People may not perform selfless acts just for an emotional reward, a new brain study suggests.

Instead, they may do good because they're acutely tuned into the needs and actions of others.

Scientists say a piece of the brain linked to perceiving others' intentions shows more activity in unselfish vs. selfish types.

"Perhaps altruism did not grow out of a warm-glow feeling of doing good for others, but out of the simple recognition that that thing over there is a person that has intentions and goals. And therefore, I might want to treat them like I might want them to treat myself," explained study author Scott Huettel, an associate professor of psychology at Duke University Medical Center, in Durham, N.C.

He and lead researcher Dharol Tankersley, a graduate student at Duke, published their findings in the Jan. 21 online issue of Nature Neuroscience.

For decades, psychologists and neuroscientists have puzzled over the tendency of humans to engage in altruistic acts -- defined by Huettel's group as acts "that intentionally benefit another organism, incur no direct personal benefit, and sometimes bear a personal cost."

Experts note that altruism doesn't seem to provide individuals with any survival edge, so how and why did it evolve?

To help solve that puzzle, Heuttel's team had a group of healthy young adults either engage in a computer game or watch as the computer played the game itself. In some sessions, the computer and participants played for personal gain, while in other sessions, they played for charity.

The researchers used high-tech functional MRI (fMRI) to observe "hot spots" of activity in the participants' brains as they engaged in these tasks.

Participants were also asked to complete a questionnaire aimed at assessing their personal levels of selfishness or altruism.

Huettel said he was surprised by the study results.

"We went into this experiment with the idea that altruism was really a function of the brain's reward systems -- altruistic people would simply find it more rewarding," he said.

But instead, a whole other brain region, called the posterior superior temporal cortex (pSTC), kicked into high gear as altruism levels rose.

The pSTC is located near the back of the brain and is not focused on reward. Instead, it focuses on perceiving others' intentions and actions, Huettel said.

"The general function of this region is that it seems to be associated with perceiving, usually visually, stimuli that seems meaningful to us -- for example, something in the environment that might move an object from place to place," he explained.

This type of perception would have allowed humans' more primitive ancestors to quickly pick out a potential threat -- a crouching lion, for example -- from amid a mass of less important stimuli.

It's much less clear why pSTC activity gets ramped up in the brains of altruistic people, however. "That was really surprising to us," Huettel said.

The researchers found that pSTC activity was highest when study participants were observing the computer play the game on its own -- not when they were playing themselves. "That gets to this idea of agency -- watching somebody else play the game," Huettel said. "You are thinking, 'Oh, the computer pressed the button -- somebody else did that.' "

The bottom line, he said, is that altruism may rely on a basic understanding that others have motivations and actions that may be similar to our own.

"It's not exactly empathy," he said, but something more primitive. "We think that altruism may have grown out of -- at least in part -- such a system."

Another expert said the Duke study raises even more questions than it answers.

"It's a really interesting study," said Paul Sanberg, director of the Center of Excellence for Aging and Brain Repair at the University of South Florida College of Medicine, in Tampa. "It would be really interesting, now though, to see if people who had damage to that [brain] area were much less altruistic."

Huettel said he's pondered that possibility. "For example, we don't know if people who are sociopaths, or people who are autistic, might show differences in this region," he said. "It's a good question, but we don't have data that shows anything one way or another. This is just a jumping-off point."

Sanberg said the study also showed only an association between heightened pSTC activity and altruism, not a direct cause-and-effect relationship. "That needs further study," he said.

But the Florida neuroscientist said this type of work is helping unravel the mysteries of human consciousness and behavior.

"These functional studies with high-level human behaviors are shedding important light on the contribution of different brain areas," Sanberg said.