Understanding how brain networks have evolved within primates is a central issue in neuroscience, both for identifying what distinguishes the human brain and for ensuring the meaningful translation to humans of findings obtained in non-human primates. Most comparative studies conducted over the past decade rely on resting-state functional magnetic resonance imaging (rs-fMRI), a non-invasive approach that measures spontaneous low-frequency fluctuations of the BOLD (blood-oxygen-level-dependent) signal and reveals temporal correlations of activity between brain regions. This method makes it possible to compare the organization of functional connectivity from one species to another. A major methodological problem, however, affects these comparisons: recordings are generally performed in anesthetized macaques and in awake humans, yet anesthesia strongly alters rs-fMRI signals.
To directly assess the effect of the state of consciousness, the authors studied the same group of macaques under two conditions, under anesthesia and then in the awake state. They focused on a well-characterized network in the human brain, the lateral cingulo-frontal network, by analyzing functional connectivity between seed regions located in the cingulate sulcus and the lateral frontal areas.
In awake macaques, the results reveal an organization structured along the rostro-caudal axis. Rostral seeds in the cingulate sulcus showed stronger correlations with rostral lateral frontal areas than with caudal areas, whereas more caudal seeds showed the opposite pattern, with stronger correlations toward caudal areas than toward anterior areas. This reversed rostro-caudal functional gradient reproduces the organization already described in the human brain, arguing in favor of a functional connectivity organization conserved from macaque to human. Crucially, this gradient disappeared entirely under anesthesia.
This work thus establishes that the organization of the cingulo-frontal network in awake macaques resembles that observed in humans, but that this similarity is only observable in the waking state. The network therefore appears sensitive to anesthesia, which calls for considerable caution when comparing functional connectivity profiles between species studied in different states of consciousness. Beyond the evolutionary question, this finding highlights an important methodological limitation of comparative approaches relying on disparate recording conditions. The demonstration rests on a within-subject comparison, which allows the observed differences to be attributed solely to the state of consciousness rather than to interindividual variations, strengthening the robustness of the conclusion.