Do machines know how to think?

When our PC 'strikes', we often curse them as if they were humans ...

When our PC 'strikes', we often curse them as if they were humans.

The question of why, in what circumstances do we assign human characteristics to machines and how such processes are expressed in the cortex studied in the project of Dr. Sören Krach and Professor Tilo Kircher of Aachen RWTH University (Psychiatry and Psychotherapy Hospital) in collaboration with the Department of Human Machine Learning (Bielefeld University) and Neuromage Nord (Hamburg).

Almost every day, every step in the field of robotics appears on the media. The work of creating increasingly human and meticulous human-like robots is widely published, so the impact of humanoid robots accumulates in everyday life. However, the question of how people perceive these machines and the mounting of abilities and 'intellectual quality' for robots is largely unexplored.

In the fMRI study, published in PLoS ONE, Krach and colleagues investigated how increasing the number of human-like partners affected participants' brain activity. Participants play a simple computer game against four other players, including: a regular notebook, a fully-featured jigsaw robot, BartHOC Jr. humanoid robot. and another person. All participants played the same game however they did not know about their game.

Picture 1 of Do machines know how to think?

Humanoid robot, Barthoc Jr.(Bielefeld humanoid robots have human-oriented communication).(Photo: Krach)

The results clearly demonstrate that brain activity in the middle frontal lobe as well as at the temporal crown joint must increase proportionally to the level of 'human' . For example, the more you play, the more human-like traits are displayed, the more participants use cortical areas that are related to mental states.

Moreover, in a short questionnaire, participants stated that they were most excited when they played their interactions that represented the most humane characteristics, from which they evaluated your intelligence.

This is the first study in the field of understanding the neurological basis of direct interaction of human-like robots at a high level of awareness such as intellectualization. Therefore, the researchers expect that the results will affect long-standing philosophical and psychological debates about the humanoid machine's interaction, especially the question of what makes humans are considered human.

Reference articles

Krach S, Hegel F, Wrede B, Sagerer G, Binkofski F, et al. Can Machines Think? Đang xác định và Perspective Taking với Investigated Robots via fMRI. PLoS One, 3 (7): e2597 DOI: 10.1371 / journal.pone.0002597

Update 11 December 2018
« PREV
NEXT »
Category

Technology

Life

Discover science

Medicine - Health

Event

Entertainment