The hollow-face illusion has been used to study the dissociation between vision-for-perception and vision-for-action. In this experiment, people used their fingers to make a quick flicking movement at a small target attached to the inside surface of the hollow—but apparently normal—face, or on the surface of a normal protruding face. The idea was that the fast flicking would engage the vision-for-action networks in the dorsal stream and thus would be directed to the actual rather than the perceived position of the target. Despite the presence of a robust illusion in which people perceived the hollow face as if it were a normal protruding face, the flicking movements they made were accurately directed to the real, not illusory location of the target. This result suggests that the bottom-up cues that drive the flicking response are distinct from the top-down cues that drive the Hollow-Face illusion.
The hollow-face illusion is weaker among people with schizophrenia and other populations with psychotic symptoms, perhaps as a result of a reduced tendency to interpret any kind of ambiguous 3D object as convex. It appears to be related to current mental state, namely in regards to current positive symptoms, inappropriate effect and need for structure. The illusion seems to strengthen among successfully-treated patients.