Affective Computing: Teaching Machines About Emotion

jbc writes "The L.A. Times is running a story about affective computing, a field in which researchers are programming computers to recognize human emotions through the use of such clues as facial expression, vocal tone, and blood pressure. Some hail it as the dawn of a new era in super-useful machines, while others warn about invasions of privacy."

Comments

I can just see the possibilities:

[Bollie]

Microsoft Software: We Can Smell Your Fear!

i'll show you emotion...

[tps12]

I think my Athlon system knows all about emotion. It sure knows that the best way to piss me off is to have the hard drive make That Clicking Sound. The Sound can only be stopped through intense swearing.

Oh just great

[dscottj]

I try so hard to convince users that it really is just a lump of plastic and metal (when I can convince them the monitor isn't their computer), and now this?!?

Poker

[Tri0de]

This would make a computerized poker game MUCH more challanging; to say nothing about the possibilities of, er, interactive, adult entertainment.

The best use for this...

[bahtama]

Combine some of this, realdoll and the ability to recognize my emotions and I think we have something!

I come home from work and "Bambi" recognizes that I'm real tense and stressed out and knows what will help me unwind. Oh yeah, I'm liking this... ;)

Uh oh

[sandidge]

So... how would the computer respond to certain... um... "physical" signs of... uh... pleasure?

May configure it to "pop up" the "needed sites"?

Clippie v2

[peterdaly]

Clippie>> It looks like you are writing a letter.
You>>grr
Clippie>> You are frusterated, would you like my help?
You>>arrrg
Clippie>> I sense you need help, I have migrated your document into the letter template I think you want to you.
You>> stop!
Clippie>> Oh, you are done with your letter? Since you are having trouble, I have taken the liberty of saving and printing you letter for you.
You>> &*^@*(&#$_#(%*&
Clippie>> I sense how difficult this is for you, relax as I help you through the end of the letter writing process. Place an envelope in the printer to print the envelope to send you letter, that's all you have to do, see how easy this is?
...

I can't wait...

-Pete

Re:Clippie v2

[bpfinn]

I hoped it might go something like this:


Clippy>> "It looks like your writing a.... Oh, erm. Never mind". (*Clippy looks terrified and runs off-screen.*)

You>> "THAT'S RIGHT CLIPPY! RUN! RUN FOR YOUR LIFE!"

Hey, that could be fun.

If you're interested

[xmedar]

I'd suggest reading AffectiveComputing by Rosalind Picard from MIT Press, her homepage is here and interview on First Monday and the MIT homepage at MIT

affective?

[BlueFall]

I'd be happy with effective computing... ;)

Dating games, anyone?

[Dark Paladin]

This could be interesting for gamer types. Anyone ever play that current cop-shooting game, where the system has a body sensor that can tell if you're leaning/crouching so you can hide behind objects in a gun fight?

Now, take a dating sim like Sakura Taisen. Not only do you have to choose the right response to the question "Does this dress make me look fat?", but your facial response can have other effects.

For some games, this can be cool. Imagine an RPG where the look on your face determines your character's mood - and your response can then be read as humorous, sarcastic, serious, threatening - who knows. It will put real role playing on the computer into a new light, because you're doing more than reacting with the game, you're interacting.

Then again, the look on my face when I play FPS's look Quake is usually the same one I get when I'm sitting on the toilet, so that might not be a good thing....

Re:Dating games, anyone?

[JWhitlock]

Now, take a dating sim like Sakura Taisen [sakurataisen.com]. Not only do you have to choose the right response to the question "Does this dress make me look fat?", but your facial response can have other effects.

I just had visions of a dark future where geeks learn how to date from computers... and they get good at it!

What monstrosities will result from the union of geeks and cheerleaders?

Invasion of privacy...

[feloneous cat]

Computer A: Did you know about Jim? He is really getting teed off at me.
Comp. B: How can you tell?

Just what we need, computers that gossip...

Ironic

[jhines0042]

I think it is highly ironic that this article has a pop-under ad on it.

You want to get rid of these things? Stop linking to the sites that carry them.

Fully functional...

[bahtama]

I can't read this article without thinking about Data and Tasha in The Naked Now.

Tasha: You are fully functional, aren't you?
Data: Yes.
Tasha: How fully?
Data: I am programmed in multiple techniques of pleasure. (And can recognize your emotions, I'm the perfect man for you!)
Tasha: You jewel! That's exactly what I hoped.

Re:Fully functional...

[DarkHelmet]

A computer that senses emotions? Actually, this article describes what would happen if Data and Troi were to have a lovechild.

It could still happen... There's always Star Trek 11!

It's been done

[johnburton]

"Good news everyone! I've taught the toaster to feel love!"

- Professor Hubert J Farnsworth

MIT Media Lab Affective Computing Home Page

[Seth Finkelstein]

The MIT Media Lab has had a Affective Computing Research Group for a long time. Check out their home page at:

http://affect.media.mit.edu/AC_affect.html, and description

Affective computing is computing that relates to, arises from, or deliberately influences emotions. Our research focuses on creating personal computational systems endowed with the ability to sense, recognize and understand human emotions, together with the skills to respond in an intelligent, sensitive, and respectful manner toward the user and his/her emotions. We are also interested in the development of computers that aid in communicating human emotions, computers that assist and support people in development of their skills of social-emotional intelligence, and computers that "have" emotional mechanisms, as well as the intelligence and ethics to appropriately manage, express, and otherwise utilize these "emotions." Embracing the latter goal of "giving machines emotions" is perhaps the most controversial, and is based on a variety of scientific findings, which include indications that emotion plays a crucial role in enabling a resource-limited system to adapt intelligently to complex and unpredictable situations.

...

We understand that this research may involve gaining access to the emotional life of a person, including information that may be highly personal, intimate, and private. This work is inherently motivated by respect for human feelings, and therefore must respond with respect to a person's desire for privacy. Our default is to protect a person's privacy throughout our research, as well as in the tools we develop. We appreciate the potentially sensitive nature of our work, and feel strongly that the work we do adheres both to the highest ethical standards and the most fundamental human values. We made an effort to detail this policy.

Sig: What Happened To The Censorware Project (censorware.org)

Computer Emotion

[Yoda2]

Here are two good links on researchers trying to model emotion using computers.

Lola Cañamero's Emotion Page

Steve Allen's Home Page

No big deal...

[nonya]

Two comments:

First, let me state the obvious: There is a big difference between a
computer recognizing emotions and a computer having emotions. The
first problem is not hard to solve. It requires we identify a set of
features that can be used to recognize emotions ("phonemes of
emotional expression" from the article), and feed these features to
some sort of classifier. From a research standpoint, the interesting
part is finding the features that identify emotions. Once we find
those features "discovering" that a computer can recognize these
features is not surprising.

Second, there is some interesting problems in AI. Really! Knowledge
representation, vision, and language design are particular
interesting. But I get very, very angry at people who hype AI to way
beyond what it can do and/or do superficial projects like kismet (Rod
Brooks is good salesman, but he is not a scientist).

Just a year late....

[LittleGuy]

"Look Dave, I can see you're really upset about this.... I honestly think you ought to calm down; take a stress pill and think things over...."

{insert daisy.mid here}

hmm

[nomadic]

Of course, the best person to teach a computer to feel is still William Shatner. His powers also work on alien women, as long as they're attractive and don't wear much clothing.

Sure it has uses -- on the road

[ejaw5]

This could prevent would-be accidents caused by chronic road-rage drivers. If the system can sense anger/rage from facial and bodily expressions, and driving behaviors like sharp cornering and spontanious accelerating, it could try to calm the driver down by changing music or cooling the cabin. If those measures don't work, the system could then reduce the available power the engine gives out momentarilly. This could also stop a drunk driver from continuing to drive...

Oh Geez, It's 1969 all over again

[pmancini]

Computers that understand emotions, perhaps even feel them too. It will be the summer of love all over again. Hippy-freak computers will protest The New War, engage in "free love", 15 pin Mini-D connectors will find themselves enjoing the wierd possibiliites of parallel ports. Oh and lets not forget about the experimentation with drugs. Computers will seek the "White Rabbit" by dropping all of your mother's drugs (Microsoft Software) and trying better, more powerful ones. LSD = Latest Software Distribution. Some will tragically crash on pre 1.0 beta software. Others will literally have their memory blown on other psycadelic software upgrades. The phrase "If you remember the turn of the century you weren't there" will be common place.

"Turning On" will have a whole new meaning.

The MP3's will be great, if they can manage to foster the hippy utopia of a world with no RIAA.

Ah yes... but is Silicon Vally read to be the next Height Ashbury? Will Bill Gates disown his computer for letting it's cables get too long? Will Steve Jobs quit existentialism and realize "you don't need to be a weatherman to know which way the wind blows?"

And in the end, the love you take is equal to the love you make.

Re:Oh Geez, It's 1969 all over again

[pmancini]

Hey man, it was the Electric Kool Aid! How was I supposed to know it was a vat of Acid laced fruit drink? :-D

Man, I feel like Ozzy Osborne looks.

Re:Oh Geez, It's 1969 all over again

[Com2Kid]

Does this mean that Gainax would no longer have to draw their own Anime as their tripped out computer would be able to do it for them?

That would rock.

or at least be insanly scary.

Ooooh SHIIIIT!!!!!!!

The world CANNOT HANDLE that much Gainax-esque anime!

Hmm, how about Fractal generators? The new Lava Lamps?

Emotion is thought.

[kmellis]

From the article:

Movellan is part of a growing network of scientists working to disprove long-held assumptions that computers are, by nature, logical geniuses but emotional dunces. The ability to interpret markers for emotion--facial expressions, vocal tones and metabolic responses such as blood pressure--may seem like crude first steps. Yet experts see machine intelligence, unswayed by human frailty and bias, as an eventual advantage. They envision machines that know us better than we know ourselves.

The idea that consciousness could occur outside the context of emotion is a pernicious misconception. It arose from the combination of a greatly oversimplified view of thought and the legacy of dualism.

It is true that what we experience as "emotion" is a subtly different kind of cognition than what we experience as deliberate thought. It's more fundamental, and more closely tied to other physical systems. So I do agree that it makes a certain sort of sense to distinguish "thought" and "emotion". Ultimately, however, both are manifestations of the same fundamental brain activity. They are deeply related and are not in opposition.

We've been spectacularly bad at analysis of our own consciousness. History has shown that much of what we don't notice and so take completely for granted are fundamental and extremely difficult problems; while what we are very aware of and have concentrated upon have proven to be trivial. The predicate calculus, in this context, is trivial.

I've long railed against the cliche of the "unfeeling" thinking machine/being one sees in popular science fiction. Neither Spock nor Data would be able to carry on a meaningful conversation if their thought didn't exist within the context of emotion. The idea that a thinking machine could imitate human consciousness without including human emotion is absurd if examined carefully.

Be that as it may, "affective computing" is only a very minor addition to computing in the context of AI. It's just another form of data acquisition, albeit one that would no doubt be very useful for an AI. None of this stuff we hear about is even remotely close to actual AI; at best it's just "smarter" computing. Real AI will only be achieved when we are able to build (or more properly, "grow") very high-level complex adaptive systems aimed at complex human interaction.

Good website for them to research...

[Uttles]

This guy has a database of predefined emotions.

Thanks for Doing LA Times' Job For Them

[GuyMannDude]

I'd suggest reading AffectiveComputing by Rosalind Picard from MIT Press, her homepage is here [mit.edu] and interview on First Monday [firstmonday.dk] and the MIT homepage at MIT [mit.edu]

Thanks for posting this. For the LA Times' article, one gets the feeling that Movellan is leading a one-man renaissance in AI. Like most articles about far-future technologies, the article is heavy on the "gee-whiz" and "what will they think of next?" stuff and light on any sort of in-depth examination of the issues involved. First, I don't understand why the media (newspapers especially) don't take the time to do a thoughtful, in-depth story about non-time-critical issues like Affective Computing. Secondly, I wish that if they were going to do a half-assed job of it, they would at least cite other, more detailed sources of information so interested readers could learn more on their own. Yeah, I suppose someone can do a web search to find this out. And thank god for slashdot where the readers usually know more about the subject matter than the article authors. But it's common curtosy to cite important people in a scientific field. At least it is when writing a scientific paper -- why should the mass media be exempt from this little niceity? Suppose you were a researcher at MIT's program in this field and saw this article. Wouldn't you be kind of pissed off? The LA Times could have replaced that paragraph about the Golem with a paragraph about the MIT program.

I'm troubled by the slipshod coverage that science and technology gets in the mass media. Do the newspaper authors think we don't care to know the details? Do they themselves not care?

GMD

Nice to see

[Salsaman]

Somebody using the word 'affective' properly. No it is not the same as 'effective'.

Metrowerks debugger

[MouseR]

...knows about emotions, and has been using it for years.

When it senses that it has ticked you long enough, it knows it has to go. And it just does.

At that point, my emotion peaks, and suddenly, my pulse rate go down, as I wait for CodeWarrior to reopen my project.

Now, I can't say it's helping me much. But it sure is prudent enough, by quitting, to preserve my TiBook.

So many possible jokes...

[mttlg]

I would make some witty comment here involving the potential for computers to mimic the psychological behavior of girlfriends (most likely citing a specific instance of this behavior), except:

A: I really don't have much experience to use as a basis, and

B: A large portion of the audience probably doesn't either.

You'll just have to make up your own joke this time, you won't be getting a Score:5, Funny from me...

Marvin

[Luyseyal]

Marvin: "I've seen it --- it's rubbish."

Faking it in AI

[Animats]

I've noticed a tendency in the AI community to work on stuff like this when they're not making any progress on the real problems. Stanford's Knowledge System Lab was into this just before they tanked, and you can still see posters for some of the drama-related projects in the abandoned cubicles on the second floor of the Gates Building.

Like Eliza, systems that seem to have emotions generate responses from humans that cause them to be overestimated. Parry, which was developed in the 1960s along the lines of Eliza and simulated a dialog with a paranoid, was probably the first program to have "emotional state". So this isn't new.

Even something as simple as the Furby has that effect. (I'm not criticizing the Furby; I've met the designer, and he's just trying to make a toy kids like. He doesn't make any unreasonable claims for the toy.) It's a great way to get press coverage, because it yields good demos.

Dolls that fake emotions have been around for a while. The classic is Baby Think It Over, the attention-demanding doll from hell used to convince teenagers not to get pregnant. Hasbro marketed, as My Real Baby, a lower-cost (and less obnoxious) version designed by some of Rod Brooks' people from MIT.

And, of course, there are the Sims.

It doesn't take much internal state to fake emotions. It's typically just a few scalar values going up and down in response to inputs.

"Anger leads to hate..."

[antdude]

user getting frustrated with computers. :)

Good for usability tests

[iabervon]

The best use I know of for affective computing is in running usability tests. You have the person use your program, and monitor their emotional state. You can then determine how they actually feel about the interface without interfering by asking them questions while they're trying to use it, or missing things by only recording after the fact.

Door prize

[Tablizer]

Now bosses have justification to yell at something besides their employees.