summary: Using mice, researchers found that passive exposure alongside active training can significantly improve the learning process. This study shows how passive exposure to stimuli such as sounds and language helps the brain form basic representations, making active learning more efficient.
The findings are consistent with previous research in humans, showing that combining low-effort passive exposure with active training accelerates the acquisition of new skills, such as learning a musical instrument or a foreign language. suggests that it is possible.
Important facts:
- In addition to active training, mice that were passively exposed to the sound learned to associate the sound with the reward faster.
- Artificial neural network simulations show that passive exposure creates a fundamental representation of the stimulus in the brain.
- The insights of this study are consistent with human studies and suggest that an approach that combines passive exposure and active training may enhance learning of complex skills.
sauce: University of Oregon
Learning a new skill requires deliberate practice over time, but passive exposure to the subject matter at hand can potentially speed up that process. This is suggested by a new study from the University of Oregon.
This finding builds on previous research in humans and shows how passive exposure can be a valuable tool for learning. How watching foreign language movies can supplement grammar drills and vocabulary flashcards, and how listening to recordings of professional piano concertos can help budding musicians improve their skills. This will help explain how it works.
The study provides further insight into the brain mechanisms that may be behind the effects and helps scientists understand why passive exposure is so powerful, said the same University of Neurosciences. said James Murray, a neuroscientist at the university who led the research along with Santiago Jaramillo, also of the Faculty of Arts. And science.
It’s much easier to study what’s going on in the brains of rodents than in humans, so we’ll discuss how both active training and passive exposure affect learning in mice. “Studying both of them opens up exciting possibilities for studying the neural mechanisms underlying the interactions between them,” Murray added.
The researchers describe their findings in a paper published in the journal e-life.
To study how mice learn, the researchers trained mice to respond to sounds that rose and fell in pitch for rewards in specific locations. All mice underwent an active training protocol during which they received feedback on their performance so they knew whether they had made the right choice. Some mice were also given passive exposure to listening to sounds when not engaged in a task.
The researchers showed that mice that were passively exposed to the sound in addition to active training learned how to choose the location of the reward faster. It did not seem to matter whether the passive exposure occurred at the beginning of training or was scattered in small chunks throughout the active training session.
Next, to better understand how learning occurs in the brain, the researchers trained and tested various artificial neural networks on simulated versions of learning tasks. Neural networks, a type of machine learning algorithm, process information in a way that mimics the way the brain processes information.
Artificial neurons represent real neurons, and learning occurs by changing the strength of connections between those neurons. Although these are not direct replicas of the brain, they can be used to generate hypotheses that can be tested experimentally.
This modeling suggests that passive exposure to a stimulus lays the groundwork for the brain to create hidden representations that capture the most salient features of that stimulus, much like drawing an outline with a pencil before diving into a detailed painting. I am. Then, during active learning, the brain associates the stimulus with a specific action. Passive exposure primes the brain to make those connections more quickly.
In the future, the researchers hope to record brain activity in mice during similar learning tasks to see if their predictions hold true.
Although the study was conducted using a simple task in mice, the researchers suggest that the findings may also have implications for more complex learning in humans. Study co-author Melissa Bays-Burke, a former UO linguist now at the University of Chicago, wrote research showing how passive exposure helps adults learn to understand new speech sounds. has been announced previously.
“Along with previous studies in humans by Melissa and colleagues, our results demonstrate that combining low-effort passive exposure with active training results in relatively low “It suggests that a given performance threshold can be achieved with effort,” Murray said.
“This insight could help humans learn musical instruments or second languages, but we wonder how this applies to more complex tasks, and how to combine passive learning with active training. Further research is needed to better understand how to optimize training schedules.”
About this learning and neuroscience research news
author: laurel hammers
sauce: University of Oregon
contact: Laurel Hammers – University of Oregon
image: Image credited to Neuroscience News
Original research: Open access.
“Passive exposure to task-relevant stimuli enhances classification learning” by James Murray et al. e-life
abstract
Passive exposure to task-relevant stimuli enhances classification learning
Learning how to perform perceptual decision tasks is typically accomplished through intense practice sessions with feedback.
Here, we investigated how passive exposure to task-relevant stimuli, which is relatively effortless and does not require feedback, influences active learning.
First, we trained mice on a sound classification task on various schedules combining passive exposure and active training.
Mice that received passive exposure showed faster learning, regardless of whether this exposure occurred completely before active training or was sandwiched between active sessions.
We then trained neural network models using different architectures and learning rules to perform the task. A network that uses the statistical properties of the stimuli to increase the separability of the data through unsupervised learning during passive exposure provided the best explanation of the behavioral observations.
Additionally, during interleaved schedules, there is better consistency between passive exposure and weight updates from active training, and some interleaved sessions are identical to schedules with long periods of passive exposure before active training. I found it to be quite effective. This is consistent with our behavioral patterns. observation.
These results provide important insights into the design of efficient training schedules that combine active learning and passive exposure in both natural and artificial systems.