Distance isn't Dead: Real World Distance Affects Online Collaboration & Community

By Peter Rothman, published at 10 May 2007 - 8:12pm, last updated 11 years 1 week ago.

Erin Badner and Gloria Mark of U.C. Irvine published a paper a few years ago in which they examined the impacts actual physical distance between online participants on collaboration within an online space. They tested both Instant Messaging (IM) and video conferencing systems as part of their experiment.

The test included three different measurements: the desert survival task, the Paulhus Deception Scales and the prisoner’s dilemma game. They found that "the geographical distance of a
collaborating partner affects one’s willingness to initially cooperate with, be persuaded by, and deceive that partner. If people believed that their partner was far away in a distant city, they initially cooperated less than if they believed the partner was close. However, cooperation increased with interaction. On the other hand, if one believed their partner to be close, the amount of
cooperation did not increase with interaction. In addition, people were less persuaded by a person that they believed to be distant from them, compared to being in the same city. Also, people were more likely to give deceptive (positive) portrayals about themselves to a partner that they believed to be in a distant city, as opposed to the same city."

Drawing on social identity theory and social impact theory, they reason that:

"When the confederate was not a local, why were subjects more likely to inflate themselves through deception? This is where we return to social impact theory. If people intuitively know they are disadvantaged by distance – i.e. ill-equipped to influence a remote person to form a favorable image of them – they may stretch the truth to compensate.In responses to our survey, subjects confirmed
that it was more likely that they would meet the Same city participant (i.e. confederate) in person some day compared to the Distant city participant (F(1,94)=10.60, p<.003). This expectation of meeting may reassure subjects that they will be able to enhance or otherwise modify the image the confederate holds of them should the need arise. With the confederate geographically remote, there is little likelihood of meeting her in person and the image portrayed must be favorable. Thus, it may be easier to deceive the confederate when one believes she is remote. On the other hand, if people believe they are more likely to meet the partner, they may be less likely to deceive them. After all, if they do meet each other some day, their deception will be 'found out'."

"A second possible explanation for our results which expands on the 'similarity' hypothesis is the ecological model of social identity proposed by Tajfel. Social identity theory suggests that we are less likely to trust, cooperate and attribute expertise to individuals who are further from us compared to those who are near. In our questionnaire, subjects responded that they considered the confederate in the same city to have significantly more expertise on the survival task than those in the distant city (F(1, 96) = 4.32, p<.05). In other words, subjects attributed expertise in a manner consistent with Tajfel’ s model. It is possible that subjects were more persuaded by the near confederate because they felt she had more expertise. Yet subjects in the near condition did not report in the questionnaire that they felt they had more in common with the confederate than did subjects in the far condition. Thus, empirical support for social identity theory to explain our findings is tentative, at best."

Badner and Mark did not find any significant difference between the two media types (video and IM) despite the significant difference in perceptual information available from these two channels. That is surprising to me. I'd like to see some of these experiments repeated to see if similar results are obtained. It is interesting to speculate how performing this experiment inside a 3D virtual environment such as a multi-player online game or Second Life which has its own illusion of space and distance would change these results. Does real world distance dominate virtual distance? Exercise left to the reader. ;)