June 13, 2001
Lost in virtual space: Gender differences are magnified
TORONTO – Well-documented gender differences in people’s ability to navigate and orient themselves in the real world are vastly exaggerated in computer-simulated virtual environments, according to studies conducted by University of Washington researchers.
The latest study, which will be described at the annual meeting of the American Psychological Society here Saturday, shows that these variances are not because men have greater skills than women in using joysticks, which are commonly used for navigating in virtual environments and in computer games.
“These gender differences are a very real effect and no one understands why this is the case. We are just now starting to look at what causes it,” said Earl Hunt, a UW psychology professor, who conducted the newest study with Maryam Allahyar, a UW psychology doctoral student.
“If you ask people how to get somewhere, men have a larger frame of reference while women will give procedural direction. And when you ask for orientation, or to point to the direction of something, women don’t do as well. There are huge individual differences. Of course, many women have an excellent sense of direction. However, of those people in a group who are really turned around – 90 degrees or more – and literally are going to get lost, the vast majority are women.”
Hunt said the actual real world male-female differences are not that great, but they are vastly enlarged in virtual reality. Some people have speculated that these differences might be hormonal or caused by boys playing more ball games than girls do.
David Waller, who worked with Hunt and earned his doctorate at the UW and is now a post-doctoral researcher at the University of California, Santa Barbara, earlier suggested that women have more trouble manipulating a joystick because boys play more computer games.
That idea, however, is disputed in the new study. Eighty subjects were asked to operate in a computer-simulated environment consisting of a large central room surrounded by hallways. Both the room and hallways were filled with a variety of objects. The subjects were asked to navigate, using a joystick, from one point to a specific object by choosing the shortest route. Hunt and Allahyar paired subjects, having one person manipulate the joystick and the other wear a virtual reality headset and watch the simulated environment on the screen. Once again, the overwhelming majority of people who got “lost” in this exercise were women.
The gender differences could become important if virtual environments are adopted as training or educational devices, unfairly penalizing women and girls, said Hunt. The U.S. Navy already has conducted simulated shipboard fire-fighting training exercises, as well simulated at-sea fueling and refueling of vessels, which is a difficult operation, according to Hunt. In addition, several universities are exploring the use of virtual reality desensitization programs for treating people with a fear of heights.
Hunt said there are a number of real-world reasons to explore this virtual phenomenon. First it offers a way of studying human orientation without having to take people into the wild or real world, which is costly and time-consuming. At the same time, as the use of virtual environments in training situations increases, it is necessary to establish a scientific basis of their effectiveness in transferring skills and learning to the real world.
“Right now, if virtual reality were used in training or screening situations, it would exaggerate male-female differences in orientation,” he said. “We look at virtual reality as a tool to study what might be the cause of these differences and how they might be reduced.”
The research is funded in part by the Office of Naval Research.
###
For more information, contact Hunt at (206) 543-8995 or ehunt@u.washington.edu; Allahyar at (206) 543-8839 or maryam@u.washington.edu. Hunt will be attending the American Psychological Society meeting in Toronto Thursday through Saturday and may be reached at (416) 361-1000.
