Tethered sensors from BIOPAC collect pulse, conductance, and temperature information from a participant's fingers.

Physiological Signals of Autistic Children Can be Useful

Physiological Signals of Autistic Children Can be Useful

By Karla Conn Welch

NOTE: This is an abstract of the entire article, which appeared in the February 2012 issue of the IEEE Instrumentation & Measurement Magazine.
Click here to read the entire article.

Three commonly used signals (Galvanic Skin Response, Heart Rate, and Body Temperature) are relatively easy to instrument in wearable clothing. The article discusses the sensors for these signals and research into their use in studying the emotional states of children with autism.

Individuals with autism are characterized as having difficulties with social interaction and communication and a tendency to fixate on limited interests and repetitive behaviors. The symptoms can range from mild to severe in degree, which is why autism is generally described as autism spectrum disorders, or ASD. Also, there are many challenges that come with living with an autistic person and for the person with autism. Monitoring physiological signals as indicators of affect can provide objective feedback about an important part of social interaction – understanding emotional cues.

This article covers the latest research concerning the measurement of physiological signals of children with autism, particularly for the study of changing emotions in various environments. Answers to important questions regarding autistic children’s physiological activity are examined, and we will see that within a non-social environment, physiological responses are the same between children with and without autism but different in environments with social contexts. Moreover, physiological signals can be used as a reliable indicator of emotions of children with autism. Also covered are the latest developments in wearable sensor technologies avail- able for measuring on-the-go. The author reviews additional research that identifies body signals in response to stimuli and may help explain core social deficits in children with autism.

Galvanic skin response, heart rate, and body temperature have good potential for being streamlined into wearable devices embedded within clothing for possible long-term monitoring beyond lab settings. Galvanic skin response (GSR), also known as skin conductance, is a commonly measured signal when observing emotional responses. The sensors are usually positioned on the fingertips of a participant’s non-dominant hand, and the Velcro strap wraps around the finger to secure it in place. (See figure below.) Any two of the pointer, middle, or ring fingers are acceptable. Although the fingertips generally produce strong readings, applicable locations for GSR measurement also include the middle phalanges (non-knuckle bone segments of fingers) of the pointer, the middle, or the ring finger; the palm of the hand; the inside of the wrist; and even the toes and the soles of the feet.

Tethered sensors from BIOPAC collect pulse, conductance, and temperature information from a participant's fingers.
Tethered sensors from BIOPAC collect pulse, conductance, and temperature information from a participant’s fingers.

Heart rate data can also be collected from a sensor placed on a fingertip. When using a photoplethysmograph (PPG) to monitor the changes in light absorption in the skin of a fingertip, pulse information can be extracted. Skin temperature can be collected from multiple locations such as the fingertip, wrist, or even the nose. Decisions on sensor placement depend heavily on the activity in which a participant will be engag- ing and their personal tolerance level. For example, placing sensors on a participant’s non-dominant hand makes for easy attachment, is usually comfortable during an activity, and frees up the dominant hand for interaction.

Monitoring physiological signals is a way to measure emotional information that may not be apparent. Children with ASD often react outwardly in ways unlike developmentally typical children, so tailoring the measurement and interpretation of physiological signals to children with autism is necessary. For example, children with ASD might smile when they are actually in pain. They might show no expression or a neutral expression when they are enjoying an activity. The article describes further the observed differences and how they relate to verbal reports made by children with ASD.

In addition to helping researchers and other people understand the emotional state of children with ASD, the measurement of physiological signals might help children with ASD understand their own emotions better. Although children with ASD may react differently to certain environments (e.g., a markedly negative reaction to eye contact when children without ASD may have no reaction or even a positive reaction as a sign of interest or a more negative reaction to an invasion of personal space than typical children have), the measured signal can be used as part of a feedback system for children with ASD to explain what emotions they are feeling. Producing a real-time affect-sensitive system to relay messages based on physiology-monitoring sensors would make a significant impact on ASD intervention, and could help therapists and caregivers in other ways.

The author describes research on a robot mechanism that is able to identify a child’s emotional state and react to it, in order to promote a learning environment which bolstered positive feelings. The autonomous robot’s prediction of the child’s emotional state (enjoyment vs. dislike) closely matched the clinician’s report on emotion for all six participants in the study.

Physiological signals can aid in improving social interactions and interpretations that even typical individuals sometimes have difficulty navigating successfully. Add to social situations a population that has performance-impairing difficulties interpreting these situations, and you will see how these signals could be vital feedback about what the individual with ASD is feeling and what others around him or her are feeling. Therefore, measuring physiological signals can help all individuals better understand the emotional world around them, and this insight could be especially informative for anyone challenged with accurately understanding affective information, like the ASD population. We will always be social creatures, so emotionally-informative technology might play a pivotal role in our next cultural evolution.


Karla Conn Welch (karla.welch@louisville.edu) is an Assistant Professor in the Electrical and Computer Engineering Department at the University of Louisville. Her research lab explores physiological signal processing, machine learning, and human-machine interaction for individuals with and without autism. She is a member of IEEE and the International Society for Autism Research (INSAR).