Avoiding the Uncanny Valley

Note: View original article for images and video.

Avoiding the Uncanny Valley

 New research suggests that robots that look almost like people are unsettling.

By Bobby Azarian

Sharp unveils the humanoid robot-shaped smartphone RoboHon at a preview of Asia’s largest electronics trade show CEATEC in Chiba, Japan, on Oct. 6, 2015.

At 7½ inches tall, RoboHon is a walking, talking voice-controlled robot that can play games, dance, project visuals onto a wall—it’s even equipped with facial recognition software that allows it to get a feel for the mood you are in. The Japanese-made robot is set to hit shelves there in the first half of 2016, and it’s a clear sign that personal robots with social skills are on the horizon.

With a physical appearance that is loosely modeled after humans, RoboHon stands upright with two legs, two arms, wide-open eyes, and a welcoming facial expression. Its childlike voice and cheery demeanor instantly make you want to give it a big hug and be best friends forever. As more robots like RoboHon come into our everyday lives—as teachers, co-workers, aides for the sick and elderly, and even as companions—it is crucially important that they make humans feel comfortable interacting with them. If we fail to design robotic systems that can establish meaningful social bonds with people, then we will not be taking full advantage of the technology.

Hundreds of studies exploring ways to promote smooth human-robot interaction have been carried out in recent years. In general, scientists agree that as robots begin to appear more humanlike, people tend to respond to them more positively. These humanoid robots engage us better, since they can communicate in natural and familiar ways through social cues—like facial expressions, body postures, eye gaze shifts, and gestures. A machine with a lifeless appearance is much less inviting than one with a recognizable likeness, which implies familiar behavior.

However, the popular idea known as the uncanny valley suggests that there’s a problem with that approach. The hypothesis predicts that the positive relationship between a robot’s degree of human likeness and our affinity for it continues to grow only until a precise point. Specifically, when robots appear almost exactly human, people experience an unsettling feeling that causes revulsion. Something just feels “not quite right,” and the machine looks “creepy.” At this stage the positive relationship sharply turns negative, where it remains for a short period of time just before turning positive again when the robot starts to look completely indistinguishable from a human—a design feat yet to be fully achieved. (30 Rock has an excellent summary of the uncanny valley.)

But the uncanny valley hypothesis, as put forth in 1970 by Japanese robotics professor Masahiro Mori, is just a logical prediction and not the result of objective experimental testing. British art curator and writer Jasia Reichardt later described it further in her 1978 book Robots: Fact, Fiction, and Prediction. Since its birth, the uncanny valley has lacked the detailed descriptions and rigorous explanations that are customary to most scientific hypotheses. Nevertheless, the paradigm has heavily influenced robotics design for decades.

Now that we’re getting closer and closer to designing robots that look like humans, testing the hypothesis is more important than ever. If the uncanny valley is in fact real, then trying our best to make robots that mirror humans exactly is a big design no-no.

Scientific investigation into the uncanny valley didn’t really start until about 10 years ago, which is roughly when researchers developed the ability to design highly realistic humanoid robots. One of the earliest studies to formally challenge the hypothesis, carried out in 2005, proposed that negative reactions to humanlike robots are more related to good or bad design aesthetics, and can occur at any level of realism. In other words, highly realistic and unrealistic humanoid bots can both cause revulsion with certain physical features—like bad skin, sickly eyes, significant asymmetry, and poor grooming. Conversely, clear skin, symmetry, and sharp grooming accompanied by oversize eyes, smaller-than-usual noses, or very large smiles can still be seen as highly attractive despite being unrealistic, if the proper balance is achieved. The study did find an uncanny valley effect when participants looked at a set of morphed images that followed a continuum from unrealistic to very realistic anthropomorphic robots, but that effect disappeared when the same images were made more attractive. The researchers concluded that although the uncanny valley may exist, careful design practices could help overcome it. A number of subsequent studies have both supported and conflicted with these results, and a recent systematic review described the empirical evidence for the hypothesis as ambiguous.

Despite the history of inconsistent research, a recent study in the journal Cognition provides compelling evidence to support the uncanny valley claim. What makes it different than earlier work is that researchers Maya Mathur and David Reichling may have avoided some pitfalls that could have obscured results in prior studies. In order to test whether the uncanny valley effect occurs with real-life robots, the researchers used 80 pictures of social robots that have actually been built, rather than using computer-generated morphed blends of human and robot faces, which often have unnatural distortions that cause strange features. They were also very picky about a lot of other factors—for instance, they only chose robots that were meant to interact with users (that is, not missing hair, facial parts, skin, or clothing), not marketed as toys (which are often made to look “adorable”), and capable of physical movement.

In addition to rating how friendly each robot seemed, participants were asked to play an economic investment game with the bots to determine how much they trusted them. This is important because social trustworthiness is a big part of our willingness to interact with one another. The subjects were given up to $100 and were told to decide how much money to give to each robot in the hopes that they would receive a return on that investment.

The results showed a strong uncanny valley effect with both measures. Specifically, as robot faces appeared increasingly more human, their likability ratings increased up until they looked almost human, at which point the ratings dropped significantly, dipping down into the valley. Similarly, the amount of money wagered by participants first increased before drastically dropping, only to increase again when robots began to look identical to humans.

These findings have important consequences, not just for engineers and scientists, but also for anyone who wants to see robots come into our daily lives. Giving robots expressive physical features that allow them to communicate through social cues will make them more trustworthy, persuasive, and fun to be around.

At the same time, the results show that we must take careful caution not to fall into the uncanny valley by making humanoids that are too lifelike for comfort. Minor flaws and imperfections in appearance can give us a feeling of seeing dead matter impersonating humans, much like watching a zombie. This can elicit fear or disgust while reminding us of our own immortality. Fortunately, past research suggests that we can mitigate these effects by using good design practices that ensure that robots look as attractive and friendly as possible.

You might look at the new study’s data and conclude that we have already solved the uncanny valley design problem, since the most realistic robots reversed the valley with their high likability ratings. But the researchers used only static images of robots, so the participants couldn’t observe how those robots actually move. Since we are far, far away from being able to design machines that move realistically, even today’s most human-looking androids will likely still fall victim to the uncanny valley. Future studies that use dynamic videos of moving robots will be required to determine just how realistic robots would have to move to become likable once again.

We are social creatures, and as such require robots that are social as well. There is no doubt that getting their design completely right will require a lot of hard work and creative thinking. But with the many scientists out there currently picking apart these tough questions and applying that knowledge, we can be sure that we are well on our way to a society where interacting with cooperative, supportive, and incredibly productive robots is an everyday occurrence.

This article is part of Future Tense, a collaboration among Arizona State University, New America, and Slate. Future Tense explores the ways emerging technologies affect society, policy, and culture. To read more, visit the Future Tense blog and the Future Tense home page. You can also follow us on Twitter.

Copied from http://www.slate.com/articles/technology/future_tense/2015/10/research_supports_the_uncanny_valley_theory_of_human_robot_interaction.html on 03.30.17

(a thought) - Dick Higgins

(a thought) is reprinted from a letter from Dick Higgins, perhaps the most respected artist involved with the field. He reflects on his role in, and participation with, the mail art network.


a lot of what i do is mail art's literary cognate, as yet unnamed. postal verbal gestalt-making.

one has as idea. should it be written up in a technical format for the benefit of a few aestheticians or other specialists digested by professional pudgies or pudgy professionals, and - filed? that would be instant academicism.

or should one simply get it to known thinkers - and/or artists - wherever encountered. and let them use it as they see fit. in the way ideas can be tools, or can be enjoyed for their own sake. one needn't worry about the politics, copyrights or the personality of the sender. the focus is on the thing, not the creator or his act.

being relatively anonymous, it doesn't pay very well to do things that way. but there are other satisfactions. e.g., the word i revived (from coleridge) fourteen years ago, "intermedia", has gotten very widespread use and rather little abuse. i used it in a lecture first, and noticed it didn't get picked up. so i wrote it up in a newsletter as an aesthetic idea and sent out some 5000 free copies. that got picked up.

same with my cognitive/postcognitive idea, rather sher idea, and so on - they went out first as postcards. if only more philosophical people would do things like that, it would be like the academics were offering us back our brains! i mean, philosophy is really just the art of thought, isn't it? yet the prof's, the specialists and other pudgies insist on gluing it all down into academic (that is, into non-implicative) contexts and formats, with an excess of rationcination, etc. result: american philosophy is not philosophy since it is not the art of thought - rather is is just the subject matter which so-called american philosophers teach. that's too bad, since philosophy should be as much as an art as the others - as collage-making, mail art, art performances, wordsmithing and so on.

post wordsmithing - maybe that is what i mean.

some texts require the interchange of mail or other deliveries - mail poems. what can be more exciting that waiting for the next lion to arrive?

above copied from : CORRESPONDENCE ART: Source Book for the Network of International Postal Art Activity. Eds - M. Crane & M. Sofflet.

The unity of the kind artwork, Roberto Casati

(Translated from Italian by Marcel Lieberman)

It is not easy to identify in a precise way the role that works of art play in our cognitive life. Yet, works of art, like all artifacts, are essentially linked to our cognitive life. A study of the relationships between art and cognition is thus a necessary step towards understanding artistic phenomena and artifacts. A variety of possible outcomes can emerge out of the study of this interaction.

1. The study of artistic artifacts considered as cognitive products can give us access to mechanisms of the mind that go unnoticed in normal cognition.

2. The study of cognitive mechanisms that form the background to artistic practices can enable us to bring into focus certain philosophical problems, for example, the question regarding the "definition" of a work of art and its conditions of identity.

In this article I'll examine an additional problem, which is different from, and in some ways more ambitious than the one posed by works of art considered as an object of cognitive and philosophical study: namely, the question that in a certain sense precedes the examination of topics such as 1 and 2:

3. The study of cognitive activities allows one to clarify and eventually solve the problem of the unity of the kind work of art.

I take it for granted that there is a problem regarding the unity of the kind. Or rather, that there is a unity of the kind and, what is more, that it is a problematic one. Besides the clear linguistic indication ("work of art"), our attitudes towards symphonies, architectural works, films, paintings and sculptures are much more similar to one another than are the attitudes towards, say, a painting by Picasso and a family photo. But this is just the problem: what is it that entities as diverse as a Picasso and a symphony have in common that outweighs the numerous and undoubtedly greater resemblances between a painting and a family photo?

We can examine two types of solutions, radically different from one another, that are based on the study of cognitive mechanisms. The first, which won't be discussed in this paper, is a "circumscribed" solution and appeals to the idea that there exists an artistic faculty or pseudo-faculty that is activated every time we encounter objects considered works of art. This would explain why such objects, as dissimilar as they are, end up in a single category. The theory of the "pseudo-module" seems to have a certain explicative power while using a limited hypothesis. I won't examine its validity; I prefer instead to present another theory, one that is completely different and much broader. This hypothesis places artistic artifacts within a social dynamic. It is in virtue of their becoming becoming elements of such a dynamic that artifacts acquire the wholly extrinsic property of being artistic. Up to this point, the "broad" theory does not differ from a sociological study of art. It adds, however, a fundamental question: why is it that not all artifacts become part of a social dynamic that makes them artistic? The explanation of this dynamic is subject to cognitive constraints, and the study of such constraints can enable us to make a prediction regarding the properties of artistic artifacts.

A mistaken but widespread theory of art

In order to characterize the second, broad theory, let us take a brief look at those things that seem obvious and that generate dubious rationalizations. When one speaks of cognitive theories applied to art, often the only thing one has in mind is a diagram: in one box there is the artist's mind, in another the mind of an observer, which are connected by an arrow that splits in the middle to make room for a box containing the work of art. (I could draw the diagram below, but refuse to do so in order to avoid spreading it even further). This diagram rationalizes and perhaps only illustrates common-sense intuitions regarding the function of cognition and the fact that art might be a type of expression. Through the work, the artist supposedly expresses himself or sends a "message" to his audience. An artist has something "to say". And the audience must reconstruct what the artist meant: the audience's task is that of an interpreter who, by observing or listening to the work and on the basis of personal knowledge and other background factors, is able to read the artist's message.

The message theory is surely a cognitive theory. But it is faced with a number of problems.

The main problem that interests us is that it doesn't explain the unity of the kind "work of art" among its diverse manifestations, apart from attributing to architectural works and dance the task of transmitting messages. In connection to the first problem, it doesn't explain why works of art are admired by people who know little about the history of art, why they survive the test of time (how is it possible to admire works by inaccessible cultures, whose message can no longer be reconstructed?), it doesn't explain why artists like talking about their works and why they apply labels to them (what purpose would it serve, given that the works already express what the artists mean?). Moreover, given that the sender of the message might not have in mind any receiver, or not know who the receiver is, one ends up losing sight of the receiver himself. At the same time, the intentions of the majority of senders are inaccessible: whether because the artist is no longer living, or because no one is so transparent to themselves and it isn't clear that artists really know what they intended to say. The result is that the work, rather than transmitting the elusive intentions of an artist, ends up with the responsibility of expressing the "spirit of an age". Lastly, the message theory does not explain why the artist chose such an implausible way of sending "messages". Why hide messages in a medium that requires so much work on the part of the one receiving the message?

There is certainly some cognitive work at play. But it doesn't seem plausible that it's the one required by the message theory.

The theory of conversational prompts

The alternative theory could be called the theory of conversational prompts. The theory claims that artistic artifacts are objects produced with the chief aim of provoking some type of conversation about their production. Artistic products don't serve as a type of "communication" between the artist and the public: they are not bearers of "messages". Rather, they are objects that must attract attention (and thus must not be instrumental, or hide their instrumental side) within a linguistic context in which they are used as objects of discussion. I won't enter into further details, which might seem somewhat definitory: definitions are notoriously useless for understanding common-sense notions. I would like to show how this hypothesis can work by showing how it is set within a series of anthropological observations regarding the use of artistic artifacts.

The theory explains why artistic artifacts are able to survive through time (if one thinks about it, this survival is quite strange, and at any rate hardly compatible with the idea that artistic products contain a message). They pass the test of time because conversation never stops; it is always in need of topics. Even when it is no longer possible to know the terms of the conversation in which the product was initially inserted as a stimulus, it remains possible to recover the product within a new conversation. It must be noted that the theory does not say that the artist must form the intention of seeing his product placed within a specific conversation (which most probably is the general one of his time), but in any conversation. This fact imposes constraints on the structure of works of art. They are objects that must be able to lend themselves to conversation.

Similarly, the theory explains why works of art pass the test of space, or rather why they can be appreciated by communities that are quite distant from the original community of the work's creator.

The theory explains why artistic products have the aspect they have. Artistic products must solve a variety of problems

- maximize novelty

- attract attention (be sufficiently different from instrumental artifacts)

- be sufficiently complex (through their apparent form, or through the history of their origin) to maximize conversational elements

The theory explains the fluctuation in the esthetic and economic values of artistic products. Having good qualities does not suffice for being a good conversational prompt: there must also be a conversation in which such qualities can be noticed. By postulating the existence of conversations, the theory explains why artistic products survive, are fashionable subjects, and die. (Likewise, it is not enough for a metal to have excellent qualities: resistant to acid, malleable, yellow, in order for it to be of great worth. There also needs to be a context of exchange that confers value to the metal). In the same way, the theory explains the difference between great art and popular art, simply by postulating the existence of different conversations with different rules: among a myriad of conversations (that provide the basis for popular art) one proclaims itself "high". Fine art is nothing other than popular art with an army behind it.

The theory thus explains the existence of degrees of artistic quality, and why certain things are considered art by some and not by others. It explains why a local artistic culture finds the works of another culture of little interest, while recognizing that they are artistic artifacts.

The conversational theory explains the origin of art and artistic artifacts. There is no origin! Works of art were discovered: or rather, it was discovered that certain objects entered into circulation within a community and caused people to talk about them.

It explains why instrumental objects can be works of art (as in the case of architecture, which some esthetic purists seek to expunge from the category of art). The possibility of being inserted within a conversation doesn't seem to depend upon the type of object to be inserted.

The theory explains why artists like to talk about their work and adorn it with explanations (this is particularly difficult to explain in a theory of communication or expression). It is a way of launching a conversation that will give life to the product.

The theory explains why paintings have labels and musical pieces have titles: they are points of entry into a conversation. It therefore explains why museum visitors head straight towards the labels, and place great importance on knowing the author and the subject.

It explains why an artist's biography interests us; and it explains why we are satisfied by the fact that the biography is in some way reflected in the work. It enables us to use the work as a narrative prompt.

The theory explains why works of art are acquired with no regard for the artist, like invitations to a conversation that are disconnected from the person of the author.

Finally, the theory's hypothesis that artists produce works with an eye towards possible conversations about their products, allows us to solve almost immediately the problem regarding the unity of the kind work of art. Works of art are objects created with the chief aim of making a conversation possible. The main proviso is meta-representational: the author must have the intention that his work be a conversational prompt. The proviso excludes cases of artifacts that are accidentally, but not essentially, currency for conversational exchange, like mathematical theories or political discourses which are not works of art.

The conversation theory and the time of a work of art

The intention of creating a work of art doesn't focus on the moment of creation, nor the moment of reception, but has a projection into the future; it is focused on the theme of the conversation. The difference between the message theory and the conversational theory concerns a deep metaphysical aspect of works of art: their relation to time. The conversational theory is by nature projective and has an articulation that unfolds in the future. Works of art have an evolution, linked to the exercising of discussions that change in the course of time. The message theory is temporally static and concerns a fixed point in the past. In the message theory time essentially concerns the packaging of the message: the moment at which the author consigns the message to the work. An eventual interpretation constitutes an extrinsic aspect, and therefore does not introduce a new temporal element.

The intuition that guides the message theory is that works of art are like packages, wrappers ("vehicles", a "medium"). One needs to unpack them in order to reach their hidden essence, the message itself. The artist leaves the message in the work just as a castaway does with a message in a bottle. The precise methods of the unpacking process cannot be foreseen a priori by the artist who doesn't know on which beach the bottle will wash up, or whose eyes will read the message. For this reason, one must leave room in the message theory for the notion of an open work: the sender's intention would be to produce a message that is at least partially indeterminate, that partly constrains interpretation, but leaves space for the receiver. The notion of an open work is a clear case of theoretic artifacts generated by the message theory. Only if one thinks that works must necessarily be interpreted in order to extract the artist's intention, must one then give an open structure to such an intention, faced with the mutability of possible interpretive contexts. But the necessity of "recovering" the artist's intention doesn't exist, given the inexistence of such an intention. Naturally, the artist can have intentions, but these concern the use of the work and not its interpretation. Emotions, messages, authors' intentions for communicating a message, substitutions of experience: these are possible, but accidental, ingredients of the dynamic that leads to the production, and assures the circulation, of a work of art. Works of art are not signs. Rather, they are more like toys.

What is a conversation? Empirical hypotheses

The conversational theory makes use of a concept, namely conversation, which is certainly indeterminate. What is a conversation? Can there be inner conversations (meditation)? Which conversation does the artist have in mind, a specific or general conversation? In fact, the theory shifts the indeterminacy of the concept of a work of art to the indeterminacy of the concept of conversation. In itself, this might be an advantage of the theory, to the extent that the concept of artistic artifact has fuzzy borders.

The fact that conversations are different in time and space does not create problems in so far as some conversational elements - the choice of topos, the way of developing it - are subject to cultural universals. Here, the theory makes an empirical prediction that will have to be tested: if what counts are not the conversations actually taking place, but the generic aspects of conversation, a study of the latter should bring to light some elements of works of art that usually go unnoticed. Where does the study of cognition come in regarding the conversational theory? In the fact that not all subjects are good for conversation and assure successful conversations. Studying the normative constraints of conversational success will enable us to make interesting empirical predictions regarding the content and form of artistic artifacts.

above copied from: http://www.interdisciplines.org/artcog/papers/4