Robot study the difference between men and women, adults and children

As this relate to human beings, it is not about the collection of data for a certain period of time or humans being conducted a test on. But rather, it is the use of a Robotic head dubbed Furhat to uncover inequalities in people’s participation when working on a shared activity, and see if a robot could help redress the balance.

The Robot achieves this by two people at a time sitting around an interactive table with a touchscreen opposite the robot. They will be asked to play a game that involves sorting a set of virtual picture cards, such as arranging images of historical inventions in chronological order. The people worked with the robot to try to solve the task. During this time, the robot’s sensors tracked how long each person spoke for.

According to the results gathered so far, when a woman is paired in conversation with another woman, she speaks more than if paired with a man. And two men paired together speak less than two women. But this only holds for adults. “Surprisingly, we didn’t find this same pattern for boys and girls. Gender didn’t make much difference to how much children speak,” says Gabriel Skantze at the KTH Royal Institute of Technology in Stockholm, Sweden, who is also one of the robot’s creators.

Mr Skantze also said that “this turned out to be a really nice opportunity to study the differences between men and women, and adults and children."
Analysis from the Machine, Furhat shows that pairs of women spoke for 45 per cent of the time on average, compared with just 26 per cent for pairs of men. When women were paired with men, the speaking time was 28 per cent, with each gender sharing the time about equally. For children, there was no significant difference between gender pairings.

Read:Flying Robotic Ambulance Undergoes First Solo Test Flight.

In cases of adults paired with children, the former dominated the conversation. The largest imbalance occurred when a male adult was paired with a female child, with the men speaking more than twice as much as the girls.

When it was the robot’s turn to say something, its behaviour was randomly selected from four options, such as addressing a question to the dominant or non-dominant speaker. This hints at how a robot could influence the conversation. “When Furhat directly addressed the less dominant speaker, they were more likely to speak,” he says. “We want to use this idea to make interactions with the robot more equal.”

Furhat shows that results regarding study on humans can't actually take longer time to process.

No comments:

Post a Comment