I recently became familiar with the term “Uncanny Valley” while having a conversation with Joshua Young who is a Digital Product Creation Consultant and a brilliant expert on 3D design and applications.

At first, avatars seem like an obvious pairing for virtual fit technologies. It seems logical that people should want to see a representation of themselves in a virtual try-on session. After all, having a virtual version of yourself to try-on the clothing should make the experience feel more real.

To test this idea on a very small scale I included multiple avatars for participant interaction.

Visitors to the website encountered 3 distinct Avatars responsible for communicating “fit” in the virtual world.

The Model

The Cyber Peer

The Android

A very interesting thing happened when participants were presented with the third example, a more “human-like” avatar: People felt uncomfortable. The responses of the survey showed that some of the participants even had emotional responses to the human-like avatars. One participant’s response to the 3D body scan avatars is below:

“ …Unless the point is you need to be in really good shape or be really skinny to buy jeans…”[ the 3D images were not helpful]

Other comments similar to this one revealed a bit of insecurity when presented with the human-like comparison. Not everyone had the same reaction to the 3D figures most people found them useful. Logically, the 3D representations would be the most accurate for understanding shape and bodily measurements when comparing denim but participants did not share these sentiments.  Had the image been an actual photo i’m not sure if they would have had the same responses but further research would need to be conducted in order to verify.

So what is going on here?

Well, what is happening can be explained through an understanding of “ The Uncanny Valley”.

The term is mostly used in robotics  but can be applied to animation. To be plain, the chart below shows what happens when animation and robotics come close to making a human like image, but the difference is noticeable to actual humans…it creeps people out. The chart  illustrates the effects of human-like images and people’s level of comfort.

Based on my study it was observed that human like representations can take the focus off the actual goal and distract people. Distinctly human-like avatars although useful in gaming industries (when people want to project an image of themselves into the virtual world) may not have the same effect in the fashion world. The point of the fashion avatar would be to give an understanding of fit but if the avatar evokes emotional responses from the consumer he or she may no longer want to shop. If the avatar is too animated, he or she may only use the avatar for entertainment. If the avatar is not attractive, prejudices could ruin the experience for the online shopper. If the avatar designed to look like the user does not flatter the user,he or she may become offended. With the internet being a fast paced environment and the number of mobile purchases rise, one second of mistrust, confusion or insecurity could cost the retailer the sale.

I have a specific idea in mind for the right type of avatar for use in fit technologies that would encourage better user experiences.

In the case of avatars and virtual fit technology, my professor and I both agreed that more research should be conducted to assess which avatar representation is most helpful  in a purchase decision journey.