Who Is More Able to Connect with Robots?

Source: TheDigitalArtist/Pixabay

Late one evening, Denice returns home exhausted from a long day at work and is expecting company for dinner. Luckily for her, John has already cleaned the floors, folded the laundry, and started cooking two hours ago. Denice sits immediately on the sofa, and John listens to her intently as she unloads the trouble that went down in the office that day. John comforts Denice with kind words and puts an arm around her shoulder.

Reading this today, we might assume that John is Denice’s roommate, close friend, or romantic partner. But, in the near future, we are expected to perceive a different reality: John is a multifunctioning social robot that is capable of fulfilling many of the needs that Denice might have sought in other humans. The question we face now is not whether this reality materializes, but who will be the first to join this trend and how far it can go.

Role of Human Attachment Patterns

Researchers at the Wroclaw University of Science and Technology in Poland sought to answer these questions, looking carefully at the role that human attachment patterns play in mediating the experience of human–robot cohabitation.

In their study, humans were generally categorized into displaying one of four types of attachment in interpersonal relationships: secure, as demonstrated by a positive view of the self and a positive view of the other; anxious, as demonstrated by a negative view of the self and a positive view of the other; dismissive, as demonstrated by a positive view of the self and negative view of the other; and fearful, as characterized by a negative view of both the self and others.

The team from Wroclaw hypothesized that different types of attachment styles are likely to create different lived experiences of cohabitation and interaction between humans and a social robot. Here, the "attachment" is not to another human, but to a social robot.

To test their idea, the scientists identified three research participants with varying attachment styles who volunteered to cohabit with a social robot for 10 days. The robot could carry out functions such as playing music, speech recognition, operating household devices, and using social networks and Internet-based communication on a human’s command.

The robot was programmed using EMYS (“Emotive Heady System”) to detect the human’s emotions and respond expressively in a way that is intended to match the emotions of the human user. Following the 10 days, the scientists used the robot’s activity log together with in-depth interviews to analyze the human participants’ experiences.

Study Results

The results showed that all of the participants became attached, at least on some level, to the robot’s function, or indeed to the robot itself. The extent to which participants were satisfied with the robot, however, was largely dependent on their attachment style.

The researchers identified three possible patterns. First, the participant with a safe attachment style attributed the most human characteristic traits to the robot, labeling it as friendly, patient, and open. This participant was the only one to admit to treating the robot as a human.

The participant with an anxious attachment style reported on technical issues with the robot, and also on fear and concerns of being observed. They treated the robot as a pet, and explained that it took time for them to learn how the robot’s emotions function.

The participant with avoidant attachment styles treated the robot most mechanically and felt more distant from the robot than the other participants. They were also the most skeptical about the potential role and benefits of said robot.

As such, this study indicates that there is a need to understand how the emotional and psychological repertoire of each individual will affect their interactions with smart robots and artificial intelligence. We have just started to grasp this. It seems that people who feel secure in relationships are those who will adopt this revolution earlier.

Be honest, are you surprised by these suggestive results? Most of us think that "weird" and "socially challenged" people are those who find robots friendly, but this study, although quite limited in scope, hints otherwise.