A team of researchers suggests that smart speakers—such as Alexa and Siri—that exhibit active listening cues could play a therapeutic role in users. Credit: Brandon Romanchuk/Unsplash.
According to a team of researchers, adding short, random expressions of understanding to a conversation could turn smart speakers, such as Alexa and Siri, into robotic therapists that allow people to people more openly without violating their privacy.
In one study, the team programmed a smart speaker to respond with reverse channels—short and non-word words, such as “hmmm,” “must” and “uh-huh”—intermittently while conversing with the user. These words, the researchers say, act as the sound equivalent of nods and facial gestures that make users think the virtual assistant is actually listening to them — or actively listening — to them. and promote more emotional expression, the researchers said.
Because users notice smart speaker Saeed Abdullah, assistant professor of information science and technology and branch of the Institute of Computational and Data Science, said that as a good listener and push yourself to reveal more, the speaker can play a role. Roles are emotional support devices for people, especially those who are isolated.
Users also tend to respond with more positive words when they interact with devices, says Abdullah, further supporting the idea that interactions with smart speakers can have therapeutic effects. helpful.
“This represents the first step in rethinking or reimagining the types of interactions that can be facilitated through smart speakers and other voice interfaces,” said Abdullah. “We’re quite interested in finding out if we can use smart speakers to provide some sort of therapeutic aid to people with diseases. mental illnessor even individuals who may be socially isolated, this is a growing concern for mental health professionals.”
The ability to randomly broadcast upstream signals is a win-win, according to the researchers, as it can promote perceptions of active listening while maintaining user privacy. use. According to S. Shyam Sundar, James P. Jimirro Professor of Media Effects, coding a smart speaker engine that actually listens to and responds to a person would require the device to get information from the user, which is what This may be frustrating for some users. Donald P. Bellisario College of Communications and co-director of the Media Effects Research Laboratory at Penn State.
“In a way, this is more socially responsible because we’re doing it in a way that protects privacy,” said Sundar, who worked with Abdullah on the study. “We don’t collect any data, but still give people the social cues they need to continue to disclose and benefit from the catalytic side of the process.”
This approach can be both useful and ethical, he added, as long as companies and designers avoid falsely promoting its capabilities. For example, companies shouldn’t fool users into thinking smart speakers are actually listening to them, says Sundar, director of Pennsylvania’s Center for Socially Responsible Artificial Intelligence (CSRAI) , said.
Abdullah adds that random reverse channel switching also solves some important technical problems.
“One technical challenge it’s quite confusing when to provide that backchannel because it requires a lot of computational processing to understand the context,” said Abdullah.
The researchers recruited 40 participants for the study. Participants were randomly assigned to two different test conditions—one with a smart speaker app programmed to randomize reverse channeling, and a control condition that didn’t provide any backflow. any. Participants were asked to interact with a smart speaker to express their thoughts on a number of issues in their personal lives.
After the interaction, the participants completed a questionnaire to determine their perception of active listening, emotional statusperceived mental support from the speaker and the perceived quality of the intelligent speaker’s ability to switch later channels.
To measure self-disclosure, the researchers recorded how long users spent interacting with the smart speaker and how many emotional words they used.
The researchers published their work in the journal ACM Proceedings on Human-Computer Interaction and presented it at the recent annual conference on Collaborative work powered by computers and social computing.
Eugene Cho et al, Alexa as Active Listener: How Reverse Channeling Can Inspire Self-Disclosure and Boost User Experience, ACM Proceedings on Human-Computer Interaction (2022). DOI: 10.1145/3555164
Provided by
Pennsylvania State University
quote: People are more open to smart speakers actively listening, find researchers (2022, December 19) retrieved December 19, 2022 from https://techxplore.com/news/ 2022-12-people-smart-speaker.html
This document is the subject for the collection of authors. Other than any fair dealing for private learning or research purposes, no part may be reproduced without written permission. The content provided is for informational purposes only.
