海角大神

Bees can experience false memories, scientists say

Researchers at Queen Mary University of London have found the first evidence of false memories in non-human animals.

What do bees and Brian Williams have in common?

Turns out, they are both prone to "conflating" their memories.

It has long been known that humans 鈥� even those of us who aren't famous news anchors 鈥� tend to recall events that did not actually occur. The same is likely true for mice: In 2013, scientists at MIT聽induced false memories of trauma in mice, and the following year, they used light to manipulate mice brains to turn painful memories into pleasant ones.

Now,聽researchers at Queen Mary University of London have shown for the first time that insects, too, can create false memories. Using a classic Pavlovian experiment, co-authors Kathryn Hunt and Lars Chittka determined that bumblebees sometimes combine the details of past memories to form new ones. Their were published today in聽Current Biology.

鈥淚 suspect the phenomenon may be widespread in the animal kingdom,"聽Dr. Chittka said in a written statement to the聽Monitor.

First, Chittka and Dr. Hunt trained their buzzing subjects to expect a reward if they visited two artificial flowers 鈥� one solid yellow, the other with black-and-white rings. The order didn鈥檛 matter, so long as the bee visited both flowers. In later tests, they would present a choice of the original two flower types, plus one new one. The third type was a combination of the first two, featuring yellow-and-white rings. At first, the bees consistently selected the original two flowers, the ones that offered a reward.

But a good night鈥檚 sleep seemed to change all that. One to three days after training, the bees became confused and started incorrectly choosing the yellow-and-white flower (up to fifty percent of the time). They seemed to associate that pattern with a reward, despite having never actually seen it before. In other words, the bumblebees combined the memories of two previous stimuli to generate a new, false memory.

鈥淏ees might, on occasion, form merged memories of flower patterns visited in the past,鈥� Chittka said. 鈥淪hould a bee unexpectedly encounter real flowers that match these false memories, they might experience a kind of deja-vu and visit these flowers expecting a rich reward.鈥�

Bees have a rather limited brain capacity, Chittka says, so it鈥檚 probably useful for them to 鈥渆conomize鈥� by storing generalized memories instead of minute details.

鈥淚n bees, for example, the ability to learn more than one flower type is certainly useful,鈥� Chittka said, 鈥渁s is the ability to extract commonalities of multiple flower patterns. But this very ability might come at the cost of bees merging memories from multiple sequential experiences.鈥�

Chittka has studied memory in bumblebees for two decades. Bees can be raised and kept in a lab setting, so they make excellent long-term test subjects.

鈥淭hey are [also] exceptionally clever animals that can memorize the colors, patterns, and scents of multiple flower species 鈥� as well as navigate efficiently over long distances,鈥� Chittka said.

In past studies, it was assumed that animals that failed to perform learned tasks had either forgotten them or hadn鈥檛 really learned them in the first place. Chittka鈥檚 research seems to show that animal memory mechanisms are much more elaborate 鈥� at least when it comes to bumblebees.

鈥淚 think we need to move beyond understanding animal memory as either storing or not storing stimuli or episodes,鈥� Chittka said. 鈥淭he contents of memory are dynamic. It is clear from studies on human memory that they do not just fade over time, but can also change and integrate with other memories to form new information. The same is likely to be the case in many animals.鈥�

Chittka hopes this study will lead to a greater biological understanding of false memories 鈥� in animals and humans alike. He says that false memories aren鈥檛 really a 鈥渂ug in the system,鈥� but a side effect of complex brains that strive to learn the big picture and to prepare for new experiences.

鈥淓rrors in human memory range from misremembering minor details of events to generating illusory memories of entire episodes,鈥� Chittka said. 鈥淭hese inaccuracies have wide-ranging implications in crime witness accounts and in the courtroom, but I believe that 鈥� like the quirks of information processing that occur in well known optical illusions 鈥� they really are the byproduct of otherwise adaptive processes.鈥�

鈥淭he ability to memorize the overarching principles of a number of different events might help us respond in previously un-encountered situations,鈥� Chittka added. 鈥淏ut these abilities might come at the expense of remembering every detail correctly.鈥�

So, if generating false memories goes hand in hand with having a nervous system,聽does all this leave Brian Williams off the hook?

鈥淚t is possible that he conflated the memories,鈥� Chittka said, 鈥渄epending on his individual vulnerability to witnessing a traumatic event, plus a possible susceptibility to false memories 鈥� there is substantial inter-person variation with respect to this. It is equally possible that he was just 鈥榮howing off鈥� when reporting the incident, and is now resorting to a simple lie to try to escape embarrassment. That is impossible for me to diagnose.鈥�

But if Mr. Williams genuinely did misremember his would-be brush with death, Chittka says he shouldn鈥檛 be vilified.

鈥淵ou cannot morally condemn someone for reporting something they think really did happen to them,鈥� Chittka said. 鈥淵ou cannot blame an Alzheimer patient for forgetting to blow out the candle, even if they burn down the house as a result. In the same way, you can't blame someone who misremembers a crime as a result of false memory processes."