What is the Uncanny Valley Effect?

I wanted to share something interesting I came across during my research!

As I’ve been diving deep into the world of virtual AI assistants, I stumbled upon a fascinating phenomenon known as the Uncanny Valley Effect (UVE). It’s this idea that when robots look almost—but not quite—human, they can actually make us feel a bit uncomfortable or uneasy, which can impact how much we trust and engage with them. 🤖 😅

I wanted to share something interesting I came across during my research!

🤖 What is Uncanny Valley Effect:

The Uncanny Valley Effect (UVE) is a phenomenon whereby as robots become more humanlike, our affinity for them increases until a certain point where it drops sharply, creating a UVE. The term "uncanny valley effect" describes a phenomenon in which users are thought to become uneasy when exposed to nearly accurate, or imperfect, pictures of actual people. The valley represents the bottom of a graph depicting the positive of human reaction to a robot or embodied agent as a function of the robot's human similarity.

Image source: Mori, M., MacDorman, K. and Kageki, N. (2012) 'The Uncanny Valley [From the field],' IEEE Robotics & Automation Magazine, 19(2), pp. 98 100. https://doi.org/10.1109/mra.2012.2192811

🦾 Understanding the UVE:

The Uncanny Valley is a concept where robots that are almost, but not quite, human-like can cause feelings of discomfort or eeriness in users, leading to decreased trust and engagement. While hyperrealistic designs might seem like the best option, research shows that they can sometimes backfire, making users feel uneasy rather than connected. On the flip side, simpler, more abstract visuals may offer a better user experience by enhancing comfort and trust.

🎓 Prior Studies in Summary:

Studies collectively offer a comprehensive view of how different visualization styles impact user experiences with virtual AI assistants. Song and Shin (2022) highlighted the negative effects of hyperrealistic visualizations in e-commerce contexts, while Go and Sundar (2019) explored the compensatory relationship between visual and interactive elements in chat interfaces. McDonell et al. (2012) emphasized the appeal of cartoon-style avatars, and Stuart et al. (2022b) provided insights into the role of rendering fidelity in medical simulations. By comparing and contrasting these findings, we can better understand how to design virtual AI assistants that balance realism, user trust, and engagement effectively.

(You can find the original papers from the resource section down below.)

Image source: https://medium.com/advanced-design-for-ai/how-to-make-an-emotionally-intelligent-machine-in-terms-of-visual-design-considering-uncanny-9bc473b0267f

✨ Results & Findings:

Studies found that highly photorealistic characters are not always the most appealing, with cartoon-like characters sometimes being rated more positively. The research highlighted that characters in the middle of the abstract-to-realistic spectrum can trigger discomfort due to their unfamiliar appearance, aligning with the Uncanny Valley theory.

Guidelines have been developed to help designers to avoid the Uncanny Valley Effect. These guidelines focus on elements such as facial features, proportions, and the amount of detail in skin texture. MacDorman(2006) discovered that altering a character’s appearance to a more cartoon-like style can effectively eliminate the uncanny effect like McDonell et al. (2012).

Sources

Stuart, J. et al. (2022b) 'The effect of virtual human rendering style on user perceptions of visual cues,' Frontiers in Virtual Reality, 3. https://doi.org/10.3389/frvir.2022.864676

McDonnell, R., Breidt, M. and Bülthoff, H.H. (2012) 'Render me real? Investigating the Effect of Render Style on the Perception of Animated Virtual Humans' ACM Transactions on Graphics, 31(4), pp. 1–11. https://doi.org/10.1145/2185520.2185587.

Song, S.W. and Shin, M. (2022) 'Uncanny Valley effects on Chatbot trust, purchase intention, and adoption intention in the context of E-Commerce: The moderating role of avatar familiarity,' International Journal of Human-computer Interaction, 40(2), pp. 441–456. https://doi.org/10.1080/10447318.2022.2121038

Mori, M., MacDorman, K. and Kageki, N. (2012) 'The Uncanny Valley [From the field],' IEEE Robotics & Automation Magazine, 19(2), pp. 98 100. https://doi.org/10.1109/mra.2012.2192811

MacDorman, K.F. et al. (2009) 'Too real for comfort? Uncanny responses to computer generated faces,' Computers in Human Behavior, 25(3), pp. 695–710. https://doi.org/10.1016/j.chb.2008.12.026.

MacDorman, K.F., 2006, July. Subjective ratings of robot video clips for human likeness, familiarity, and eeriness: An exploration of the uncanny valley. In ICCS/CogSci-2006 long symposium: Toward social mechanisms of android science (Vol. 4).

Go, E. and Sundar, S.S. (2019) 'Humanizing chatbots: The effects of visual, identity and conversational cues on humanness perceptions,' Computers in Human Behaviour, 97, pp. 304–316. https://doi.org/10.1016/j.chb.2019.01.020

Arzu Cetinkaya

UX/UI Designer

Arzu Cetinkaya

UX/UI Designer

Arzu Cetinkaya

UX/UI Designer