Stimulation of Category-Selective Brain Areas Modulates ERP to Their Preferred Categories

Abstract

Neural selectivity to specific object categories has been demonstrated in extrastriate cortex with both functional MRI [1-3] and event-related potential (ERP) [4, 5]. Here we tested for a causal relationship between the activation of category-selective areas and ERP to their preferred categories. Electroencephalogram (EEG) was recorded while participants observed faces and headless bodies. Concurrently with EEG recording, we delivered two pulses of transcranial magnetic stimulation (TMS) over the right occipital face area (OFA) or extrastriate body area (EBA) at 60 and 100 ms after stimulus onset. Results showed a clear dissociation between the stimulated site and the stimulus category on ERP modulation: stimulation of the OFA significantly increased the N1 amplitude to faces but not to bodies, whereas stimulation of the EBA significantly increased the N1 amplitude to bodies but not to faces. These findings provide the first evidence for a specific and causal link between activity in category-selective networks and scalp-recorded ERP to their preferred categories. This result also demonstrates that the face and body N1 reflects several nonoverlapping neural sources, rather than changes in face-selective mechanisms alone. Lastly, because early stimulation (60-100 ms) affected selectivity of a later ERP component (150-200 ms), the results could imply a feed-forward connection between occipital and temporal category-selective areas.

Extracting Visual Evoked Potentials from EEG Data Recorded During fMRI-guided Transcranial Magnetic Stimulation

“…The subtraction technique described, which was validated 19 and applied in several studies 7,26,27 , has several noteworthy advantages: it allows elimination of the residual long-lasting TMS artifact covering the time-window of most significant ERP components; it equally eliminates the artifact components from muscular, mechanical (electrical interference to the electrodes) and non-desired cortical (e.g. auditory) origins; and it is robust and reliable even in electrodes lying directly under or in proximity to the coil.…”

“…Use a high sampling rate for a better representation of the noise artifact. It is recommended to use 1 kHz or above, as most previous studies using this method have done 7,[26][27][28] . 5.…”

“…Place the reference and ground electrodes as far from the coil as possible. In this example, areas in the occipital cortex were targeted using a nose reference and an Fz ground 7 . For other examples see 3,4,27,29,30 .…”

“…To illustrate all of the above, we will provide an example of a study we conducted previously, in which EEG was recorded concurrently with TMS guided by fMRI activations 7 . In this study, a double dissociation was made between face-selective and body-selective ERPs: although face and body ERPs peak around the same latency and electrode sites, targeting individually defined face-selective and body-selective areas in the lateral occipital lobe enabled us to dissociate the neural networks underlying each ERP response.…”

“…Indeed, in recent years researchers have started using TMS in combination with EEG to directly relate cortical areas with event-related potentials (ERP) or oscillatory activity patterns (e.g. [2][3][4][5][6][7] ). In this methodological paper we will describe a particular and useful framework for combining TMS and EEG: fMRI-guided TMS during an ERP experiment.…”