Parametric design and correlational analyses help integrating fMRI and electrophysiological data during face processing

• Published in: NeuroImage (Orlando, Fla.) Vol. 22; no. 4; pp. 1587 - 1595
• Main Authors: Horovitz, Silvina G, Rossion, Bruno, Skudlarski, Pawel, Gore, John C
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01-08-2004; Elsevier Limited
• Subjects: Adult; Algorithms; Attention - physiology; Brain; Brain Mapping; Cerebral Cortex - physiology; Color Perception - physiology; Discrimination Learning - physiology; Dominance, Cerebral - physiology; Electroencephalography; ERP; Evoked Potentials, Visual - physiology; Face; Face perception; Female; fMRI; Humans; Image Enhancement; Image Processing, Computer-Assisted; Magnetic Resonance Imaging; Male; Medical imaging; Methods; Nerve Net - physiology; NMR; Nuclear magnetic resonance; Oxygen - blood; Parahippocampal Gyrus - physiology; Pattern Recognition, Visual - physiology; Perceptual Masking - physiology; Signal Processing, Computer-Assisted; Statistics as Topic; Studies; Temporal Lobe - physiology; fMRI; Face perception; ERP
• ISSN: 1053-8119; 1095-9572
• DOI: 10.1016/j.neuroimage.2004.04.018

Face perception is typically associated with activation in the inferior occipital, superior temporal (STG), and fusiform gyri (FG) and with an occipitotemporal electrophysiological component peaking around 170 ms on the scalp, the N170. However, the relationship between the N170 and the multiple face-sensitive activations observed in neuroimaging is unclear. It has been recently shown that the amplitude of the N170 component monotonically decreases as gaussian noise is added to a picture of a face [Jemel et al., 2003]. To help clarify the sources of the N170 without a priori assumptions regarding their number and locations, ERPs and fMRI were recorded in five subjects in the same experiment, in separate sessions. We used a parametric paradigm in which the amplitude of the N170 was modulated by varying the level of noise in a picture, and identified regions where the percent signal change in fMRI correlated with the ERP data. N170 signals were observed for pictures of both cars and faces but were stronger for faces. A monotonic decrease with added noise was observed for the N170 at right hemisphere sites but was less clear on the left and occipital central sites. Correlations between fMRI signal and N170 amplitudes for faces were highly significant (
P < 0.001) in bilateral fusiform gyrus and superior temporal gyrus. For cars, the strongest correlations were observed in the parahippocampal region and in the STG (
P < 0.005). Besides contributing to clarify the spatiotemporal course of face processing, this study illustrates how ERP information may be used synergistically in fMRI analyses. Parametric designs may be developed further to provide some timing information on fMRI activity and help identify the generators of ERP signals.