The sensors developed initially at MIT have been improved upon and commercialized by the super team at Empatica.
Full disclosure: Picard is a shareholder and co-founder of Empatica, and part-time chief scientist, also serving on their board of directors.
Empatica sells a wearable autonomic and activity sensor optimized for clinical and medical researchers, EmbracePlus. Their wearable, platform, data and biomarkers are now cleared by FDA and used in clinical trials.
Empatica also makes a consumer-facing wearable, which provides clinical quality data and runs onboard AI to provide alerts for grand mal seizures, Embrace. Embrace is the only smart watch cleared by FDA for seizure monitoring. The watch also shows sleep parameters (e.g. sleep fragmentation and duration), steps, and provides raw data of temperature, electrodermal activity, and 3-axis acceleration for research purposes.
(Note that the above replaces the earlier "Q sensor". Here is a brief description of what happened to the Q Sensor.)
If you are a Media Lab Member, then please access the member website pages . Members also have access to all our pre-publication work and should feel free to contact us directly for any details, software, etc.
Whenever possible, we try to publish details about our work for the public, and list all of these on the AC Publications page. Years ago, Picard (with Profs. Pentland, Adelson, and Bobick) was part of the Vision and Modeling group, a computer vision research team, and together we amassed a large tech-report database including pdfs of all our published papers. All of our group's early papers are now managed by Pentland; they can be found occupying the early numbers of his current group's tech report database.
First, get into MIT.
Most of our students come from MIT's Program in Media Arts and Sciences (MAS), while others have come from Health Sciences and Technology or Electrical Engineering and Computer Science. We are open to students from any department.
We occasionally accept students from non-technical backgrounds (e.g., physics, design, psychology, cognitive science) as long as you are willing to learn to create and build new things (software or hardware). You have to apply separately to each department at MIT in which you might want to work. MIT's Admissions information can guide you to the information for each department.
If there is a hardship paying the application fee let the department you want to apply to know and we will try to help. Note that students have traditionally received full funding once they are admitted to MIT MAS (provided they remain in good academic standing and graduate in a reasonable amount of time), so cost of attending graduate school in our lab should not be a concern.
While there are no special advantages for any group, we especially encourage female and minority applicants to apply. We also encourage anyone who "doesn't fit" in the usual places. We are always looking for ways to expand the diverse viewpoints in our group, which gives us strength in thinking outside the box. If you have a tendency to underestimate your abilities, set aside your doubt and take the risk to apply.
What advice can we give to improve your chances to get in? The best thing to do is to first learn about our latest research projects before you write your statement of objectives. You can download publications, and contact the students and postdocs working on the projects that interest you most. Because of so many talented applicants, the admission decisions sometimes come down to choosing those whose passions and strengths are most closely aligned with current research goals. Learn what interests us and see if it is a good fit to your interests. Feel free also to suggest things you think we should be interested in. Great design, visualization, and communication skills are always a plus.
Professor Picard rarely accepts more than 1-2 new students a year and some years she accepts none. It is wise to target several areas of interest in your admissions application, to increase your chances of #1 getting into MIT, and #2 coming to our group. Once you are at MIT you will find there are few walls. Thus, you are encouraged to find 2 or 3 groups at MIT that interest you and apply broadly.
Picard personally goes through the applications to Media Arts and Sciences that mention Affective Computing or Autism, so list either of these if you want to consider working with this group. If you make the short-list during admissions, then Picard will plan to follow up with you by phone or by arranging a face-to-face meeting before final decisions are made.
We wish we had the resources to engage in email dialogue or otherwise interact with all applicants, their families, and others who email us on your behalf. Please do not expect personal responses to all your mails; Prof. Picard gets over 1000 mails/week and has time to respond to only a small handful.
Each year many outstanding applicants are turned away, despite that they have enormous talent and we would love to have them all here. A rejection does not imply anything negative about your talent or ability. Sometimes people are just not a fit for the current research focus. Some try repeatedly and get in on a 2nd or 3rd attempt. The staff in the office of the Program in Media Arts and Sciences can be of help if you have questions about the process or the status of your application.
Faculty are not allowed to respond to individual queries about the status of an application, or why a student was or was not admitted. You will not help your application by mailing the faculty with such inquiries.
Note that sometimes Prof. Picard serves on student thesis committees outside of MIT as well as outside the USA, so it may be possible to collaborate even if you are at another school.
Unfortunately with new US Immigration and MIT regulations, we are ALWAYS REQUIRED to pay money for international visiting students, and some other categories of visitor, even if they bring their own money. We get an enormous number of these requests and apologize that we do not have the resources to respond personally to all of them.
Unfortunately, almost never. Our group receives almost weekly unsolicited postdoc requests and accepts about one person every 3 years. We apologize that we do not have the resources to personally respond to all these requests.
These sensors were never for sale. Due to our research needs and an overwhelming number of requests over the years we developed the iCalm sensor, an improvement over the galvactivator and the Handwave, which then became spun out into a commercial product, the Q Sensor. The Q Sensor was then made and sold by Affectiva and measured electrodermal activity as skin conductance (better quality than the galvanic skin response of the old galvactivator) and also measured activity and temperature on the surface of the skin. It is no longer sold, having been replaced by something much better available from Empatica, which measures a higher quality version of the data and provides analytics and applications for health. A brief description of what happened to the Q Sensor and how to find out more is here.
We have a data share page that contains links to data sets. This page includes instructions for proper citations for the work.
There is no simple answer to this; it is an active research question that depends on many factors. "Best" is always with respect to some criteria: recognition accuracy, comfort and ease of sensing in a given situation, respect for user's privacy concerns, and more. For several examples, look at our publications.
Research at the MIT Media Lab is funded by sponsoring corporations, governments, and generous individuals. Our group shares ideas and prototypes with these companies and they take some of the ideas into development.
Sometimes spin-out companies develop our ideas. Prof. Picard has co-founded two spin-out companies, Affectiva, making video-based analytics software for emotion measurement and communication (mostly focused on facial and vocal affect analysis) and Empatica, focused on wearable sensors and their analytics for activity, stress, sleep, seizures, viral infection monitoring and supporting clinical studies for many kinds of medical conditions.
The MIT Media Lab is not in the business of making products.
That said, many products have been inspired by the work at the MIT Media Lab, and by our Affective Computing Research group. Some that have directly grown out of our group, include robots and software agents that sense and respond to a person's emotional expressions, new kinds of smartphone software that senses respiration and heart rate from your "motionless" subtle movements, software that helps people give better presentations, wearables and apps that help people manage stress, software that improves driver safety, and wearable/AI technology that detects and alerts to seizures and other important events. Our algorithms are also now being used in many commercial wearables.
Yes, if we have time (sorry if our schedule is already packed).
All press/interview/media requests are handled by the Media Lab Press Office.
Please contact them via the Contact Us form or email press@media.mit.edu.
Thank you for visiting this site. We apologize that we do not have time to respond individually to all the email we receive. We do value your comments and try to at least read most of the non-spam email. We hope that you will feel free to contact us at "affect" at "media dot mit dot edu" (making the usual substitutions) with additional comments or suggestions. Thanks!