Elsevier

NeuroImage

Volume 140, 15 October 2016, Pages 76-82
NeuroImage

Different coupling modes mediate cortical cross-frequency interactions

https://doi.org/10.1016/j.neuroimage.2015.11.035Get rights and content

Highlights

  • Frequency-specific tACS effects on cross-frequency coupling.

  • Distinct coupling modes differentially affect cross-frequency coupling.

  • Selective modulation of alpha power by tACS.

  • Successful cross-frequency analysis in combined EEG-tACS recordings.

Abstract

Cross-frequency coupling (CFC) has been suggested to constitute a highly flexible mechanism for cortical information gating and processing, giving rise to conscious perception and various higher cognitive functions in humans. In particular, it might provide an elegant tool for information integration across several spatiotemporal scales within nested or coupled neuronal networks. However, it is currently unknown whether low-frequency (theta/alpha) or high-frequency gamma oscillations orchestrate cross-frequency interactions, raising the question of who is master and who is slave. While correlative evidence suggested that at least two distinct CFC modes exist, namely, phase-amplitude-coupling (PAC) and amplitude-envelope correlations (AEC), it is currently unknown whether they subserve distinct cortical functions. Novel non-invasive brain stimulation tools, such as transcranial alternating current stimulation (tACS), now provide the unique opportunity to selectively entrain the low- or high-frequency component and study subsequent effects on CFC. Here, we demonstrate the differential modulation of CFC during selective entrainment of alpha or gamma oscillations. Our results reveal that entrainment of the low-frequency component increased PAC, where gamma power became preferentially locked to the trough of the alpha oscillation, while gamma-band entrainment enhanced AECs and reduced alpha power. These results provide causal evidence for the functional role of coupled alpha and gamma oscillations for visual processing.

Introduction

Cognition and conscious perception are thought to arise from neuronal interactions between functionally specialized but widely distributed cortical regions (Siegel et al., 2012). While phase synchronization between task-relevant cortical areas might integrate information across several spatial scales (Fries, 2005), cross-frequency coupling (CFC) has been suggested to constitute a flexible mechanism for information integration across temporal scales (Canolty and Knight, 2010). Hence, it might serve as a key mechanism for selective gating and processing of information within coupled or nested cortical networks and thus, subserve numerous cognitive functions in humans (Canolty and Knight, 2010, Engel et al., 2013, Voytek et al., 2010). However, the functional role of CFC is currently extensively under debate, given that (I) various methodological constrains hamper its interpretation (Aru et al., 2014), (II) the evidence supporting its role for cognitive processing was only correlative in nature (Canolty and Knight, 2010, Voytek et al., 2010), and (III) it remained unclear whether different coupling modes (e.g. PAC or AEC) subserve distinct cortical functions. In particular, alpha–gamma PAC has been suggested to constitute a powerful mechanism to organize visual processing (Jensen et al., 2014, Spaak et al., 2012), while theta–gamma PAC might subserve memory processes and long-range cortico-cortical communication (Axmacher et al., 2010, Canolty and Knight, 2010, Lisman and Jensen, 2013, Tort et al., 2009). Most studies focused on PAC (Canolty and Knight, 2010, Voytek et al., 2010), where gamma power is preferentially phase-locked to the trough of the theta (4–7 Hz) or the alpha (8–12 Hz) rhythm. So far, it remained elusive whether the fast or the low-frequency spectral component drives their interaction (Jiang et al., 2015, Schroeder and Lakatos, 2009, Spaak et al., 2012).

In addition, it has recently been observed that the cerebral cortex exhibits a large-scale correlation structure which is independent from phase-locked signaling and can best be analyzed by quantifying AEC (Engel et al., 2013, Hipp et al., 2012). While first applied to uncover envelope correlations within one frequency range, AEC can also be applied to the cross-frequency domain (Helfrich et al., 2014a). Fig. 1A depicts a schematic how both CFC measures (PAC and AEC) can be obtained from raw data.

To investigate the mechanisms underlying the alpha–gamma interplay, we took advantage from a novel non-invasive brain stimulation technique, namely, transcranial alternating current stimulation (tACS; Herrmann et al., 2013, Thut et al., 2011), which has recently been demonstrated to selectively entrain neuronal oscillations within canonical frequency boundaries (Helfrich et al., 2014b, Ozen et al., 2010). Thus, this approach provided the unique opportunity to selectively drive one spectral component and study subsequent cross-frequency effects in coupled frequency bands. Here, we reanalyzed data from two combined tACS-EEG (electroencephalography) studies on visual perception, which was obtained during stimulation. Our aim was to study how cross-frequency interactions were modulated when either alpha (8–12 Hz; Helfrich et al., 2014b) or gamma (> 35 Hz; Helfrich et al., 2014a) oscillations were entrained by tACS, i.e., when one of these processes was set up as the master through synchronization to the external driving force.

Section snippets

Participants

A total of 30 healthy, right-handed participants were recruited from the University of Oldenburg, Germany, and the University Medical Center Hamburg-Eppendorf, Germany. Sixteen subjects (8 female; mean age: 24.6 ± 2.8 years; 2 excluded due to technical difficulties during data acquisition) participated in study 1 and fourteen subjects (7 female; mean age: 27.5 ± 6.7 years) participated in study 2. All participants reported no history of neurological or psychiatric disease and were medication free

Results

In a previous study, we found that alpha power was increased during 10 Hz tACS as compared to sham (Helfrich et al., 2014b; + 49.3% ± 21.8% (mean ± SEM); t13 =  2.26, p = 0.04). Entrainment was quantified by means of circular statistics, revealing a non-uniform phase distribution in presence of the externally applied electric field, as well as by increments in inter-trial coherence and phase-locking to the externally applied sine wave (for details and statistics, please see Fig. 4 in Helfrich et al.,

Discussion

Our results demonstrate that selective driving of alpha or gamma oscillations differentially modulated their physiological interaction. Entrainment of alpha oscillations increased PAC and promoted a preferential phase-locking of gamma power to the alpha trough, while gamma-band entrainment modulated AECs. These results indicate that both can serve as master and slave through distinct coupling modes, possibly working as opponent processes to subserve different cortical functions (Schroeder and

Acknowledgements

The authors declare no competing financial interests. This work was supported by grants from the European Union (ERC-2010-AdG-269716, AKE), the German Research Foundation (SFB936/A3/Z1, AKE; SFB/TRR 31, CSH; SPP1665 EN 533/13-1, AKE; SPP1665 HE 3353/8-1, CSH), the German National Academic Foundation (RFH) and the Alexander von Humboldt Foundation (Feodor Lynen Program, RFH).

References (43)

  • J.E. Lisman et al.

    The θ-γ neural code

    Neuron

    (2013)
  • E. Maris et al.

    Nonparametric statistical testing of EEG- and MEG-data

    J. Neurosci. Methods

    (2007)
  • T. Neuling et al.

    Good vibrations: oscillatory phase shapes perception

    NeuroImage

    (2012)
  • T. Neuling et al.

    Friends, not foes: magnetoencephalography as a tool to uncover brain dynamics during transcranial alternating current stimulation

    NeuroImage

    (2015)
  • M.A. Nitsche et al.

    Transcranial direct current stimulation: state of the art 2008

    Brain Stimul.

    (2008)
  • R. Polanía et al.

    The importance of timing in segregated theta phase-coupling for cognitive performance

    Curr. Biol.

    (2012)
  • R. Schulz et al.

    Non-invasive brain stimulation in neurological diseases

    Neuropharmacology

    (2013)
  • D.J.L.G. Schutter et al.

    Retinal origin of phosphenes to transcranial alternating current stimulation

    Clin. Neurophysiol. Off. J. Int. Fed. Clin. Neurophysiol.

    (2010)
  • E. Spaak et al.

    Layer-specific entrainment of γ-band neural activity by the α rhythm in monkey visual cortex

    Curr. Biol.

    (2012)
  • S. Yuval-Greenberg et al.

    Transient induced gamma-band response in EEG as a manifestation of miniature saccades

    Neuron

    (2008)
  • J. Aru et al.

    Untangling cross-frequency coupling in neuroscience

    Curr. Opin. Neurobiol.

    (2014)
  • Cited by (0)

    1

    These authors contributed equally to this work.

    View full text