Study Shows People Connect Better with a Computer-Generated Avatar Representing Building Management

Researchers at the University of Southern California (USC) see a lack of communication when it comes to buildings and their occupants. However, if there is improved dialogue between the two, smart buildings can work better for a sustainable society.

In a recent study, investigators discovered that slight modifications in the design of virtual assistants lead to behavioral changes that can prove beneficial to the environment. The scientists observed that people are able to connect better with a computer-generated avatar representing building management. They noted that better results are obtained when there is social banter between people and machines. These findings show how social interactions and personal connections, which are integral to human relations, also encourage cooperation between machines and people.

The study titled, “Establishing Social Dialog between Buildings and Their Users,” has been published in the International Journal of Human-Computer Interaction on December 27^th, 2018. The study is a product of convergent disciplines at USC, including communications, engineering¸ and behavioral sciences, and is the latest one by USC researchers focused on the dynamics of humans and machines. The study authors are from the USC Institute for Creative Technologies and the USC Viterbi School of Engineering.

Virtual assistants key to communication between people and smart buildings

Virtual assistants are as old as HAL 9000 from 2001: A Space Odyssey and as new as Siri or Alexa. The rapid emergence of smart buildings equipped with automated systems for lighting, heating, and cooling is significant to improve the health and productivity of workers, safeguard the environment, and conserve energy.

You can think of things like the computer interface on the Star Trek ship Enterprise as closer to reality than science fiction. We’re beginning to explore where the line is between people and buildings that perform as machines. We are trying to get people to feel more comfortable and make smart buildings perform better.

Gale Lucas, Research Assistant Professor and Study Corresponding Author, USC Viterbi School of Engineering.

Americans are known to spend 90% of their time within buildings, either studying, or working, or shopping¸ or sleeping. The operation of buildings is directly affected by quality of life, comfort, and worker safety and productivity.

The U.S. Green Building Council has revealed that buildings are responsible for 39% of carbon dioxide emissions in the United States. The state Air Resources Board in California has ranked the building sector as the second largest source of greenhouse gas emissions and has directed the sector to reduce its carbon emissions by 80% by the year 2050. Behavior practices have also been identified by the board as a critical influence for the consumption of energy in buildings.

Communication between people and smart buildings improves performance

However, the promise of smart buildings usually does not correspond with their performance. In spite of automation, human actions within office floors or four walls can affect the energy efficiency capabilities of a building. Buildings can work more efficiently when there is cooperation within the building occupants.

If a building spoke to you, it could ask for things that might help the environment, like ‘turn off lights’ or ‘open windows’ or ‘save energy. If the building were to ask people, ‘Why don’t you do something environmentally-friendly?, we might get people to engage in healthy behaviors for themselves and the environment.

Gale Lucas, Research Assistant Professor and Study Corresponding Author, USC Viterbi School of Engineering.

To achieve that objective, buildings and people need better cooperation, computers need to utilize established behavioral change interventions to prompt people to act, and people need to have faith in the technology. The USC study explored how to achieve that in the most optimized way.

The researchers chose 200 participants, predominantly college students, of mixed ethnicities and race. Initially, participants were subjected to an office environment through virtual reality, while a smaller group of participants were exposed to a real office setting. The investigators created pro-environmental messages for a virtual assistant—an agent dubbed Ellie—to put forth questions, like “If I open the blinds for you to have natural light, would you please dim or turn off the artificial lights?” and subsequently monitored whether the participants cooperated with these pro-environmental requests.

Who is involved in communication between people and smart buildings?

The outcomes, mimicking similar previous study conducted by the USC group, demonstrated that subjects responded better when the virtual human Ellie acted on behalf of the building manager, rather than when she conducted as the building’s personification.

Moreover, when the messages were part of a social dialogue instead of monologue, people were more cooperative. This element was shown to be crucial, indicated the study. For instance, subjects tend to respond better to small talk, like, “Hi, how are you? …. What’s your name?… I’m glad to see you.” Bias against the avatar can be overcome through social dialogue. The scientists discovered that when Ellie used social dialog, irrespective of whether she functioned as an agent of the building manager or as the face of the building, people responded well to her.

According to the research, “Including a social dialog may have helped to overcome the difference between personas by making the building persona more relatable. Indeed, people associate monologue with strangers and dialog with closer relationships.”

Whether study participants operated in a virtual reality simulation or an actual office, the researchers noted similar results. The team also discovered that when they again performed the experiment a week later, the participants responded in a more positive way, indicating that familiarity owing to repeated interactions helped.

We are trying to build a relationship between buildings and their users, akin to a friendship, so users are empowered to improve individual performance as well as building performance.

Burcin Becerik-Gerber, Associate Professor, USC Viterbi School of Engineering.

The findings indicate that design decisions accounting for the human-machine dynamic will be key for realizing the potential for smart buildings.

“Our research contributes to the fundamental understanding of human-machine teamwork,” stated Becerik-Gerber. “The impact is beyond just smart buildings. The work changes the way we perceive and experience today’s built environments and artifacts, environments and artifacts that are attentive and have an identity that can have two-way interactions with people.”