NEUROSCIENCE RESEARCH ARTICLE V. V. Lazarev et al. / Neuroscience 426 (2020) 50–58 Experimental-design Specific Changes in Spontaneous EEG and During Intermittent Photic Stimulation by High Definition Transcranial Direct Current Stimulation V. V. Lazarev, a* N. Gebodh, b T. Tamborino, a M. Bikson b and E. M. Caparelli-Daquer c,d a Laboratory of Neurobiology and Clinical Neurophysiology, National Institute of Women, Children and Adolescents Health Fernandes Figueira, Oswaldo Cruz Foundation (FIOCRUZ), Rio de Janeiro, Brazil b Department of Biomedical Engineering, The City College of New York of The City University of New York, New York, NY, USA c Laboratory of Electrical Stimulation of the Nervous System, Department of Physiological Sciences and Neurosurgery Unit, Rio de Janeiro State University (UERJ), Rio de Janeiro, Brazil d Hospital Universita´rio Gaffre´e e Guinle, Federal University of the State of Rio de Janeiro (UniRio), Rio de Janeiro, Brazil Abstract—Electroencephalography (EEG) as a biomarker of neuromodulation by High Definition transcranial Direct Current Stimulation (HD-tDCS) offers promise as both techniques are deployable and can be integrated into a single head-gear. The present research addresses experimental design for separating focal EEG effect of HD- tDCS in the ‘4-cathode  1-anode’ (4  1) montage over the left motor area (C3). We assessed change in offline EEG at the homologous central (C3, C4), and occipital (O1, O2) locations. Interhemispheric asymmetry was accessed for background EEG at standard frequency bands; and for the intermittent photic stimulation (IPS). EEG was compared post- vs pre-intervention in three HD-tDCS arms: Active (2 mA), Sham (ramp up/down at the start and end), and No-Stimulation (device was not powered), each intervention lasting 20 min. The asymmet- ric background EEG changes were only in the central areas with right-side amplitude spectra prevalence, most pronounced in the no-stimulation arm, where they depended on comparison time-points and were consistent with markers of transition between drowsiness and vigilance – bilateral decrease in the delta and asymmetric central increase in the alpha and beta1 bands. For the active arm, similar but less pronounced changes occurred in the alpha band. In contrast, responses to IPS developed similar asymmetric amplitude increase at four harmonics of the IPS of 3 Hz only in the active arm, against a background of a brain-wide symmetric increase in both active and sham arms. Our protocols and analyses suggest methodological caveats for how EEG of tDCS studies could be conducted to isolate putative brain polarization outcomes. Ó 2019 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: transcranial direct current stimulation (tDCS), electroencephalography (EEG), Intermittent photic stimulation, EEG photic driving. INTRODUCTION use of electroencephalography (EEG) with the application of tDCS is compelling mechanically, as both techniques Even as trials of transcranial direct current stimulation integrate into a single head-gear (Woods et al., 2016); (tDCS) for clinical rehabilitation and cognitive and biophysically, through the principle of reciprocity enhancement progress (Brunoni et al., 2012; Bikson (Dmochowski et al., 2017). Measurement of EEG ‘‘online” et al., 2016; Antal et al. 2017; Baudewig et al., 2018; during tDCS is complicated by inherent artifacts (Gebodh Brunelin et al., 2018; David et al., 2018; Osoegawa et al., 2019). Several trials have measured EEG ‘‘offline” et al., 2018), questions remain about the most robust comparing post- to pre-tDCS (Fregni et al., 2006; biomarkers of neuromodulation, including application- Zaehle et al., 2011; Faria et al., 2012; Maeoka et al., specific target engagement (Bikson et al., 2018). The 2012; Auvichayapat et al., 2013; Schestatsky et al., 2013; Castillo-Saavedra et al., 2016; Liu et al., 2016; *Corresponding author. Kang et al., 2018), including one trial testing changes by E-mail addresses: [email protected] (V. V. Lazarev), ngebodh01@ a sham protocol (Holgado et al., 2018). High Definition citymail.cuny.edu (N. Gebodh), [email protected] transcranial Direct Current Stimulation (HD-tDCS) is (T. Tamborino), [email protected] (M. Bikson), [email protected] (E. M. Caparelli-Daquer). advantageous over conventional sponge-pad tDCS since Abbreviations: BGR, background; EEG, electroencephalography; HD- HD-tDCS allows for seamless mechanical integration tDCS, High Definition transcranial Direct Current Stimulation; IPS, (both techniques use gel cups, with no risks of saline leak- intermittent photic stimulation. https://doi.org/10.1016/j.neuroscience.2019.11.016 0306-4522/Ó 2019 IBRO. Published by Elsevier Ltd. All rights reserved. 50 V. V. Lazarev et al. / Neuroscience 426 (2020) 50–58 51 ing as with sponge-pads) and for focal stimulation using applied for 20 min, utilizing either active HD-tDCS, sham stimulation montages like the ‘4-cathode  1-anode’ HD-tDCS, or no-stimulation HD-tDCS. EEG was then (4  1) ring (Datta et al., 2008; Caparelli-Daquer et al., acquired after the intervention arm, prior to (BGR3, 2012; Edwards et al., 2013; Kuo et al., 2013). 3 min), during (4 min), and after (BGR4, 3 min) IPS Any study of EEG changes by tDCS is limited by (Fig. 1A). Here spontaneous BGR EEG indicates EEG technical implementation (e.g. stimulation and recording acquired in the absence of photic or electrical hardware interactions (Gebodh et al., 2019); but also stimulation (HD-tDCS). experimental design; the latter is addressed in detail here. In this exploratory study, we examined neurophysiological HD-tDCS outcomes that indicated topographically restricted (focal) effects of HD-tDCS on EEG, namely locations inside In all conditions, Ag/AgCl sintered electrodes were used and outside the stimulation region. In addition to back- for electrical stimulation (tDCS) and were inserted into ground EEG, intermittent photic stimulation (IPS) evoked gel-filled plastic holders (High Definition electrodes – EEG responses were assessed, which can reveal latent Minhas et al., 2010) arranged in a ‘4-cathode 1-anode’ bioelectrical oscillators not present or weak in sponta- (4  1) configuration. The central anodal stimulation elec- neous EEG during resting state (Lazarev et al, 2001). trode was placed on the left hemisphere over standard Our objective was not to identify explicit EEG responses EEG location C3 (International 10/20 System) and the to tDCS but to explore methodological issues surrounding remaining four cathodal stimulating electrodes were the reproducible detection of real changes in EEG includ- placed in 5-cm radius around the anode (Datta et al., ing: the validity of traditional sham approaches, the com- 2009). Stimulation current was supplied with a 4  1 stim- parison of background EEG vs. IPS, the consideration of ulation set-up consisting of a 1  1 current generator con- multiple time-points and EEG frequency bands, and the nected to 4  1 adaptor (Soterix Medical Inc., New York, use of focal HD stimulation to resolve emergent NY, USA). asymmetries. During the active HD-tDCS intervention arm, 2 mA of current was applied over 20 min including 1-min ramp- up and ramp-down periods. In the sham HD-tDCS EXPERIMENTAL PROCEDURES intervention arm, current ramped up to 2 mA then down, Participants only during the first and the last minutes of the intervention period (without stimulation during the Twenty-five healthy individuals (eight males) aged 20– remaining 18 min). In the no-stimulation intervention 57 years (mean ± SD: 33.6 ± 11.1) with no history of arm, the stimulation electrodes were placed at their neurological, psychiatric, chronic or drug-related illness respective locations, however the stimulation device volunteered for the study, without financial was not powered on. compensation. Subjects included, but were not limited to, staff and students at the Fernandes Figueira Institute EEG recording and IPS and the Hospital Universita´ rio Gaffre´ e e Guinle. All subjects underwent neurological and psychiatric EEG signals were acquired with a 32-channel Bio-logic evaluations by a neuropsychiatrist through an oral Ceegraph device (Bio-Logic Systems Corp.) at 8 scalp interview and physical examination. Subjects were locations. In the left hemisphere, the recording medication free and had no implanted electrical devices electrodes were arranged according to the HD-tDCS or intracranial implants. The study was approved by the electrode positions. Three of them were placed inside Ethics Committee of the Hospital Universita´ rio Gaffre´ ee the stimulated area. The central position (C3) Guinle of the Federal University of the State of Rio de corresponded to anode; the posterior central position Janeiro (HUGG UNIRIO). All individuals gave written (C3p, rearwards and to the right from the anode) in the informed consent and received no compensation. ‘‘outer ring” of HD-tDCS corresponded to cathode; and the intermediate position, 3 cm anteriorly from the anode (C3a), was between stimulation electrodes. The Experimental design recording electrode outside the stimulated area was The study consisted of three intervention conditions placed in the left occipital region (O1). In the anode and (arms). These included ‘‘Active” HD-tDCS (20 cathode positions, the recording and stimulation participants), ‘‘Sham” HD-tDCS (14 participants), and electrodes replaced each other before and after HD- ‘‘No-stimulation” HD-tDCS (13 participants). Several tDCS. In the right hemisphere, which was not subjects participated