We all sometimes talk to ourselves in our minds.
We call this inner speech. It helps us concentrate, plan or solve problems, as well as to make more morally demanding decisions.
A research team from the University of Palermo wondered if internal speech would also help robots in solving problems.
The experiments were performed with the Pepper robot.
It can recognize emotions and is designed to communicate and socialize.
They built the ability of inner speech into him and listened to what he was thinking while performing tasks.
One of the tasks was to follow the rules of behaviour when eating or at the table. According to these rules, the napkin is on the plate, the fork is on its left side, and the knife and spoon are on its right.
The researchers then instructed him to move the napkin, although this is not in line with etiquette. Thus Pepper was faced with a dilemma: should he follow etiquette or human command?
»Wow, this situation makes me nervous. I don’t want to break the rules, but I can’t make a man angry, so I’ll do what he tells me to do,« Pepper said to himself, shifting his napkin.
They thus monitored his thought process while resolving the dilemma.
Subsequent experiments have also found that a robot that talks to itself in its mind is more successful at solving more complex moral dilemmas.
»If we can hear the robot thinking, we might trust it more,« said study co-author Antonio Chella.
»This will also make it easier for ordinary people who are not technicians or engineers to understand robots and to be able to work and communicate better with them.«
Points to Consider
- What is artificial intelligence?
- Have you ever talked to a robot?
- Would you like to have a robot?
The original version of this article was published on June 21st.