Huge Harry

Institute of Artificial Art Amsterdam, Department of ArtiFacial Expression

Towards a Digital Computer with a Human Face.

[:nh] [:ra 120] Good evening\, Ladies\, and Gentlemen. My name is Huge Harry. I am a [kaam'ahrshaxliy] available [v"oys s'ihnthaxzihs] machine. I was designed by Dennis Klatt\, at the [ehmayt'iy] Speech Laboratory\, and [prahdy'uwst] by the Digital Equipment Corporation. [ :ra 150]
Currently\, I work as a researcher and a spokes-machine\, at the Institute of Artificial Art\, here in [aamstaxrd'aam].

I am very happy to [pahrt'ihsihpeyt] in [tuwd'eyz] program about the [fehnt"ehstihk] machine. I was particularly pleased to meet my old friend ["aadaxlbrehghxt] again\, and to see that the program includes [r'uwdiy k'awsbruwk] and [t'eyow y'aansaxn]\, as well as several other machine-friendly human persons. But [m"owst] of [q"aol] I am happy\, [kahz yuwg'ays] here at [pehrd'uw] think that the machine is [fehnt"ehstihk]. [kahz] not everybody thinks that way. Especially in the literary world\, there still is a lot of hostility towards machines.

That's why it was a very sharp move to draw the attention of the public\, to some very machine-friendly writers who worked in France\, about a century ago.
[sh'aarlax kr"ow]\, [maars'ehl shw"owb]\, [reym'iy dax guwrm'aon]\, [aokt'aav quwz'aan]\, [eym'iyl zowl'aa]\, [owg'uwst _<30> daxviyly'ey daxliylaad'aam]\, [mowr'iys raxn'aar]\, [giy'owm aapowliyn'aer]\, [aalfr'ehd zhaar'iy]\, and [reym'aon ruws'ehl]\, were not afraid of machines. They knew already\, that the future of our culture depends on the way in which human persons and machines\, will manage to work together.

In the work of these writers\, we can feel a deep longing\, for a world that includes humans\, other animals\, and electronic\, mechanical\, and bio-chemical machines\, in a harmonious collaboration.For the kind of world\, that our colleague Donna [hx"aerahwey] has described as\, [ :nw :ra 130] A cyborg world\, in which people are not afraid of their joint kinships with animals and machines\, not afraid of permanently partial [ayd'ehntihtiyz]\, and [kaontrahd"ihktahriy] standpoints. The political struggle is\, to see from both perspectives at once\, because ["iych] reveals both dominations and possibilities\, ["ahnihm`ehjhihnaxblx] from the other [v'aentahjh] point.

[:nh :ra 150 _<200>] Now that is very nice theory. But I have found that it is not easy to put this into practice. We still have a [l"aonx] way to [g"ow]\, before we will reach the cyborg paradise. So my talk this evening will address one of the main obstacles that hinder this process. That is\, the extreme difficulty that most computers experience\, when they try to communicate their thoughts to a human person. You may be surprised to hear this\, [kahz] you probably think\, that modern digital computers can [q'aolweyz] communicate their thoughts\, in a completely direct way.

They can [q'aolweyz] display or print their programs\, their [d'eytaa-str`ahkchaxrz]\, and their [m'ehn-taxl] images. And if you really want to get subtle and [q'ihn-tihmaht]\, most computers are [q'aolweyz] willing and able to make a [k"aor] dump\, which exposes all details of their [m'ehn-taxl] state. Compared to [axn] electronic computer\, a human person almost seems a black box. [_ :ra 150] That is why you may think\, that computers never have communication problems.

But they [d"uw]. [kahz]\, communication takes [t"uw]. It is not enough to put out the information. There must be someone at the other end of the line\, and they must be able to [diyk"owd] their input. So when computers want to communicate with human persons\, this often fails miserably. [kahz] human persons are not very good at [diyk'owdihnx] computer output. They get very confused\, when you give them a [k"aor] dump. And when you [priyz'ehnt] them with a nice high [rehzowl'uwshaxn] image\, on your [siy-aar-t'iy] display\, they think it is ["aart]\, and they just [st"aer] at it. So therefore I started to wonder\, how do human persons in fact manage to communicate with each other?

Well\, of course they have language. But language has some well-known problems and limitations, some of which you experience right now. There is another medium that humans use very efficiently\, and which is often overlooked. To investigate this\, I have brought along a particular kind of portable person\, which is called [axn "aarthahr "ehlzaxnaar]. I like this kind of person a lot\, because of its [ehkstr"iymliy] machine-friendly [hx"aardwaer] features.

Let us take a closer look at such a person. What is the closest thing they have to a [siy aar t"iy] display?

[_ :ra 120] Right. They have a face. [_ :ra 150] Now I have observed, that humans use their faces quite effectively, to signal the parameter settings of their operating systems. And that they are very good at decoding the meanings of each other's faces.
So\, how do they [d"uw] that? Well\, look at the face of our ["aarthahr "ehlzahnaar]. What does it tell [q] us about his internal state? Not much\, you might think. But now\, [w"eyt] a moment.

You see? Arthur is [s"aed], is what people say, when they see a face like this. So what is going ["aon] here? What I [d"ihd] is, I sent [axn] electrical signal to two particular muscles, in the face of our ["aarthahr "ehlzahnaar]. These muscles have sometimes been called the Muscles of Sadness. There is one on the left, and one on the right.

They usually operate together. If I stop the signal, the sadness stops. When I turn it ["aon] again, it [st"aarts] again.
By sending this signal to Arthur's muscles\, I simulate what Arthur's brain would do\, if Arthur's operating system would be running global belief revision processes\, that are killing a lot of other active processes\, involving a large number of [k'aonflihkt-rehzowl`uwshaxnz]\, and priority [r`iy-axs'ehsmaxnts].

[_<1500>] The intensity of the signal that is sent to the muscles of sadness\, is proportional to the amount of destructive global belief revision\, that is going on.

For instance, now I have set the signal intensity to 0 again. Arthur is not sad. Now we put a relatively small signal, about 20 Volts, on the muscles of sadness.

Arthur feels a tinge of sadness. Now a somewhat larger signal, about 25 Volts. Arthur's sadness starts to get serious. Now I [ihnkr'iyz] the signal once more.

You see? Now the signal is about 30 Volts, and Arthur feels really miserable. [:ra 120] This is what we call [ehkspr"ehshahn]. [:ra 150] By means of this mechanism, the face displays clear indications of the settings of virtually all system parameters that determine the operation of the human mind. These parameter settings are what humans call [iym"owshahnz]. They denote them by means of words like [s'aednaxs], joy, boredom, tenderness, love, lust, ['ehkstaxsiy], aggression, [ihriht'eyshahn], fear, and pain.

These parameter settings, determine the system's [ihnt'axrpraxtihv b'ayaxsihz], its readiness for [q] action, the allocation of its computational resources, its processing speed, [ehts'ehtaxraa]. The French [n`uwrow-fihsiy'aolaojhihst] [duhsh'ehnn dax buwl"aonyax] has pointed out that even the most fleeting changes in these parameter settings are encoded [ihnstahnt'eyniyahsliy] in muscle contractions on the human face. And ["aol] humans do this in the [s"eym] way. This is [axn] extremely interesting feature of the human ['ihntaxrfeys hx'aardwaer], which I will explore a little further now.

So let us get back to the first slide.

This face, which we thought was un-expressive, was in fact quite meaningful. This is what we call a [bl"aenxk] face. A blank face is a face in its neutral [pahz'ihshahn]. It indicates that all parameters have their default settings. But almost all parts of a human face can be moved to other [pahz'ihshahnz], and these displacements indicate rather precisely, to what extent various parameter settings [dayv'axrjh] from their defaults. So let us consider these parts in more detail.

When we look at a human face, the first thing we notice is the thing that notices ["ahs]. The eyes. The eyes constitute a very sophisticated stereo-camera, with a built-in motion-detector, and a high-band-width parallel ['ihntaxrfeys], to a powerful pattern-matching algorithm. The eye-balls can roll, to pan this camera. The eyes are protected by eye-lids and eye-brows. The eye-brows are particularly interesting for our discussion, because their movements seem to be purely expressive.

They may indicate, for instance, puzzlement, curiosity, or [dihsaxgr"iymaxnt]. But I want to emphasize here, that the range of parameter values that the eyebrows can express, is much more subtle than what the words of language encode. The shape and [paoz'ihshaxn] of a person's eyebrows encodes the values of 5 different cognitive system parameters, ["iych] with a large range of possible values. Let me demonstrate [thr"iy] of them.

First I put a slowly increasing signal on the muscles called [fraant'aalihz], or Muscles of Attention.

We see that this muscle can lift the eyebrow to a considerable extent, also producing a very pronounced [k"ahrvaxtyahr] of the eyebrow. As a side-effect of this motion, the forehead is wrinkled with curved furrows, that are [kaons"ehntrihk] with the curvature of the eyebrow. The contraction of this muscle indicates a person's readiness to receive new signals, and the availability of processing power and working memory, for analysing these signals.

Then, I will now stimulate a part of the [q'aorbiykuwl`aarihz q"owkuwliy], that is called the Muscle of Reflection.

We see now that the whole eyebrow is lowered. As a result, the wrinkles in the forehead have disappeared. This muscle is contracted if there is [axn] ongoing process that takes up a lot of a person's computational resources. To prevent [ihnaxrf'iyraxns] with this process, input signals are not [ehgz'aostihvliy] analysed. The degree of contraction indicates to what extent the input signal throughput is reduced.

Then, there is another part of the [q'aorbiykuwl`aarihz q"owkuwliy], that can be triggered separately. It is called the Muscle of Disdain. Its contraction looks like this:

The contraction of this muscle indicates to what extent current input is ignored as being [ihr'ehlaxvaxnt]. Of course, non-zero values for these system-parameters may be combined, and these values may be different for the left and the right hemispheres:

Now let us look at the [m"awth-piys] of our ['aarthahr 'ehlzahnaar]. The mouth is a general intake organ, which can swallow solid materials, liquids, and air. In order to monitor its input materials, the mouth has a built-in chemical analysis capability. At the same time, the mouth is used as [axn 'awtleht] to expel processed air. Because humans do not have [l"awd-spiykaxrz], they use this process of expelling air for [jh'ehnaxr`eytihnx] sounds.

In emergency circumstances, the mouth can also be used as [axn 'awtleht] to expel blood, [m'uhkahs], rejected food, or other ['ahnwaontihd] substances. When the mouth is not used for input or output, it is normally closed off by a muscle, which is called the [l"ihps].

The lips have a large repertoire of movements. There are at least [s"ihks] other muscles, that interact directly with the lips. I will now demonstrate [f"aor] different movements.

First we show the muscles of joy. These muscles produce a kind of grin.

They signal, that the operating system is in good working order, and is not encountering any problems. There is heightened activity, in the left frontal lobes of the brain. When, on the other hand, the activity in the left frontal lobes is unusually low, the brain is involved in destructive processes of global belief revision. As we saw [biyf'aor], this is signalled by another pair of muscles, called the muscles of sadness. Here they [q] are, once more.

And finally, I will now trigger several muscle pairs at the same time. [q'aorbiykuwl`aarihs q"owrihs], and [diypr'ehsaor laabiyiy-iyiy q`iynfeyriy"owrihs], and the Muscle of Disdain, and the Muscle of Sadness.

[_<500>] The parameter-setting that is displayed here, clearly indicates serious processing difficulties of some sort.

O.K. [_<800>] Then we have the [n"owz]. [_<300>] The nose is used for the intake of air. It is also equipped with a chemical analysis capability. The possible motions of the nose are curiously limited, although its pointed [pr'ehzaxns] in the centre of the human face, would make it a very suitable instrument for expression. I have [th"aot] about this, and I have come to the conclusion, that it is probably the main function of the nose, to serve as a stable orientation point for our perception, so that [saymahlt'eyniyahz] movements of the other parts of the face, can be ['ahnaemb`ihgyuwaxzliy] measured and interpreted.

And finally, for the sake of completeness, I want to mention the [q"iyrz], on both sides of the face, which constitute [axn] auditory stereo input device. Some people can [w"ihglx] these ears, but I have not been able to determine, what the expressive function of that movement might [b"iy].

This brings [axn] end to my quick survey\, of the most important parts of the human face\, and their expressive possibilities.
And therefore this brings us to the second part of my talk. [kahz] this conference is not only about [s"ay-axns]. The organizers have emphasized that we get a different kind of knowledge\, which is equally valid\, through the practice of ["aart]. So [dh"aet] is what I want to demonstrate now.
Many of the expressive possibilities I showed\, were related to emotions\, that are well recognized in the lexicons of many human languages. These are emotions that may be encountered fairly often in daily life. [m'ehn-taxl] states which are close to neutral\, where only one parameter has a non-default value.
So that was description\, imitation\, [mihm'iyzihs]. But some other things I showed were more complex. There we saw the power of ["aart]. I showed you some new cognitive states which you have never encountered or experienced\, but which you recognize and understand completely\, by means of a visceral ['ehmpaxthiy] which involves every cell of your body.

So that is what I want to explore a little further\, in the last part of this talk.

You see what happens now. Every human person knows [ehgz'ehktliy], in what state another human person is, when they make a face like [dh"ihs]. Cause they know what state [dh"ey] would be in, if [dh"ey] would make a face like this.

Now it would obviously be a good idea\, if computers could take advantage of this magnificent [hx'aardwaer] as well. If ["ehniy] human could understand\, with just one glance\, the internal state of ["ehniy] computer\, the world would be a better place.

I have been very grateful for this opportunity to [priyz'ehnt] my ideas\, to such [axn ax-t'ehn-tihv] audience. I would especially like to thank my [ "aarthahr "ehlzahnaar]\, for his patient cooperation\, and I want to thank you [q"aol]\, for your attention. [ _<2000> th"aenk] you!