Copyright ©2013 NPR. For personal, noncommercial use only. See Terms of Use. For other uses, prior permission required.

DAVID GREENE, HOST:

On a Monday, it's MORNING EDITION from NPR News. I'm David Greene.

STEVE INSKEEP, HOST:

And I'm Steve Inskeep.

Today in Your Health, we'll hear from scientists who want to identify people at risk for depression. They want to do it using the power of computers.

GREENE: In particular, they're using a virtual therapist. This computer turns its cameras on patient analyzing how they move and searching for clues about what they're really feeling.

INSKEEP: Now, the virtual therapist has a name, Ellie. The NPR reporter who brings you Ellie's story is named Alix Spiegel.

ALIX SPIEGEL, BYLINE: Her hair is brown, and tied back into a professional looking pony tail. She wears a blue shirt, tan sweater and delicate gold chain. From the screen - where she lives out her life - she looks out at the man sitting across from her with curious eyes. He's a real man sitting in a real chair and this is the first time that they've talked. So Ellie starts simply, trying to build rapport.

ELLIE: So how are you doing today?

UNIDENTIFIED MAN: I'm doing well.

ELLIE: That's good. Where are you from originally?

UNIDENTIFIED MAN: I'm from Los Angeles.

ELLIE: Oh, I'm from L.A. myself.

SPIEGEL: Ellie is from L.A. She lives there in a computer at the University of Southern California, the creation of a group of scientists who believe that what they're doing will eventually revolutionize mental health care. They've been working on Ellie for months now, experimenting with giving her different personalities, programming her to ask provocative open ended questions. Laboring, says project manager Louis-Phillippe Morency over every element of her interactions.

LOUIS-PHILLIPPE MORENCY: Everything has been thought of. Just a simple what we call back channel like uh-huh, we have recorded more than 200 of these uh-huh.

ELLIE: When was the last time you felt really happy?

UNIDENTIFIED MAN: Uh, when was the last time.

ELLIE: Hmm.

UNIDENTIFIED MAN: Probably a couple months ago.

MORENCY: And these are so powerful, because a simple uh-huh and a silence, if they are done the right way - can be extremely powerful. So we spend a lot of time on these little details.

SPIEGEL: But Ellie isn't really intended to replace actual human beings as a therapist. Her real value - the reason she was created at all - is to help actual therapists diagnose patients because Ellie can measure people in a way that most humans can't. Underneath the wide screen, where Ellie's image sits, there are three devises: a video camera which tracks a person's face, a Microsoft Kinect that tracks their body movement, and a microphone which records their voice.

The point of these, says Skip Rizzo, a psychologist who is working on Ellie, is to analyze in almost microscopic detail the way that people talk and move, to read their body language.

SKIP RIZZO: We can look at the position of the head - the eye gaze and the head gaze - does the head tilt, does it lean forward, is it static and fixed.

SPIEGEL: In fact Ellie tracks and analyzes 60 different body and face and vocal movements. The theory of this is that a detailed analysis of those movements can give therapists new insight into their patients, because micro changes in the body, face and voice express things that words sometimes obscure.

RIZZO: You know, people are in a constant state of impression management. They've got their true self and the self that they want to project to the world. And we know that the body displays things that sometimes people try to keep contained.

SPIEGEL: And so, as Ellie gets the person in front of her to ruminate about when they were happy, when they were sad, the machines below her screen measure. Cataloguing how much the person smiles and for how long, how often they touch their head. Morency says they record 30 measurements a second .

MORENCY: Yeah, 30 times 60 seconds. So about 1,800 measurement per minute.

SPIEGEL: Literally every wince, pause and verbal stumble is captured.

ELLIE: What advice would you have given yourself 10 or 20 years ago?

UNIDENTIFIED WOMAN: Um, to, uh to not believe. Uh, to not be so gullible. To not be so gullible.

ELLIE: Hmm.

SPIEGEL: Now, Ellie was commissioned by the military. After all the deployments in Iraq and Afghanistan, the military was seeing a lot of suicides and wanted to find a way to help military therapists catch those suicides before they happened. Soldiers don't always like to confess they're having problems but maybe their bodies would. This is why Ellie is being programmed to produce a report after each of her sessions; a kind of visual representation of the 60 different movements that she tracks.

MORENCY: For each indicator, we will display three things.

SPIEGEL: First the report will show the physical behavior of the person Ellie just interviewed - for instance, how many times they smiled and for how long. Then the report will show how much depressed people typically smile; and finally, how much healthy people typically smile. Essentially, says Morency...

MORENCY: It's a visualization of the persons behavior compare to a population of depressed and non-depressed.

SPIEGEL: If their physical behaviors are similar to someone who's depressed, the person will be flagged. And again, Morency insists it's not that Ellie will actually diagnose these people, she's just there, he says, to give insight by providing some objective measurements.

MORENCY: Think about it as a blood sample, you send a blood sample to the lab and you get the result. The person doing the diagnosis is still the clinician, but they use these objective measures to make the diagnosis.

SPIEGEL: Now obviously, this work has all kinds of issues. Even on a practical level there are real obstacles. Jeff Cohen is a psychologist at the University of Pittsburg who researches the relationship between physical movements and emotion, and he says signals from the face, body and voice are incredibly complicated to interpret.

JACK COHEN: Individuals vary a lot in how expressive they are. You know, if I'm someone who's very expressive and I smile frequently; when I'm depressed and smiling less, I may still smile more than you do, you know, if you're a tight lipped, not very emotive individual.

KIRK SIEGLER, BYLINE: This means Cohen says, that using Ellie in the way blood tests are used - as proof positive of one emotional state or another - will be really difficult.

COHEN: It strikes me as unlikely that face or voice will provide that information with such certainty.

SPIEGEL: But Rizzo - the psychologist working on Ellie - genuinely believes these technologies will eventually change mental health. If you can collect enough data from enough humans, he says, you'll be able to see human behavior in a new way. One of the central problems with humans, he says, is that they bring their own biases to whatever they encounter, and those biases often make it hard for them to see what's directly in front of them.

RIZZO: You can get training to be a health care provider or psychologists and try to put those things on hold and be very objective. But it's still a challenge. It's always going to be biased by our past experience. So what computers offer is the ability to look at massive amounts of data and begin to look for patterns and that I think far outstrips the mere mortal brain.

SPIEGEL: So are you saying that the computer can see people in a more objective way?

RIZZO: Yes. I would say that you're objectively picking up physical events in the world. And the ability to do that on such a grand comprehensive scale with a massive database of past observations, I think you can pull information out of that by using various methodologies in computing - that goes beyond a human brain.

SPIEGEL: This summer Ellie's being tested. She scheduled to sit down with about a hundred veterans from Iraq and Afghanistan. She'll ask them about their lives, encourage them to open up, then silently, Ellie will measure their answers.

Alix Spiegel, NPR News, Washington.

(SOUNDBITE OF MUSIC)

INSKEEP: It's MORNING EDITION from NPR News.

Copyright © 2013 NPR. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to NPR. This transcript is provided for personal, noncommercial use only, pursuant to our Terms of Use. Any other use requires NPR's prior permission. Visit our permissions page for further information.

NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR's programming is the audio.

Comments

 

Please keep your community civil. All comments must follow the NPR.org Community rules and terms of use, and will be moderated prior to posting. NPR reserves the right to use the comments we receive, in whole or in part, and to use the commenter's name and location, in any medium. See also the Terms of Use, Privacy Policy and Community FAQ.

Support comes from: