When autistic children get stressed, they often don’t show it. Instead their tension might build until they have a meltdown, which can result in aggression toward others and even self-injury. Because autistic children often don’t understand or express their emotions, teachers and other caregivers can have a hard time anticipating and preventing meltdowns.
A new device developed by Affectiva, based in Waltham, Massachusetts, detects and records physiological signs of stress and excitement by measuring slight electrical changes in the skin. While researchers, doctors, and psychologists have long used this measurement–called skin conductance–in the lab or clinical setting, Affectiva’s Q Sensor is worn on a wristband and lets people keep track of stress during everyday activities. The Q Sensor stores or transmits a wearer’s stress levels throughout the day, giving doctors, caregivers, and patients themselves a new tool for observing reactions. Such data could provide an objective way to see and communicate what might be causing stress for a person, says Rosalind Picard, director of the Affective Computing Research Group at MIT and cofounder of Affectiva. She demonstrated the sensor last month at the Future of Health Technology Summit 2010 in Cambridge, Massachusetts.
“This certainly sounds like interesting technology,” says autism specialist Helen Tager-Flusberg, director of the Lab of Developmental Cognitive Neuroscience at Boston University. She says the sensors will need rigorous data proving their accuracy, but “the promise of new technologies like this may well improve our effectiveness to work with individuals with autism in daily life.”
When a person–autistic or not–experiences stress or enters a “flight or fight” mode, moisture collects under the skin (often leading to sweating) as a sympathetic nervous system response. This rising moisture makes the skin more electrically conductive. Skin conductance sensors send a tiny electrical pulse to one point of the skin and measure the strength of that signal at another point on the skin to detect its conductivity.
“When you see this flight-or-fight response, it doesn’t tell you it’s definitely stress, it just tells you something has changed,” says Picard. “Are they excited, hurting, are they stressed by a sound or person in the room? It doesn’t perfectly correspond to stress as it can also go up with anticipation and excitement, but when you see it change, you know something’s going on and you can look for the cause.” She adds that having clues to a person’s stress levels, which might not otherwise be detectable, could give caregivers and researchers more insight–and possibly a way to anticipate–the harmful behaviors of autism, such as head banging. Caregivers can try to identify and block sources of stress and learn what activities restore calm.
“I’ve been doing this for 25 years, and it’s one of the most exciting things I’ve seen,” says Kathy Roberts, founder and executive director of the Giant Steps School, an institute in Fairfield, Connecticut, for children with autism, many of whom are nonverbal and use assistive technologies, like the iPad’s touch screen, to communicate. The school has been using the Q Sensors for about six months to let therapists see which activities–such as relaxation techniques like breathing exercises–affect the well-being of individual students. Aside from having difficulty communicating, many of the students have trouble understanding their feelings. “Often students can’t really describe their internal state to us at all. What these sensors are allowing us to do is to have direct feedback, which allows us to see this internal state in a very concrete way,” Roberts says. She adds that the Q Sensor is much easier and less obtrusive for autistic students than sitting at a monitor for biofeedback, a traditional method for analyzing emotional states. Roberts believes that the sensors have the potential to reveal more about sleep–which troubles many autistic children–and could even provide early detection for seizures.
The beta version of the device will be available to researchers and educators in November for around $2,000, says Picard. She cautions that heightened skin conductance is not an absolute measurement of stress, because it applies to excitement as well as distress. She says the information needs to be evaluated in context. Additionally, stress levels can be hard to accurately detect when the wearer is taking medication or has attention deficit hyperactivity disorder or attention deficit disorder.
The Q Sensor can be worn as part of a wristband or a smaller module that can slip beneath a sweatband or baseball cap to make it discreet. After field testing, Picard’s company designed it for kids: the actual sensor–which is flat and a little less than four by four centimeters–can be wiped down, and the wristband itself can go in the washer. The device also has a temperature sensor to help correct for mistakes: it can tell, for example, when a user is entering a hot room rather than having an emotional reaction. It also has a clock, a rechargeable battery that lasts a day, an external button that lets a person put an event marker on the data, and a motion sensor that tracks movement in three directions. (It’s able to distinguish, for example, when you’re sitting or plummeting down a roller coaster.) To download data, a caregiver or user can plug the sensor into a PC or Mac with a USB and use software to view, compare, and annotate the data with descriptions of events during high- and low-stress periods.
Though Picard has largely focused on using the sensor with autistic children, a team at Children’s Hospital in Boston is using the Q sensors with epileptic patients in order to understand more about the onset of seizures. And a research group at Massachusetts General Hospital plans to place the sensors on babies to monitor normal growth of the autonomic nervous system.
Monica Werner, director of the Model Asperger Program at the Ivymount School, in Rockville, Maryland, for children with learning disabilities and autism, plans to use the Q Sensor to help second-grade through 10th-grade students better moderate their emotions. She hopes it can lead to more subtle methods for reducing a child’s stress, because some kinds of intervention can compound a child’s anxiety.
“The beauty of this is, it’s a way of providing feedback and intervention that is much less socially intense,” Werner says. The program will couple the sensor information with an app on the iPod Touch that lets students report on how they feel during a class. At the end of the day, the teachers will discuss the students’ reports and physiological signs with them to figure out what went wrong and how to better solve the problems. Eventually, she hopes to make use of what she calls “the holy grail”: real-time feedback, which Affectiva plans to enable in a later version of the device.
“It’s important for those of us who are therapists and teachers,” Werner says, “to know when to get in there.”