Bisexuality is not a phase or a period of experimentation that inevitably leads toward same-sex partnerships, new research has found, but a "third orientation" that remains distinctive over the years.
The study, which followed a group of 79 non-heterosexual women over a 10-year period, was published in this month's edition of Developmental Psychology.
The findings fly in the face of the popularly held notion within both straight and gay communities that bisexuality is not a real form of sexuality, but a temporary attraction, said Lisa Diamond, the University of Utah psychology professor who conducted the study.
"We're a culture that still has a very rigid notion of sexual categories: If you're not totally gay you must be totally straight," Dr. Diamond said. "Bisexuality throws that right out the window. So it's easier to dismiss bisexuality as not being real."
Of the women who identified as bisexual in 1995, 92 per cent identified as bisexual or unlabelled in 2005. Of the women who identified as lesbian in 1995, 66 per cent identified as lesbian 10 years later, 19 per cent had switched to bisexual and 16 per cent to "unlabelled." None of the women who identified as lesbians in 1995 switched to the heterosexual label.
But Dr. Diamond found that her subjects' definition of their own sexuality was quite fluid.
Seventeen per cent of respondents switched from a bisexual or unlabelled identity to a heterosexual identity at some point during the study.
But more than half of those women switched back to bisexual or unlabelled by the end of the 10-year period.
And of the women who identified as lesbian during the last round of interviews, 15 per cent reported having sexual contact with a man during the prior two years.
"The distinction between lesbian and bisexual women is not a rigid one," she said. "Like with most people, a lot seems to depend on who you happen to meet."
Bisexuality has not been the subject of much academic study over the years, Dr. Diamond said, and subjects who identify as bisexual are often excluded from studies of human sexuality because researchers do not know how to interpret their results.
"I've had journal editors say it would make for a much cleaner study if you just took them out," she said. "And that's exactly what keeps happening, so we know almost nothing."
Some of her subjects themselves said they were worried about participating because they felt their sexual histories involving both men and women were "bizarre."
"They didn't realize how normal they really were, and that's why I think it's important to get this information out there," she said.
Dr. Diamond believes her research will help many people understand their own sexuality and feel less uncomfortable about their seemingly competing attractions to men and women.
Over the 10-year study, the most common identity for women to switch to was unlabelled, she added, a message that may comfort young people who feel pressure to conform to either straight or gay lifestyles.
"One of my subjects explained it to her mom by saying, 'It's kind of like a garage,' " Dr. Diamond recounted. " 'I'd be happy driving a red car, I'd be happy driving a blue car, but I've only got a one-car garage.' "
Of the subjects who had babies during the period of the study, the majority were with male partners.
Most of the women who had children with female partners self-identified as lesbian.
Although her study followed only women, Dr. Diamond believes that bisexuality means different things to different genders.
"I do think there's enough evidence now that women's sexuality does appear to be more fluid than men's," she said. "I think it's a combination of biological and social factors. For women, there's a closer link between emotional connections to other people and their sexual feelings."