Rett syndrome (RTT) is a neurodevelopmental disorder, including autonomic nervous system dysfunctions and severe communication impairment with an extremely limited ability to use verbal language. These individuals are therefore dependent on the capacity of caregivers to observe and interpret communicative signals, including emotional expressions. People in general, including therapists tend to focus on changes in facial expressions to interpret a person's emotional state or choices, but with this population it is difficult to know if the interpretations are correct. The aims of this study were to investigate if the Facial Action Coding System (FACS) could be used to identify facial expressions, and differentiate between those that expressed emotions and those that were elicited by abnormal brainstem activation in RTT. The sample comprised 29 participants with RTT and 11 children with a normal developmental pattern, exposed to six different musical stimuli during non-invasive registration of autonomic brainstem functions. The results indicate that FACS makes it possible both to identify facial expressions and to differentiate between those that stem from emotions and those caused by abnormal brainstem activation. This knowledge may be a great help to an uninitiated observer, who otherwise might incorrectly interpret the latter as an expression of emotion. (Contains 5 tables.)