International Meeting for Autism Research: Externalization and Interpretation of Autonomic Arousal In Teenagers Diagnosed with Autism In a Relaxation Experiment

Externalization and Interpretation of Autonomic Arousal In Teenagers Diagnosed with Autism In a Relaxation Experiment

Friday, May 13, 2011
Elizabeth Ballroom E-F and Lirenta Foyer Level 2 (Manchester Grand Hyatt)
9:00 AM
J. C. Lee1, M. S. Goodwin2 and R. W. Picard3, (1)The Media Laboratory, Massachusetts Institute of Technology, Cambridge, MA, (2)Media Lab, Massachusetts Institute of Technology, Cambridge, MA, (3)Massachusetts Institute of Technology, The Media Laboratory, Cambridge, MA

Heart rate (HR) is a robust physiological indicator of stress and arousal, and has been shown to have atypical ranges in school-age children diagnosed with autism (Goodwin, 2006). Progressive muscle relaxation is routinely used as a technique to decrease and manage stress and arousal. However, it is often difficult for teachers and caregivers to objectively determine a child with autism’s internal arousal state and thus to know if relaxation techniques are effective. Researchers in the MIT Media Lab have developed a wireless ear-mounted sensor capable of comfortably recording HR and displaying real-time data in various ways.


We wanted to assess HR using a wireless ear-mount sensor in a group of children diagnosed with autism while they participated in progressive muscle relaxation to see if it lowered their HR as it is expected to do.  We also experimentally manipulated teacher access to the data (i.e., no view, laptop display of HR in real-time, LED-vibrotactile device that pulsed in synch with real-time HR) to see how visualizing HR affects teacher’s appraisal of student arousal during the procedure.


7 participants (age 7-17, 5 male and 2 female) diagnosed with autism (via ADOS) enrolled in this MIT IRB-approved study involving an ABAB single case design of progressive muscle relaxation. In a quiet testing room, a familiar teacher led students through the following 7 phases: 0. Preparation (e.g. helped put ear-mounted sensors on student and began recording), 1. Baseline (2min), 2. Guided Relaxation with intervention 1 (2min), 3. Rest (1min), 4. Guided Relaxation with intervention 2 (2min), 5. Rest (1min), 6. Guided Relaxation with intervention 3 (2min), 7. Baseline (2min).

In the three guided relaxation phases the teacher experienced three intervention conditions, in randomized order: 1) no real-time data display, 2) screen-based display showing the student's real-time HR in beats-per-min (BPM), 3) physical tactile device that blinked and vibrated according to the student’s real-time BPM. Teachers also rated how behaviorally aroused the student appeared (1: very relaxed to 7: very aroused) after each experimental phase, and were interviewed after the study.

For data analysis, we first computed average HR in each phase per student. We then compared HR averages in each adjacent phase per student using two sample one-tail T-tests. Finally, we compared teacher ratings with each student’s per phase HR averages.


a.      5 out of 7 participants completed the study. The 2 who did not appeared to be tactile averse to the ear-mounted sensor.

b.     Comparing teachers’ ratings and students’ average HR per phases, we found 77% agreement overall when using the screen-based display, 46% agreement overall when using the tactile device, 58% agreement overall when no data was displayed.

c.      In several cases, HR did not decrease during guided relaxation.


Emerging physiological sensors can comfortably reveal arousal characteristics during guided relaxation exercises for some people on the autism spectrum. When viewed, it may help teachers more accurately assess their students’ internal state of arousal and enhance efforts to teach self-regulation.

| More