TGC most prominent in within-electrode analyses - Phase-amplitude coupling increased in some areas during WM and decreased in others - In most coupling patterns, high-frequency bursts are synchronized across locations | - WM processes affect coupling heterogeneously, depending on location and frequencies involved - Phase-amplitude coupling important to coordination of neural activity spatiotemporally |
vii. Chaieb et al. (2015) [34] | 14 patients with pharmacoresistant temporal lobe epilepsy | iEEG | Hippocampus | Modified Sternberg paradigm featuring facial stimuli | Maintenance | - Increases in phase-phase TGC observed for multi-item but not single-item trials for coupling factor of 6 (e.g., 6 gamma cycles to 1 theta cycle) - Changes in coupling measures across WM loads predicted WM capacities | TGC particularly important to multi-item WM |
Studies utilizing scalp-level recordings | |||||||
viii. Schack, Vath, Petsche, Geissler, & Mӧller (2002) [70] | 10 right-handed female adults (aged 25 to 35 years) | Scalp EEG | Frontal, prefrontal, frontopolar regions | Sternberg tasks featuring number words and random figures | Encoding, maintenance and retrieval | - Phase-coupling between theta activity over frontal areas and gamma activity over frontopolar areas during maintenance of number words | TGC is prominent during human maintenance of information in WM, supporting Lisman & Idiart’s [1] model |
ix. Baerbel Schack & Weiss (2005) [109] | 23 females (aged 23.7 ± 2.7 years) | Scalp EEG | Scalp-wide | Single-trial word list learning paradigm | Encoding | - Increased frontocentral phase-phase TGC for words later recalled vs. those not - Phase-phase TGC also observed between Fz and Cz | Gabor expansion is a useful means of assessing phase-phase TGC from scalp EEG |
x. Sauseng, Klimesch, Gruber, & Birbaumer (2008) [69] | 22 analyzed patients (19 females, average age of 23.6Â years) | Scalp EEG | Scalp-wide | Cued visual attention task | Target presentation | - Phase-phase TGC over posterior areas enhanced by attention shift to cued visual field - Stronger TGC over areas contralateral to cued visual field | Phase-phase TGC over posterior areas reflects matching of stored information to incoming visual input |
xi. Sauseng et al. (2009) [32] | 21 right-handed participants (15 females, aged 23.9 ± 4.4 years) | Scalp EEG | Scalp-wide | Visuospatial delayed match-to-sample task | Maintenance | - Phase-amplitude TGC over bilateral parietal and occipital sites - Phase-phase TGC modulated by WM load over right posterior sites, exclusive to relevant information - Degree of phase-phase TGC modulation by WM load predictive of WM capacity | - TGC, and particularly phase-phase relationships, are central to visual WM function - Independent mechanisms for maintenance of relevant vs. suppression of irrelevant info |
xii. (Holz, Glennon, Prendergast, & Sauseng (2010) [68] | 18 participants (13 female, aged 23.67 ± 2.99 years) | Scalp EEG | Scalp-wide | Visuospatial delayed match-to-sample task | Retrieval | - Increased phase-phase TGC over parieto-occipital brain regions soon after probe presentation - Matching increased TGC in right hemisphere; non-matching increased TGC in left hemisphere | Comparison of stored information to visual inputs leads to phase-phase TGC over parieto-occipital regions |
xiii. Park, Lee, & Lee (2011) [36] | 31 older participants (19 females, aged 66.45 ± 5.75 years) | Scalp EEG | Scalp-wide | Spatial delayed match-to-sample task | Maintenance | - TGC between theta phase and gamma power across scalp, enhanced in left centro-parietal regions - Strength of TGC at FP1 predicted task performance - Strength of TGC at P8 predicted delayed figure recall performance | TGC is important to visual memory function (e.g., parietal correlation) but may also reflect attentional processes (e.g., prefrontal correlation) |
xiv. Park, Jhung, Lee, & An (2013) [35] | 13 right-handed participants (5 females, aged 20.8 ± 3.1 years) | Scalp EEG | Scalp-wide | Visuospatial 2-back task, vigilance task | Maintenance | TGC between theta phase and gamma power most prominent over prefrontal areas in WM task but not vigilance task | TGC recorded from scalp represents WM processes rather than attentional processes |
xv. Lee & Yang (2014) [110] | 9 right-handed males (aged 21.22 ± 0.97 years) | Scalp EEG | Scalp-wide | Visuospatial delayed match-to-sample task | Maintenance | Strength of phase-phase TGC at P8 predicted task performance | Phase-phase TGC can be used as a measure of WM ability |
xvi. Vosskuhl, Huster, & Herrmann (2015) [111] | 33 right-handed participants (14 females, aged 25.74 ± 2.69 years) | Scalp EEG | F3, Fz, F4, C3, Cz, C4 | Transcranial alternating current stimulation; digit span task; 3-back task | Encoding during stimulation | - Short-term memory capacity as assessed using forward digit span was increased during stimulation - Theta amplitude increased after stimulation | Supports theory that the length of theta cycle (e.g., theta frequency) limits how many gamma cycles can be superimposed upon it |
xvii. Heusser, Poeppel, Ezzyat, & Davachi (2016) [77] | 20 right-handed males (aged 21–25; mean age 28) | MEG | Scalp-wide; left lateral and posterior clusters of sensors used for analyses | Visual temporal order memory paradigm | Encoding | - Peak of gamma power shifted progressively later with memory stimuli progressively later in presented sequence - Relative gamma peaks reflected the order of stimulus presentation in sequence | Strongly supports notion that the sequential order of stimuli is encoded using a theta-gamma code |
xviii. Alekseichuk, Turi, de Lara, Antal, & Paulus (2016) [112] | 47 healthy people (25 females, aged 19–28) | Scalp EEG | Scalp-wide | Transcranial alternating current stimulation; 2-back visuospatial memory test | Encoding/retrieval during stimulation | - Transcranial alternating current stimulation improved task performance and increased phase connectivity across scalp - 80Hz to 100Hz stimulation during theta peak produced optimal effects | - Transcranial current in gamma frequency range improved WM performance - High gamma may be particularly important to WM-related coupling |