Their findings appear to challenge the popular concept of the "virtual community", where a support network of virtual mates is always there to ease the user's path through life.
The research, published in the journal American Psychologist, was carried out for a study dubbed "HomeNet", designed to look at the impact of the Internet on the social life and mental wellbeing of the average American. Some 169 human guinea pigs in 73 formerly Internet-free homes were recruited for the study.
Feedback from the group has revealed that, while the majority of users were quickly hooked on the Net, the increased time spent at the computer was accompanied by a decline in interaction with family and friends. A rise in the reported incidence of loneliness and depression followed.
Teenagers, who tend to use the Internet in longer bouts than adults, were found to be the most vulnerable to negative effects. After two years, the team found that they could predict the changes in a user's emotional state according to the number of hours spent online.
Professor Bill Scherlis, who worked on study, confirms the disturbing results but says he hopes it will not be used as a stick to beat the medium. "We should simply become more discriminating in our use of the Net," he told The Independent. "The challenge is for us to become better users."
The increase in signs of depression among the study group can be clearly associated with their use of the Net, Scherlis says, because of the number of controls that the team introduced during the trial. There was no question, for example, that the study group happened to be full of people who were predisposed to depression. Yet, he emphasises, the findings of HomeNet should be seen in context. "It was a very mild effect, but it was statistically significant," he says. "We want to explore it further, so all other interpretations are speculative at this stage."
On the up side, the urge to communicate has been underlined by the study as the main human motivation for using the Net, as opposed to, say, simply information gathering. The problem is, Internet contact does not appear to be as valuable a form of communication as many of those it replaces.
Bob Kraut, the professor of social psychology who led the research, explains the displacement that may occur in some users' lives, regardless of the fact that they think they are keeping up lots of friendships by e-mail. "People can't easily report on what they are giving up," he says. "For example, I find the Internet very convenient for keeping up with colleagues from my old job. The question is, does the time and energy I devote to these e-mail messages hinder me from forming strong friendships with people in my current workplace or community?"
Professor Scherlis thinks Net users are often deceived into believing they are strengthening their relationships through the Net. What they don't spot is the fallout around them in the real world. "The research shows that although Internet use is in fact a socially positive experience, it may not be as positive as other activities it replaces," he says.
"Sociologists distinguish between the kind of friendships you have with people who might actually offer to drive you to the airport and those you have with people you might just say `hi!' to when you pass in the corridor."
The $1.5m study was funded in part by a consortium of IT companies, including Apple Computer, Hewlett Packard, Intel Panasonic and Lotus Development Corporation, who may now wish they had not bothered.