Recording and quantifying time-domain face-selective activity in the VOTC.

A. SEEG (depth) electrode arrays (white circles) shown on the reconstructed white matter surface of one of the participants (ventral view of the left hemisphere). Electrodes penetrate both gyral and sulcal cortical tissues. B. The frequency-tagging paradigm to quantify face-selective neural activity: images of nonface objects appear at a rate of six stimuli per second (6 Hz) with variable face images presented every five stimuli (i.e., every 0.835s). Each stimulation sequence lasts for 65s (2s showed here). C. Representative examples of natural face images used in the study (actual images not shown for copyright reasons). D. Top: example raw intracranial EEG signal measured at the bipolar recording contacts shown in panel A (red). The signal is shown from −1.5 to 20 s relative to the onset of a stimulation sequence. The time-series displayed is an average of 2 sequences. Above the time-series, red vertical ticks indicate the appearance of face image every 0.835s and small black vertical ticks indicate the appearance of non-face objects every 0.167s. Middle: time by frequency representation of SEEG data in the HFB range (30-160 Hz). The plot shows the percent signal change at each frequency relative to a pre-stimulus baseline period (−1.6s to −0.3s). Periodic burst of HFB activity at the frequency of face stimulation (i.e., 1.2 Hz) are visible. Bottom: modulation of HFB amplitude over time obtained by averaging time-frequency signals across 30-160 Hz. E. HFB signal is transformed in the frequency domain to quantify face-selective amplitude as the sum of 12 face-selective frequency harmonics. F. Time-domain averaged HFB response to face images shows both the periodic response to non-face objects at 6 HZ (cycle duration = 0.167 s) and the larger face-selective response starting ∼0.1 s after face onset. Shaded area around the curve is standard error of the mean across face trials. G. Mean face responses from two separate example recording contacts (in PTL and ATL) in which the 6 Hz response to non-face objects has (black traces) or has not (blue traces) been filtered-out.

Time-domain face-selective periodic increases.

Left column: Time by frequency response (percent signal change, psc; see scale value at the top of each plot and color scale at the bottom) in the HFB range (30-160 Hz) averaged over face-selective contacts in each main VOTC region (OCC: occipital, bottom; PTL: Posterior temporal lobe, middle; ATL: anterior temporal lobe, top). For each recording contact, the time-frequency data was segmented in epochs of about 3 face cycles (i.e. 3 x 0.833 s), averaged by contacts and then averaged over the three groups of contacts. Frequency axis is on the left. Green traces are the HFB amplitude envelope obtained by averaging over the 30-160 Hz range. Amplitude (psc) axis is on the right, show in green. Right column: Frequency spectra averaged over the corresponding groups of recording contacts and showing the face-selective response (red circles) at multiples of 1.2 Hz (i.e., face stimulation frequency) and visual response at 6 Hz (black square, other harmonics not shown). This highlights the sharp decrease in general visual response (i.e., 6Hz) relative to face-selective response from posterior to anterior VOTC.

Spatial organization and increase in abstraction of face-selective HFB activity in VOTC.

A. Map of all VOTC recording contacts across the 140 individual brains displayed in the Talairach space in a transparent reconstructed cortical surface of the Colin27 brain (ventral view). Each circle represents a single recording contact. Face-selective contacts are color-coded according to their anatomical location in the original individual anatomy. White-filled circles correspond to contacts without significant face-selective activity. Values along the y-axis of the Talairach coordinate system (postero-anterior) are shown near the interhemispheric fissure. B. VOTC maps of the local proportion of face-selective contacts relative to the number of recorded contacts. Black contour lines delineate local proportions significantly above zero (p < 0.01, percentile bootstrap). C. The number of face-selective contact is show for each anatomical region (region defined in each individual participant) and hemisphere. D. Face-selectivity index (FSI) along the postero-anterior axis collapsed along the X dimension (medio-lateral). The shaded area shows the 99% confidence interval computed using a percentile bootstrap. E. Map of the proportion of face-exclusive (i.e., face-selective without significant response to nonface-objects at 6Hz and harmonics) relative to face-selective contacts across VOTC.

Concurrent face-selective activity across VOTC.

A. Mean time-domain face-selective HFB activity in each VOTC region (OCC, PTL, ATL) and hemisphere. HFB time-series were filtered to remove the general visual response at 6 Hz and harmonics so only face-selective signal remains. The maximum amplitude of each averaged waveform was normalized to 1 for visualization purposes only (see Figure S4 for non-normalized waveforms). Shaded area represents the standard error of the mean between contacts. Colored vertical lines indicate onset and offset latencies for each VOTC region with shaded horizontal bar around each line representing a 95% confidence interval (percentile bootstrap). B. Timing parameters split by VOTC region and hemisphere: onset (top left) and offset (top right) latency of face-selective response, correlation of time-series between regions (bottom left) and between-region area under the curve (AUC) overlap (bottom right). For the AUC overlap, the arrow shows the directionality of the computed overlap. For instance, the arrow from ATL to OCC indicates that we are describing the percentage of the total AUC of ATL (measured between onset and offset latencies) occupied by the AUC of its temporal overlap (also determined using onset and offset) with OCC. Acronyms: LH: left hemisphere, RH: right hemisphere.

Concurrent functional connectivity between face-selective regions.

(A) Group-averaged Pearson’s correlations between single-trials face-selective amplitude measured at left IOG and left latFG computed across time, representing functional connectivity between these 2 regions. Black contour lines indicate significant positive correlations (p < 0.01, fdr-corrected). Correlations were computed across −150 to 150 ms lags between regions to infer direction of connectivity. The black dashed diagonal line represents a 0 ms time-lag between regions. Correlations centered above the diagonal would indicate that face-selective activity in latFG correlates with but precedes activity in IOG, suggesting information flow from latFG to IOG, and the reverse for correlations centered below the diagonal. B-F: functional connectivity between pairs of face-selective regions. Same conventions as in panel A.

Mapping concurrent face-selective response onset latency in VOTC.

A. Face-selective onset latency map across VOTC. B. Variation of face-selective response latency along the postero-anterior axis. Each data point represents the onset latency measured from the time-series averaged over contacts collapsed across the medio-lateral X dimension in each hemisphere within 20 mm segments (in the Y dimension). Thick lines are estimated onset latencies and shaded areas show the 99% confidence intervals expected under the null hypothesis that the postero-anterior location has no influence on the onset latency. Dark blue (full lines) is used for latencies computed when excluding contacts with SHC < 0.4 (as in panel A) and gray (dashed lines) are for latencies computed when excluding contacts with SHC < 0.7. Green line and shaded area show the expected increase in response onset latency based on simple conduction delays(van Blooijs et al., 2023) due to increasing distance from the occipital region, with reference to the latency averaged over the 2 most posterior face-selective response measured here. The dotted line is the mean expected conduction velocity for direct cortico-cortical connections (∼3.2 m/s) and the shaded area represents the minimum (1.7 m/s) and maximum (5 m/s) expected conduction velocity (i.e., based on measurements from other brain regions (van Blooijs et al., 2023)).

Number of contacts showing significant face-selective response increase in each anatomical region.

The corresponding number of participants in which these contacts were found is indicated in parenthesis. Acronyms: VMO: ventro-medial occipital cortex; IOG: inferior occipital gyrus; medFG: medial fusiform gyrus and collateral sulcus; latFG: lateral fusiform gyrus and occipito-temporal sulcus; MTG/ITG: the inferior and middle temporal gyri; antCoS: anterior collateral sulcus; antOTS: anterior occipito-temporal sulcus; antFG: anterior fusiform gyrus; antMTG/ITG: anterior middle and inferior temporal gyri.

VOTC anatomical parcellation scheme and face-selective recording contacts anatomical labels.

A. Anatomical regions were defined in each individual hemisphere according to major anatomical landmarks. The ventral temporal sulci (COS, OTS, and midfusiform sulcus, i.e., MFS) serve as medial/lateral borders of regions, whereas three coronal reference planes containing anatomical landmarks (posterior tip of the hippocampus, i.e., HIP, anterior tip of the parieto-occipital sulcus, i.e., POS, limen insulae) serve as an anterior/posterior boundary for each region. We considered contacts in the ATL if they were located anteriorly to the posterior tip of the hippocamps and posteriorly to the limen insulae. The schematic locations of these anatomical structures are shown on a reconstructed cortical surface of the Colin27 brain. Acronyms: TP: temporal pole; ATL: anterior temporal lobe; PTL: posterior temporal lobe; OCC: occipital lobe; PHG: parahippocampal gyrus; CoS: collateral sulcus; FG: fusiform gyrus; ITG: inferior temporal gyrus; MTG: middle temporal gyrus; OTS: occipito-temporal sulcus; CS: calcarine sulcus; IOG: inferior occipital gyrus; LG: lingual gyrus; ant: anterior; lat: lateral; med: medial. B. Map of all face-selective recording contacts and displayed in the Talairach space. Each circle represents a single face-selective contact color-coded according to its anatomical location in the original individual anatomy (see legend on the right).

Parameters for timing analyses.

The timing of face-selective activity was characterized using 4 parameters: (1) latency of response onset and (2) offset, as well as (3) response overlap and (4) response correlation across VOTC regions. A. The first two parameters, onset and offset latencies of face-selective response, were computed on HFB time-series averaged by groups of contacts (left). Averaged HFB time-series were converted to Z-score values by subtracting the mean amplitude in the baseline window of the time-series (i.e. before face onset: [-0.166 to 0s]) and dividing by the standard deviation of the amplitude in the same baseline window (right). Z-score values were further converted to p-values and corrected for multiple comparison using FDR. Onset latency was defined as the first time-point at which the p < 0.05 (two-tailed, corresponding to a z-score of 1.96), for at least 30 ms. Offset latency was the first time bin (after the onset) at which the response was no longer significantly different from baseline. B. A third parameter estimated the overlap between time-series for pairs of regions (e.g. regions A and B). The overlap is asymmetrical and is calculated separately for region A and region B as the ratio between the area under the curve (AUC) of the overlap between regions A and region B (i.e. summing the amplitude between the maximum of onset A and B and the minimum of offset A and B) and the total AUC for region A (for overlap of region A to B) or for region B (for overlap of region B to A). C. As the fourth timing parameter, we computed Pearson correlations between the time-series of pairs of regions using data between 0 and 0.6 s relative to face onset (gray shaded area). This provided an estimate of the similarity between the response function of the two regions compared. The shape of the response is more similar between region A and B than between region A and C. Between region correlations were compared against within-region correlations.

non-normalized time-courses.

Time-domain face-selective HFB activity in contacts showing response increase, averaged by main VOTC region (OCC, PTL, ATL) and hemisphere. HFB time-series were filtered to remove the general visual response at 6 Hz and harmonics so only face-selective signal remains. Time-series are original non-normalized versions of what appears in Figure 4 of the main paper. Shaded area represents the standard error of the mean between contacts.

Response timing in signal-contacts.

A. Time-domain face-selective HFB activity in signal-contacts averaged by main VOTC region (OCC, PTL, ATL) and hemisphere. HFB time-series were filtered to remove the general visual response at 6 Hz and harmonics so only face-selective signal remains. The maximum amplitude of each averaged waveform was normalized to −1 for visualization purposes only (see Figure S3 for non-normalized waveforms). Shaded area represents the standard error of the mean between contacts. B. Timing parameters split by VOTC region and hemisphere: onset (top left) and offset (top right) latency of face-selective response, correlation of time-series between regions (bottom left) and between-region area under the curve (AUC) overlap (bottom right). For the AUC overlap, the arrow shows the directionality of the computed overlap. For instance, the arrow from ATL to OCC indicates that we are describing the percentage of the total AUC of ATL (measured between onset and offset latencies) occupied by the AUC of its temporal overlap (also determined using onset and offset) with OCC. Acronyms: LH: left hemisphere, RH: right hemisphere. Onset latencies were not statistically different across VOTC main regions in the left hemisphere (OCC: 95 ms, [67-118] ms; PTL: 116 ms, [95-126] ms; ATL: 122 ms, [81-200] ms; OCC vs. PTL: p = 0.31; OCC vs. ATL: p = 0.48; PTL vs. ATL: p = 0.97, two-tailed permutation test, fdr-corrected), latencies in the right hemisphere were significantly higher in the ATL compared to PTL and OCC (OCC: 93 ms, [71-114] ms; PTL: 118 ms, [99-124] ms; ATL: 161 ms, [122-188] ms; OCC vs. PTL: p = 0.29; OCC vs. ATL: p = 0.005; PTL vs. ATL: p = 0.043, fdr-corrected). Offset latencies in the left hemisphere were slightly, but significantly, later in the PTL (362 ms, [352-384] ms) compared to ATL (304 ms, [155-341 ms], p = 0.045), and to OCC (325 ms, [306-356 ms], p = 0.046). No difference in offset latencies were found across regions in the right hemisphere (p’s > 0.11). The temporal overlap between regions computed using AUC was on average of 94%, ranging from 82% (right OCC AUC occupied by the AUC of its temporal overlap with the right ATL) to 100% (right and left ATL in PTL, right and left ATL in OCC). Correlating time-series across regions reveals that the lowest correlation was between right OCC to ATL (Pearson’s r=0.85, also lowest overlap in the AUC measure) and the highest was between right PTL and ATL (r=0.95), indicating that time-series across regions share 72% to 90% of their variance. Comparing within-to between-region correlations revealed only a small borderline difference when comparing OCC to PTL in the left hemisphere (within OCC and within PTL: r=0.97 and 0.98 respectively, between OCC and PTL: 0.91, p = 0.05, one-tailed percentile boostrap test, fdr-corrected), and OCC to ATL in the right hemisphere (within OCC and within ATL: r=0.96 and 0.91 respectively, between OCC and ATL: r=0.83, p = 0.05).

Response timing in face-selective contacts in main VOTC face-selective regions.

Response timing in face-selective contacts showing response increase in main VOTC face-selective regions: Inferior Occipital Gyrus (IOG, (N=65), lateral Fusiform Gyrus and OTS (latFG, N=123) and antFG+ (anterior fusiform Gyrus, Anterior Occipitotemporal Sulcus, Anterior Collateral Sulcus, N=184). A. Time-domain face-selective HFB activity averaged by main face-selective region and hemisphere. HFB time-series were filtered to remove the general visual response at 6 Hz and harmonics so only face-selective signal remains. The maximum amplitude of each averaged waveform was normalized to 1 for visualization purposes only. Shaded area represents the standard error of the mean between contacts. B. Timing parameters split by VOTC region and hemisphere: onset (top left) and offset (top right) latency of face-selective response, correlation of time-series between regions (bottom left) and between-region area under the curve (AUC) overlap (bottom right). For the AUC overlap, the arrow shows the directionality of the computed overlap. For instance, the arrow from ATL to OCC indicates that we are describing the percentage of the total AUC of ATL (measured between onset and offset latencies) occupied by the AUC of its temporal overlap (also determined using onset and offset) with OCC. Acronyms: LH: left hemisphere, RH: right hemisphere. Onset latencies of face-selective activity were not significantly different across regions either in the left hemisphere (IOG: 124 ms, [96-136] ms; latFG: 116 ms, [104-126] ms; antFG+: 97 ms, [69-116] ms; ps > 0.43 for all comparisons, two-tailed permutation test, fdr-corrected, Figure S5B) or in the right hemisphere (IOG: 106 ms, [91-114] ms; latFG: 87 ms, [67-100] ms; antFG+: 106 ms, [93-138] ms; ps > 0.43 for all comparisons, two-tailed permutation test, fdr-corrected, Figure S5A). The lack of significant difference in onset latencies of face-selective activity was confirmed when collapsing responses across hemispheres, being virtually identical between core face-selective regions (IOG: 106 ms, [95-118] ms; latFG: 100 ms, [95-108] ms; antFG+: 99 ms, [81-110] ms; ps > 0.7 for all comparisons, two-tailed permutation test, fdr-corrected). Offset latencies increased from the IOG (left: 370 ms, 95% confidence interval: [347-413] ms; right: 391 ms, [349-425] ms), to the latFG (left: 431 ms, [372-481] ms; right: 456, [401-483] ms) and antFG+ (left: 491 ms, [448-514] ms; right: 532 ms, [456-620] ms) but were not significantly different, unlike when collapsing across the two hemispheres (IOG vs. latFG: p = 0.031; latFG vs. antFG+: p < 0.005; IOG vs. antFG+: p < 0.005). Despite the difference in offset latency, there was a large temporal overlap between regions (computed either using AUC, ranging from 87% of the right antFG+ AUC included within its temporal overlap with the IOG, to 99-100% of each region with more anterior regions, or using time-series correlations, ranging from r = 0.95 to 0.99) indicating that most of the neural response occurs simultaneously across these 3 face-selective regions.

VOTC map of face-selective onset latency in contacts showing face-selective response decrease.

A. Face-selective onset latency was computed in 20 x 20 mm voxels by averaging the time-series of the signal-contacts in each voxel before computing onset latency. To reduce noise in the latency estimates, contacts with split half correlation (SHC) lower than r=0.4 were rejected from this analysis. B. Variation of face-selective response latency in signal+ contacts along the postero-anterior axis. Each data point represent the onset latency measured from the time-series averaged over contacts collapsed across each hemisphere (i.e. collapsed across the medio-lateral x dimension) within 20 mm segments (in the y dimension). Thick lines are estimated onset latencies and shaded areas show the 99% confidence intervals expected under the null hypothesis that the postero-anterior location has no influence on the onset latency.

VOTC map of face-selective onset latency in high SNR contacts showing face-selective response increase.

VOTC map of face-selective onset latency in contacts showing face-selective response increase, computed when excluding contacts with split half correlation (SHC) lower than r=0.7 to further limit noise in onset latency estimates. A. Face-selective onset latency was computed in 20 x 20 mm voxels by averaging the time-series of the signal-contacts in each voxel before computing onset latency. To reduce noise in the latency estimates, contacts with split half correlation (SHC) lower than r=0.4 were rejected from this analysis. B. Variation of face-selective response latency in signal+ contacts along the postero-anterior axis. Each data point represent the onset latency measured from the time-series averaged over contacts collapsed across each hemisphere (i.e. collapsed across the medio-lateral x dimension) within 20 mm segments (in the y dimension). Thick lines are estimated onset latencies and shaded areas show the 99% confidence intervals expected under the null hypothesis that the postero-anterior location has no influence on the onset latency.