Network dynamics of human face perception

Abstract

Prevailing theories suggests that cortical regions responsible for face perception operate in a serial, feed-forward fashion. Here, we utilize invasive human electrophysiology to evaluate serial models of face-processing via measurements of cortical activation, functional connectivity, and cortico-cortical evoked potentials. We find that task-dependent changes in functional connectivity between face-selective regions in the inferior occipital (f-IOG) and fusiform gyrus (f-FG) are bidirectional, not feed-forward, and emerge following feed-forward input from early visual cortex (EVC) to both of these regions. Cortico-cortical evoked potentials similarly reveal independent signal propagations between EVC and both f-IOG and f-FG. These findings are incompatible with serial models, and support a parallel, distributed network underpinning face perception in humans.

Fig 1. icEEG analytic methods.

(A) Representative subject (RH, subject 1) showing analytic methods applied to the cortical activity measured using intracranial EEG (icEEG) during a face-naming task. All data are trial-aligned and from highlighted electrodes localized over early visual cortex (EVC; green) and face-selective fusiform gyrus (f-FG; red). fMRI activations depicted on the cortical surface indicate higher responses to faces than non-faces (faces > animate, inanimate, and scramble; p <0.01). fMRI data provided one of three selection criteria used to identify face-selective electrodes (via co-localization). (B) Time-frequency plot of percent power change (relative to pre-stimulus baseline; -700 to -200 ms; t = 0 ms, stimulus onset; face task only) following spectral decomposition of the icEEG signal. Horizontal dashed lines denote range of broadband gamma activity (BGA, 60–120 Hz) used in this study. (C) (Top) BGA profile for face (orange) and non-face control stimuli: animate (purple), inanimate (cyan), scrambled (gray) stimuli. Representative stimuli are provided in Fig 2. Shaded regions denote 1 SEM (across trials; n = 50–60 trials-per-condition). Horizontal orange bar denotes face-selectivity onset in f-FG (faces > non-faces; q <0.01, FDR corrected). Below f-FG x-axis, time window (100 to 400 ms) used for the icEEG d’-analysis is indicated, which independently evaluates electrode face-selectivity. To avoid circularity, odd-trials are used for selectivity onset latencies and even trials for d’ selectivity indices. (D) Functional connectivity assessed using amplitude envelope correlations between electrode pairs (face task only). (Top) For the two electrodes here, trial-by-trial variance of instantaneous BGA envelope is obtained by subtracting the average envelope (black trace) from each trial. (Middle) Noise correlations performed across trials to compute connectivity between electrode pairs. To estimate information flow, zero time-lag correlations are computed (black box), and repeated for both positive (+100 ms, green box) and negative (-100 ms, red box) lag values. (Bottom) Temporal cross-correlograms summarize connectivity across all time-lags (-200 to +200 ms lags). Correlation coefficient values are plotted as a heat map. The black dashed line represents a 0-ms lag. Above this line, EVC activity leads f-FG (positive lag; information flow from EVC to the f-FG), while below the dashed line f-FG activity leads EVC (negative lag; information flow from f-FG-to-EVC). Contours represent significant correlations (q = 0.05, trial re-shuffling, 2000 resamples).

Fig 2. Right hemisphere subject electrodes, fMRI, and icEEG time-series.

(A) Cortical surface models and subdural electrodes are shown for the subjects with right hemispheric coverage in all three regions of interest. Highlighted electrodes (colored spheres) met all anatomical and functional criteria for early visual cortex (EVC, green), and face-selective inferior occipital gyrus (f-IOG, blue) and fusiform gyrus (f-FG, red). Subject-specific fMRI activations, also depicted on cortical surfaces (subjects 1–3), indicate higher responses to faces than non-faces (faces > animate, inanimate, and scramble; p <0.01). fMRI data provided one of three selection criteria used to identify face-selective electrodes (via co-localization). Subject 4 was unable to participate in fMRI recordings (dashed black boxes). fMRI activations depicted in this last case are derived from grouped analysis in 18 healthy volunteers, co-registered to the subject’s own cortical model using a surface-based normalization technique. (B) icEEG time-series representations of normalized (z-scored) mean broadband gamma power changes (BGA; 60–120 Hz; -50 to 700 ms after stimulus onset; stim. on at t = 0 ms), with respect to pre-stimulus baseline (-700 to -200 ms), for EVC, f-IOG, and f-FG electrodes identified in each subject. Time-series traces are color-coded to respective stimulus category–faces (orange) vs. animate (purple) vs. inanimate (cyan) vs. scramble (gray) stimuli. Shading denotes 1 SEM (across electrodes/region/subject; n value). Horizontal orange bars below each trace represent onset of BGA face-selectivity used in f-IOG and f-FG latency difference contrasts (face > non-face stimuli; q < 0.01, FDR corrected for time-points; computed using odd-trials). Subject 4 did not undergo icEEG recordings. Note: Line drawings of non-face stimuli (Figs 1–4) were adapted with permission from: Snodgrass J.G. and Vanderwart M. "A standardized set of 260 pictures: Norms for name agreement, image agreement, familiarity, and visual complexity.” Journal of Experimental Psychology: Human Learning and Memory, Vol 6(2), 1980, 174–215, APA. The official portrait of President Barack Obama by Pete Souza (Figs 1–4), obtained from the White House website (https://www.whitehouse.gov/sites/whitehouse.gov/files/images/Administration/People/president_official_portrait_hires.jpg), is licensed under CC BY 3.0 US (http://creativecommons.org/licenses/by/3.0/us/) as pursuant to White House copyright policy (https://www.whitehouse.gov/copyright/). Original images have been converted to gray-scale and overlaid with grid.

Fig 3. Left hemisphere subject electrodes, fMRI, and icEEG time-series.

A) Cortical surface models and subdural electrodes are shown for the subjects with left hemispheric coverage in all three regions of interest. Highlighted electrodes (colored spheres) met all anatomical and functional criteria for early visual cortex (EVC, green), and face-selective inferior occipital gyrus (f-IOG, blue) and fusiform gyrus (f-FG, red). Subject-specific fMRI activations, also depicted on cortical surfaces (subject 5), indicate higher responses to faces than non-faces (faces > animate, inanimate, and scramble; p <0.01). fMRI data provided one of three selection criteria used to identify face-selective electrodes (via co-localization). For subjects that did not participate in fMRI recordings (dashed black boxes; subjects 6–9), the fMRI activations depicted derive from a grouped analysis in 18 healthy volunteers, which was co-registered to the subject’s own cortical model using a surface-based normalization technique. B) icEEG time-series representations of normalized (z-scored) mean broadband gamma power changes (BGA; 60–120 Hz; -50 to 700 ms after stimulus onset; stim. on at t = 0 ms), with respect to pre-stimulus baseline (-700 to -200 ms), for EVC, f-IOG, and f-FG electrodes identified in each subject. Time-series traces are color-coded to respective stimulus category–faces (orange) vs. animate (purple) vs. inanimate (cyan) vs. scramble (gray) stimuli. Shading denotes 1 SEM (across electrodes/region/subject; n value). Horizontal orange bars below each trace represent onset of BGA face-selectivity used in f-IOG and f-FG latency difference contrasts (face > non-face stimuli; q < 0.01, FDR corrected for time-points; computed using odd-trials).

Fig 4. Functional connectivity during face perception.

(A) Group temporal cross correlograms of right hemisphere EVC-f-IOG connectivity, computed by averaging individual amplitude envelope correlations (n = 3 subjects, contours denote significant connectivity, q = 0.01, FDR corrected) for face stimuli only. Amplitude envelope correlations are measured across lag ranges of -150 to +150 ms. The black dashed diagonal line represents a lag of 0 ms. Above the dashed line activity in EVC activity leads f-IOG (information flow from EVC to the f-IOG), while below the dashed line f-IOG activity leads EVC (information flow from f-IOG to EVC). (B) Connectivity between EVC and the f-FG, right hemisphere (n = 3 subjects). (C) Connectivity between f-IOG and f-FG, right hemisphere (n = 3 subjects). (D) Connectivity between EVC and the f-IOG, left hemisphere (n = 4 subjects). (E) Connectivity between EVC and the f-FG, left hemisphere (n = 5 subjects). (F) Connectivity between f-IOG and the f-FG, left hemisphere (n = 4 subjects).

Fig 5. Cortico-cortical connections of the face-network.

(Top) Cortico-cortical evoked potentials (CCEPs) visualized on subject cortical surface models during stimulation at a pair of early visual cortex (EVC; left) and face-selective fusiform gyrus (f-FG; right) electrodes. Cyan electrodes denote stimulation pairs (bipolar pulses; 10 mA, 500 micro-second pulse width; 1 Hz for 50s). Amplitude (radius of electrode) and latency (color) of the N1 responses are represented. Electrodes without CCEP responses are depicted as white spheres. Normalized evoked potentials (waveform deflections) are plotted for the encircled electrodes recording from (left inset) f-IOG and f-FG electrodes and (right inset) EVC and f-FG electrodes. Shadings represent 1 SEM (n = 50 stimulation trials). (Bottom) Scatter & box plots for all subjects that underwent CCEP recordings during either EVC stimulation or f-FG stimulation. Points represent single-trial N1 latency differences between f-IOG and f-FG electrodes during EVC stimulation (left), and between EVC and f-IOG electrodes during f-FG stimulation (right).