Press "Enter" to skip to content

Rats use mental schemas to speed learning

How easy is it to make a memory? We've previously discussed how episodic memories form in the hippocampus, and gradually get translated into long-term memory in the cortex. Research that will appear later today in Science suggests that the process of consolidating these memories doesn't have to be gradual. Instead, if we have previous long-term memories that provide a framework for understanding the new information—a mental schema—then we can solidify new memories rapidly. HangZhou Night Net

The researchers worked with rats that were placed in an "arena" containing a number of sand-filled cups. Six types of food were used in the experiments, and each type was consistently hidden in the same cup (bacon was always in cup six, for example); only one type of food was present in the arena at a given time. As they were placed in the arena, the rats were given a small sample of the type of food that was hidden in that test. The researchers then determined whether the small taste was enough to allow the rats to go directly to the cup that contained more of that food.

Over the span of a month, the rats improved their accuracy so that whenever they were given a priming taste of a specific type of food, they zeroed in on the appropriate location at a rate well above random. This learning depended on the normal memory process, as rats with lesions in the hippocampus stayed at levels of accuracy consistent with random searching.

But the rats had apparently learned more than the simple taste/location association;: they had learned to understand the process. After a single example using a new food type was provided, the rats were able to go to the correct location at the same rate as their success with the original foods. By understanding the process via a mental schema, the rats were able to form new associations almost instantly.

How does this play out on the biological level? It turns out that once a schema is in place in the cortex, the process of transferring memories that fit the schema from the hippocampus to the cortex occurs rapidly. Lesions to the hippocampus within three hours of the first use of a new taste cue still blocked the process, but lesions made 24 hours after didn't, suggesting that less than a day is required to consolidate schema-based memories.

Distilled down to its most basic principles, the data is not much of a surprise: how an animal learns depends on what it already knows. But the authors suggest this plays out in more complex ways in humans using an example any scientist reading the paper would understand, that of comprehending a seminar. "Our ability to do so," they write, "depends as much on our possession of an appropriate mental schema as on the communicative skill of the speaker in logically conveying his or her message. In the absence of such mental frameworks, we are unable to follow scientific inferences in a talk and have the phenomenological experience of being functionally amnesic for its content a surprisingly short time later."

In other words, if we can't fit new information into the mental schemas we've developed over the years, we might as well not be listening, because we'll forget it all shortly anyway.