Revisiting the effective connectivity within the distributed cortical network for face perception

Effects of Face Repetition on Ventral Visual Stream Connectivity using Dynamic Causal Modelling of fMRI data

“…While our 6-ROI results were generally comparable with our 3-ROI results, the striking differences between our 2-ROI and 3-ROI results highlight the importance of the model architecture when testing hypotheses with DCM. While our results show that the addition of an EVC region has important effects on the conclusions one draws, it is possible that the results would change again if other regions were added, like anterior temporal lobes, amygdala (Xiu et al, 2015), or in particular superior temporal sulcus (STS), which is wellknown to have face-responsive neurons (Furl et al, 2015;Kessler et al, 2021;Lohse et al, 2016). Indeed, using an exhaustive data-driven approach called "Group Iterative Multiple Model Estimation" (Gates & Molenaar, 2012), Elbich et al (2019) found that connections from STS to OFA/FFA were also modulated by faces.…”

“…While superior temporal sulcus (STS) is often associated with face processing (Babo-Rebelo et al, 2022; Haxby et al, 2000) and has been included in some previous DCM analyses (Fairhall & Ishai, 2007; Kessler et al, 2021), we did not find it in our group-level univariate results and so did not include it in the present analysis. However it should be kept in mind that some of the present effects in OFA and/or FFA could emerge from interactions with STS (or other regions like ATL; see Introduction), which could be investigated in future studies.…”

“…The first DCM study of face processing (Fairhall & Ishai, 2007) found that face perception modulated connections from OFA to FFA, favouring the conventional feedforward, hierarchical view. A more recent meta-analysis of four DCM fMRI experiments (Kessler et al, 2021) replicated the increased "forward" connectivity from OFA to FFA for faces, but also found more negative "backward" connectivity from FFA to OFA. However, neither of these studies allowed for input from EVC to both ROIs, which would allow, for example, face-related input to FFA that bypasses OFA, as suggested by the patient fMRI data of Rossion and colleagues.…”

“…The previous meta-analysis by Kessler et al (2021) made inferences about individual parameters (connections) in their DCM models of a face network. However, this parameter-level inference ignores potential covariances between the posterior estimates of those parameters, which limits their reproducibility and interpretability (Rowe et al, 2010).…”

Perception of Faces and Other Progressively Higher-Order Properties

“…Yet the consensus view in cognitive neuroscience is that a sub-region in the highorder visual cortex, labelled the fusiform face area (FFA; see Fig. 6), plays a critical role in the processing of facial properties, exerting the doming influence on the core and extended networks of facial regions (Kessler et al 2021). Neuroimaging studies have shown that this high-order visual region responds preferentially to facial properties in a variety of tasks and paradigms (for review, see Kanwisher and Yovel 2006).…”