Science Watch

Misery loves company for reasons science is beginning to understand. Researchers at the University of Tokyo used three different measures to confirm that the mere presence of a partner alleviated the stress response in rats, reducing stress-induced behaviors such as "freezing," helping them to literally keep their cool, and staving off high levels of stress-related brain chemicals.

What's more, the presence of another rat that hadn't been stressed by foot shocks was even more protective than the presence of another rat that had been shocked.

In the study, reported in the August issue of APA's Behavioral Neuroscience, the researchers studied--at the behavioral, physiological and neurochemical levels--three converging measures of the effects of so-called social buffering, the general term for the protective health benefits conferred by relationships with one's fellow creatures.

The research team included neuroscientists Takefumi Kikusui, DVM, Yukari Takeuchi, DVM, Yuji Mori, DVM, and veterinary graduate student Yasushi Kiyokawa. Their findings bear out other evidence of the health benefit of sociability in animals such as guinea pigs, rats, monkeys and humans.

The likely reason? "When there's a companion available, the physiological measure of stress--cortisol levels in the blood--can be alleviated somewhat," explains Jim Winslow, PhD, a specialist in behavioral neuroscience and pharmacology and head of the National Institute of Mental Health (NIMH) Intramural Research Program, Non-human Primate Core.

You've got a friend

To investigate stress buffering among rats, Kikusui and his colleagues assigned young male rats to either a subject or a partner group. A week before the experiment, researchers implanted temperature transmitters in the subject rats because stressed animals show sudden, temporary increases in body temperature--"stress-induced hyperthermia"--triggered perhaps by a deregulating rush of endocrine hormones.

On the first day, the researchers randomly delivered nine half-second foot shocks over 15 minutes to "subject" rats in test boxes, thus teaching them to fear the boxes. Next the researchers assigned the "partner" rats to one of two groups, shocked or nonshocked. They shocked only the shocked-partner rats the same way they had shocked the subject rats. Nonshocked partners received no shocks; they just stayed in the boxes for 15 minutes.

On the second day, the researchers began the experiment only when each subject rat had a body temperature of less than 37.5 degrees Celsius for at least five minutes. The experiment itself compared the behaviors, body temperatures and stress chemicals of three seven-rat groups:

  • Solitary: Subject rats in the test box alone.

  • Shocked Partners: Subject rat and shocked partner in the test box together.

  • Non-shocked Partners: Subject rat and nonshocked partner in the test box together.

The solitary and partnered rats each stayed in the test box for an hour while the researchers videotaped behavior and recorded body temperature. For the third measure, after the experiment, the researchers deeply anesthetized the subject rats and examined brain tissue from the paraventricular nucleus (PVN)--a part of the hypothalamus that controls the release of stress hormones such as cortisol.

On the behavioral level, experimenters meticulously coded either the frequency or duration of social (walking over or crawling under, anogenital contact, grooming and nosing) and nonsocial behaviors (freezing--not moving except for breathing, which rodents do when they sense danger--resting and self-grooming).

On the physiological level, the researchers analyzed the stress-induced hyperthermia response as the temperature change from baseline for each group.

And on the neurochemical level, Kikusui and his associates counted the number of nerve cells that react to the "Fos" protein in the subject rats' paraventricular nuclei--a marker of neuronal activity in a region significantly involved in mediating the stress response in the brain.

Analysis of paralysis

The presence of both shocked and nonshocked partners attenuated subject rats' nonsocial behaviors (including freezing) and increased the time they rested, confirming the buffering effect of social contact. What's more, nonshocked partners, who had no reason to fear the box they shared with the subject rats, were the more calming of the partner types.

All subject rats had higher body temperatures after going into the test box, but again, the partners made a difference. Both types significantly weakened the physiological stress response, but again, nonshocked partners were better at helping the subject rats keep their cool.

Finally, the immunoreactive nerve cells were found in all subject rats. However, the numbers differed significantly between groups. Here, results favored only the nonshocked partner. Only a nonshocked partner reduced the level of stress-related Fos protein in the brain. With unafraid partners by their sides, the subject rats' brains simply didn't get as excited under stress, demonstrating, "on a much more mechanistic level what goes on in the brain to mediate social buffering--not only that it makes you feel better, but why it makes you feel better," says John Disterhoft, PhD, a professor of physiology at Northwestern University's Institute for Neuroscience and editor of Behavioral Neuroscience.

The company you keep

More specifically, the researchers speculate that corticotropin releasing factor (CRF), the stress hormone that stimulates cortisol production, could be the mechanism through which social buffering helps animals to "lighten up," behaviorally and physiologically. Although NIMH's Winslow would have liked to see cortisol measured, as it is by other stress researchers, he says, "It's likely that the paraventricular nucleus is telling us the same thing."

The study also raised an intriguing question: Given that the rats weren't exactly saying "There, there" to each other, why were nonshocked partners better at helping others relax?

The researchers found no signs that the rats transferred anxiety-reducing information through behavior, although Winslow points out that because rats vocalize at ultrasonic frequencies, measuring auditory cues might have been helpful. Kikusui's team believes that shocked partners may give off mixed signals--some that buffer stress, some that signal "danger!" This theory wins by a nose: According to prior research, sniffing the "alarm pheromone" aggravates stress responses.

Adds Kikusui, "Some kinds of odors also can have calming signals. Humans also seem to communicate their emotional states using odors." The authors conclude that severely stressed partners may convey aversive information and compromise social buffering by releasing alarm pheromones from the surface of their bodies.

The Tokyo research on rats and similar research on monkeys, Winslow adds, "may help us to tease out the physiological basis of social support." A moderate but not complete reduction in stress may reflect the fact that organisms need to experience some stress to survive. Thus, social comfort is more like a modulator. Says Winslow, "You want people to experience appropriate fear; you just don't want it to be all-consuming."

Extending the findings to clinical work with people, Winslow hypothesizes that social buffering "may accelerate recovery from stress and change your experience of it. This has important implications for treating depression and anxiety, in which recovery from stress is impaired. Social companions seem to help with recovery, which seems to fit what clinicians think about trauma, stress and social support."

He believes the study best reveals that "it's best to get reinforcement from someone who's not in the same lifeboat after you've gone over the side."

Rachel Adelson is a science writer in Raleigh, N.C.