Where'd I put that? Maybe it takes a bird brain to find the car keys.

Should humanity get a little too full of itself and its intellectual prowess, there's always Clark's nutcracker to think about. This pale-gray bird with black wings and a long beak flits through woodlands in the West, collecting seeds during times of plenty and tucking them away for a hungry winter's day. During a year, each bird buries 22,000 to 33,000 seeds in up to 2,500 locations, and scientists estimate that the bird recovers two-thirds of them up to 13 months later.

Just how seed cachers do this has fascinated biologists for decades. Scientific investigation of the topic has broadened and deepened in recent years. Cognitive scientists pose seed-storage puzzles to birds as a way of sorting out how their brains work and might resemble our own. Ecologists are looking for links between seed-caching powers and the perils of a species' environment.

ROBOTS NO MORE Thirty years ago, biologists took a very different view of seed caching, reminisces one of the pioneers of the field, Russell Balda of Northern Arizona University in Flagstaff. Bird-watchers knew that certain species store food. These include some members of the family encompassing jays and crows, as well as that of chickadees and tits. Russian and Scandinavian scientists in particular had documented the remarkable industry and seed-recovery accuracy of birds surviving in far-northern regions. Yet, says Balda, speculation about how the birds manage these retrieval heroics centered on the simplest of mental powers.

Scientists had trouble imagining that the birds have the brainpower to remember where seeds are. Instead, biologists typically speculated that the birds must follow a few simple rules, perhaps having favorite types of hiding places. The birds would hide seeds there; later, returning to the same preferred places, they would happen onto their caches.

Biologists of that era tended to think of animals in general as "very robotic," says Balda. "In the early 1970s, the idea that animals were cognitive creatures was still in its infancy."

Balda started studying pinyon jays and Clark's nutcrackers in the late 1960s. He noticed the nutcrackers heading high into the San Francisco Peaks in northern Arizona, and he organized his own flock of extreme bird trackers. To glimpse seed cachers at work, the researchers took advantage of most forms of transportation then available, from snowshoes to motorcycles. The project showed that nutcrackers carry their seeds as far as 25 kilometers to cache them, often hiding their supplies above the tree line. The shy birds were reluctant to cache when observed and often made fake deposits. "It was maddening, Balda says.

The overwhelming logistical problems of observing the birds in the wild led Balda and his colleagues to begin to study caching in the laboratory in 1977. Their first experimental subject, a Eurasian nutcracker, obligingly buried seeds in a large sand floor while Balda watched from a corner. After each caching session, Balda dug up seeds and let the bird back into the aviary to search for its stores. "The bird caught on and refused to cache ever again" Balda says. "Now, we make sure our birds see us as providers of seeds, not robbers." The researchers don't let the birds know they're being watched while hiding seeds.

Once the researchers got the kinks out of their laboratory system, the results began to undermine the idea that the birds just follow simple-minded rules. When Balda gave individual Clark's nutcrackers a second chance to cache seeds in the same sandy aviary floor, the birds often choose different hiding places. Also, when Balda removed all the seeds and smoothed any disturbances in the sand, the birds still searched in 60 percent to 90 percent of the correct locations in a half-hour recovery session. Moreover, the birds were able to find their treasures whether or not the researchers put landmarks such as cinder blocks, wall posters, and two by-fours in the room. These findings challenged the idea that the birds were following some subtle sensory clue to the seeds themselves instead of relocating particular sites.

In a further refinement of the experiments, Balda's research team built a floor that was 30 feet by 50 feet, with 380 holes drilled in it to hold either Dixie cups of sand or plugs. This way the people, not the birds, determined the possible cache locations. Still, the birds maintained their high success rate in relocating their hiding places.

Could it be that the birds were just poking around randomly until they happened on their caches? The rates of recovery, from half to virtually all the seeds cached, made that seem unlikely.

Balda also tested one of his graduate students at caching. The student hid seeds and 30 days later found only about half as many caches as a bird typically did.

Evidence grew that the remarkable birds indeed remembered spatiallocations for their caches, says Balda, and bird minds began to seem more interesting than many scientists had expected.

Moreover, scientific opinion about the minds of animals in general shifted during the 1970s and 1980s, and biologists started looking at mental capability as another trait that natural selection shapes. Just as the fastest and biggest might best survive, so might the smartest.

Caching research, says Balda, "played some role in changing how people viewed the animal mind."

MIND GAMES When Nicola Clayton of the University of Cambridge in England starts to describe how seed-caching birds can illuminate animal cognition, she turns to a cartoon by Gary Larson. It shows a moose saying, "I was down by the lake at me time," and other forest creatures offering similar reminiscences. The cartoon caption reads, "All forest animals, to this very day, remember exactly where they were and what they were doing when they heard that Bambi's mother had been shot."

That recollection of a specific moment, called episodic memory, separates people from animals--at least according to some scientists. Thomas Suddendorf of the University of Queensland in Brisbane, Australia, for example, has argued that evidence so far only shows that people can whoosh back to an instant in their memories or ahead to an imagined future in a process sometimes called mental time travel.

Clayton, however, disagrees. Recent experiments show evidence in birds of mental time travel both backward and forward, she and her colleagues contended in the August 2003 Nature Reviews/Neuroscience.

To test for time traveling into the past, Clayton and her colleagues had Western scrub jays cache mouth-watering wax worms and ho-hum nuts in trays of sand (SN: 9/12/98, p. 181). When some of the jays later recovered their treats, they found that the wax worms spoiled after 5 days. For other birds, the researchers substituted fresh wax worms before each recovery session to create the illusion that both wax worms and nuts stay palatable.

In the next part of the experiment, all the jays tended to go for the wax worms first when they were permitted to rummage for treats soon after burying them. When forced to wait 5 days before recovering treats, the birds that had been shielded from the sad facts of decay continued to search first for the wax worms. However, the jays that had experienced rotten wax worms tended to seek out nuts instead. Clayton and her colleagues say that the decay-savvy birds were remembering the specifics--the what, where, and when--of caching. That resembles an aspect of human episodic memory, she says.

The jays' memories seem to show flexibility, another quality of human episodic memory, Clayton and her colleagues reported in the January 2003 Journal of Experimental Psychology: Animal Behavior Processes. Clayton's researchers this time let their jays cache perishable crickets and long-lasting nuts in two trays. Then, the researchers replaced the crickets that the birds had hidden in one of the trays with decayed crickets to make it seem as if the treats degrade extrafast. After the birds harvested that tray, they were given a chance to recover the first tray's caches in the period between the two apparent spoilage dates, these birds incorporated the new information and focused mostly on the less perishable nuts.

The jays time travel into the future, too, the researchers contend. In an experiment, some birds seemed to anticipate problems yet to come. Clayton and her colleague Nathan Emery looked at how the jays reacted to the possibility that some sleazy neighbor would steal a food cache.

Some of the jays in the laboratory had criminal histories of snagging food that another bird had buried. The researchers let the criminal jays as well as ones with clean records hide treats. When the birds stashed their seeds in private, they didn't take opportunities to move their treasure to a new hiding place. However, if the researchers let another bird get near enough to watch the caching, the criminal jays took the next opportunity to recover the treat and hide it in a different place. The birds with no experience of thievery didn't recache.

Evidence for mental time travel in animals hasn't convinced everybody, though. Suddendorf, for example, says that although he doesn't think mental time travel is theoretically impossible for animals, experiments so far haven't shown it conclusively. The jays in Clayton's work might be retrieving their cache information by some process very different from episodic memory, he and Janie Busby caution in the September 2003 Trends in Cognitive Science.

Clayton is careful not to say that animals mentally travel in time the same way people do. The two phenomena "share elements" as she puts it. Still, she argues for continuing to push questions about animals' mental capabilities.

CACHE FLOW Birds vary in their caching enthusiasm--and their prowess. Even the families of such storied cachers as the nutcrackers and the black-capped chickadee contain species that manage only halfhearted food storage, if that. Some of these lackluster relatives retrieve their food in a matter of hours instead of waiting weeks or months, and some don't hide food at all.

Birds' different environments may be responsible for this variety. Skill in hiding food and remembering where to find it could make a big difference in a harsh climate with unpredictable food supplies, explains Vladimir V. Pravosudov of the University of California, Davis. He and Clayton found some supporting evidence of this notion in their study of chickadees published in 2002.

The researchers caught black-capped chickadees in southern Alaska and in the milder climate of lowland Colorado. They brought all the chickadees to Davis to live in their laboratory with the same food-and-light regimen for 45 days. Then, the researchers pitted the two groups against each other in various tests.

The Alaskan chickadees outperformed the Coloradoans on seed storage and retrieval. The Alaskans cached about twice as many seeds during the hiding sessions and later recovered them with only a quarter as many tries per seed. Yet the Alaskans didn't seem smarter in mental aspects other than their spatial memories. When the birds had to learn to associate a color with a food, the two groups of chickadees came out about the same. However, the brains of the Alaskan chickadees had a larger hippocampus, an area associated with memory, say the researchers. They argue that their evidence supports the idea that tougher life up north has focused stronger evolutionary forces on the local birds' caching capabilities.

Pravosudov is also examining the physiology of caching. Working with mountain chickadees caught in California, he has inserted implants that release corticosterone, a hormone associated with stress. Researchers studying seed-caching willow tits found that blood concentrations of corticosterone double during the worst winter months.

Pravosudov's souped-up chickadees, which had roughly double the normal blood concentrations of corticosterone, beat the all-natural birds at the caching tests, he reports in Dec. 22, 2003 Proceedings of the Royal Society of London B. The birds with implants cached more than twice as many seeds as the other chickadees did and needed only half as much probing around before finding the right spots to recover their bounty. Thus, winter food shortages that Pravosudov has found to drive up hormone concentrations could trigger a potentially lifesaving burst of memory improvement.

The idea of links between tough times and caching wizardry has kicked up what Clayton calls a lively debate. Johan J. Bolhuis of Utrecht University, for example, cautions against reading too much into the recent results. "As the authors of the chickadee paper acknowledge, there are many differences between the Alaska and the Colorado chickadees apart from differences in food storing," he says. As for the hippocampus connection, Bolhuis points out that other studies have not found a link between a large hippocampus and seed-caching prowess. The data so far aren't sufficient to show that the hippocampus is the right area to monitor for bird memory, he adds.

To resolve the question of whether tough times have contributed to the evolution of caching wizardry is "currently difficult," says Clayton. But biologists do have marvelous phenomena to examine experimentally. Birds will cache seeds in the lab without training, and researchers can raise flocks under tight controls with clear records.

What started out as a fidgety search for the operating rules of feathered robots has turned into studies of how thinking works.