Orthogonal planar gradiometers

Participants were seated comfortably in the gantry of a 306-sensor TRIUX Elekta system with 204 orthogonal planar gradiometers and 102 magnetometers (Elekta, Finland), in a dimly lit magnetically shielded room. We used a PROPixx DLP LED projector (VPixx Technologies Inc., Canada) mounted outside of the shielded room. Importantly, the projector supported a refresh rate of 1440 Hz, which allowed the application of high-frequency flicker or tagging, which was invisible to participants. Entering through an aperture in the shielded room and being routed by a reflecting mirror, the video output was projected onto a 71 * 40 cm screen placed 1.5 m away from the participant. During the experiment, horizontal and vertical eye movements were tracked with EyeLink 1000 Plus, SR Research Ltd, Canada. MEG signals passed through an embedded analogue filter, 0.1–330 Hz, and sampled at 1000 Hz. Different types of data, including MEG, eye-tracking, and triggers, were combined into a single data-stream in the acquisition computer. The combined data were then split into 100 ms blocks and sent to the stimulus computer using the Fieldtrip buffer (Oostenveld et al., 2011 (link)). Finally, these data were used to train a classifier to decode the direction of covert attention based on the power of the frequency-tagging response.

MEG data were acquired using a 306-sensor TRIUX Elekta Neuromag system with 204 orthogonal planar gradiometers and 102 magnetometers (Elekta, Finland). The data were band-pass filtered online using anti-aliasing filters from 0.1 to 330 Hz and then sampled at 1000 Hz. We used a Polhemus Fastrack electromagnetic digitizer system (Polhemus Inc., USA) to digitize the locations for three bony fiducial points: the nasion, left and right preauricular points. Then, four head-position indicator coils (HPI coils) were digitized: two coils were attached on the left and right mastoid bone and another two were on the forehead with at least 3 cm distance in between. Furthermore, at least 200 extra points on the scalp were acquired for each participant in order to spatially co-register the MEG source analysis with individual structural MRI image. After preparations, participants were seated upright under the MEG gantry with the back rest at a 60°angle.

We used a MEGIN system to collect the neuromagnetic signals from 306 sensors, of which 204 were orthogonal planar gradiometers and 102 were magnetometers (Elekta, Finland). For the preparations, we first placed 4 head-position indicator (HPI) coils on the participant’s head: 2 were on the forehead with more than 3 cm distance in between, and the other 2 were on the left and right mastoid bone behind the ears. Then, we used the Polhemus Fastrack electromagnetic digitizer system (Polhemus, United States of America) to digitise the head position by 3 fiducial anatomical markers (the nasion, left and right preauricular points). Next, we digitised the 4 HPI coils. For the final step, we digitised at least 200 points on the scalp to obtain the whole shape of the head. With the help of head digitalization, the MEG head position can be spatially co-registered with the individual structural MRI images for the source modelling analysis.
Afterwards, participants walked to a dimly lit room and sat in the MEG gantry (60 degrees upright position). The distance between the participant and the projector screen was 145 cm, which yielded 1 visual degree from 35 pixels. The MEG data were sampled at 1,000 Hz after the application of an anti-aliasing 0.1 to 330 Hz band-pass filter.