Revision of “Warm Faces” Paper Submitted

We have revised our “warm faces” paper – led by Mario Messer – and resubmitted our manuscript. The manuscript can be downloaded for free here, and abstract pasted below.


Humans rely on temperature cues to navigate and make sense of their social world. In two experiments we find that part of this process can be detected in the perception and memory of faces. We conducted two studies to find this effect through a method called “reverse correlation”, where we average a face over a large amount of trials. For participants who experienced physical warmth (vs. cold) at the initial confrontation with a face we found that the average of the trials displayed a psychologically warmer face (judged by independent raters). Excluding the possibility of a “carry-over” effect, in Study 2 we replicated the Study 1 effect, but only if physical warmth (vs. cold) preceded the face. We conclude that the effects are due to biased encoding rather than retrieval of the face. A meta-analysis across these two and two unpublished studies (available at our project website on the Open Science Framework) establishes the strength of the temperature manipulation as another boundary condition. These findings speak to the conceptual validity of heatedly debated studies on social thermoregulation, while at the same time providing a theoretical advance through showing boundary conditions of our effect.