Heart Rate Variability and Brain Function During Emotional Face and Voice Processing in Autism

Background: Heart rate variability (HRV) refers to the beat-to-beat alterations in heart rate.  HRV is predominantly mediated by parasymphathetic influences on the heart. In general, the greater the HRV, the stronger the parasympathetic influence, or vagal tone.  HRV can be affected by acute stress or mental load.  Central nervous system regulation of HRV is believed to play an important role in the relationship between psychological state (e.g. stress, mental load) and acute changes in HRV.

Objectives: The aim of this study was to investigate HRV during rest and during multisensory integration of visual and auditory emotional cues and it’s relation to underlying brain activation patterns associated with these processes in individuals with autism versus typically developing individuals. 
Methods: A sample of 17 male and 6 female (age:  M = 15.5, SD = 4.9) individuals with a diagnosis of autism spectrum disorder (ASD) participated in the study.  A sample of 17 male and 6 female (age: M = 13.26, SD = 3.91) typically developing (TD) individuals served a comparison group.  Pulse oximetery and brain functional images were acquired while participants performed an event related facial emotion discrimination task.  Images of emotional human faces and audio clips of emotional voices were presented simultaneously in the MRI scanner.  Participants were asked to judge the emotional facial expression by pressing one of two buttons.

Results: The TD group performed significantly better (M = 97.4%) on the emotional face identification task compared to the ASD group (M = 87.8%; p = .016).  The ASD group had lower HRV during the faces plus voices task (M = 6.58, SD = 1.07) compared to the TD group (M = 7.43, SD = 0.93; p = .02).  The two groups did not differ in HRV during a 7 min resting phase nor did they differ in heart rate (HR) during rest or the task.  HRV was negatively associated with SCQ (used here as an index of autism severity) in the ASD group (r = -.709, p =.04) but not in the TD group (r = .38, p = .26; Z = -2.21, p = .027).  The TD group showed significant negative correlations between HRV and brain activation in the right anterior insula (r = -.63, p =.002), right amygdala (r = -.52, p = .01) and left fusiform gyrus (r = -.53, p = .01) during the task.  These correlations were not found for the ASD group.

Conclusions: While the ASD groups had similar resting state HRV, the ASD group’s HRV did not change during the emotional face/voice task while the TD group showed an increase in HRV during the task.  HRV was associated with brain activation in predicted regions during the task for the TD group, but not for the ASD group.  These findings suggest differences in central and peripheral nervous system integration during emotion processing in ASD vs. TD group.