Prediction-Error Neurons Whisper Our Mistakes

Summary: A new study identified a distinctive class of neurons, termed “prediction-error neurons,” which activate solely when audible expectations are breached, signaling an error in anticipation.

These neurons remain dormant and unresponsive to typical sounds, springing to life only when an auditory outcome deviates from the expected. Through a methodical study involving mice, researchers explored how these neurons might illuminate various learning processes and potentially explain discrepancies in abilities like speech and music across individuals.

The findings spotlight the potential role of these neurons in learning through trial, error, and unexpected auditory outcomes.

Key Facts:

  1. Single-Purpose Neurons: “Prediction-error neurons” are exclusive in their function, remaining silent until a sound defies expectation, thus signaling an error.
  2. Versatility in Error Recognition: Different subsets of these neurons respond to various types of auditory discrepancies, such as unexpected quietness or timeliness of a sound.
  3. Potential Learning Implications: These neurons could be pivotal in understanding and refining learning processes involving trial and error in sound-associated behaviors, like speech and musical instrument playing.

Source: NYU

Whether improperly closing a door or shanking a kick in soccer, our brains tell us when we’ve made a mistake because these sounds differ from what we expect to hear. While it’s long been established that our neurons spot these errors, it has been unclear whether there are brain cells that have only one job—to signal when a sound is unexpected or “off.”

A team of New York University neuroscientists has now identified a class of neurons—what it calls “prediction-error neurons”—that are not responsive to sounds in general, but only respond when sounds violate expectations, thereby sending a message that a mistake has been made.

This shows a person.
More specifically, they found that individual prediction-error neurons in the mice’s auditory cortex not only signaled when something went wrong, but they also signaled what went wrong. Credit: Neuroscience News

“Brains are remarkable at detecting what’s happening in the world, but they are even better at telling you whether what happened was expected or not,” explains David Schneider, an assistant professor in NYU’s Center for Neural Science and the senior author of the study, which appears in JNeurosci.

“We found that there are specific neurons in the brain that don’t tell you what happened, but instead tell you what went wrong.”

The paper’s authors add that the results could potentially help better illuminate the learning process, identify the causes behind certain afflictions, and spot sound-related aptitudes. 

“Neurons like these might be vital in learning how to speak or how to play a musical instrument,” observes Nicholas Audette, a postdoctoral fellow in NYU’s Center for Neural Science and the paper’s lead author.

“Both of those behaviors involve lots of trial and error, lots of mistakes, and lots of learning from mistakes.”Schneider added, “Do expert musicians have better prediction error neurons than novices? And in diseases in which speech is underdeveloped, are prediction error neurons malfunctioning?”

Behaviors often have predictable sensory consequences. For example, when shutting a car door, we expect to hear an anticipated “thump” at a particular phase of our arm movement. And these behaviors are not limited to humans—monkeys, mice, and other animals can learn to predict what sound a movement will produce and when that sound will occur. 

Previous research has shown that in the brains of humans and other animals, neurons have significant responses when a sound violates the animal’s expectation and weaker responses when a sound matches expectation. But it had been unclear whether there were neurons that only had one job—to signal when a sound was unexpected.

To address this question, Schneider and Audette built upon their previous work, which uncovered how the brain makes distinctions between “right” and “wrong” sounds. In the new JNeurosci study, they studied responses in mice through a series of experiments and by isolating neuronal activity.

In the experiments, the mice heard a particular sound after pressing on a lever. This was done repeatedly, until they associated this sound when pressing. The researchers then altered the sounds the mice heard in making subsequent presses—a method designed to mimic unexpected sounds humans hear when making errors.

The scientists found that many of the mice’s neurons (“prediction-error neurons”) were silent—meaning that they did not produce signals that could be used by other parts of the brain—except for when the mice made a movement and heard something unexpected. 

More specifically, they found that individual prediction-error neurons in the mice’s auditory cortex not only signaled when something went wrong, but they also signaled what went wrong. 

For example, every time the sound was too quiet, one group of prediction-error neurons was activated. However, when the sound was the expected volume but came too late, a completely different group of prediction-error neurons was active.

“When a movement makes an unexpected sound, it can violate our expectations in a lot of different ways,” explains Schneider. “Different neurons are active when a movement makes too quiet a sound, and other neurons when the movement makes the wrong sound.”

Funding: This research was supported by a grant from the National Institutes of Health (R01-DC018802).

About this neuroscience research news

Author: James Devitt
Source: NYU
Contact: James Devitt – NYU
Image: The image is credited to Neuroscience News

Original Research: Closed access.
Stimulus-specific prediction error neurons in mouse auditory cortex” by David Schneider et al. Journal of Neuroscience


Stimulus-specific prediction error neurons in mouse auditory cortex

Comparing expectation with experience is an important neural computation performed throughout the brain and is a hallmark of predictive processing. Experiments that alter the sensory outcome of an animal’s behavior reveal enhanced neural responses to unexpected self-generated stimuli, indicating that populations of neurons in sensory cortex may reflect prediction errors – mismatches between expectation and experience.

However, enhanced neural responses to self-generated stimuli could also arise through non-predictive mechanisms, such as the movement-based facilitation of a neuron’s inherent sound responses.

If sensory prediction error neurons exist in sensory cortex, it is unknown whether they manifest as general error responses, or respond with specificity to errors in distinct stimulus dimensions.

To answer these questions, we trained mice of either sex to expect the outcome of a simple sound-generating behavior and recorded auditory cortex activity as mice heard either the expected sound or sounds that deviated from expectation in one of multiple distinct dimensions.

Our data reveal that the auditory cortex learns to suppress responses to self-generated sounds along multiple acoustic dimensions simultaneously.

We identify a distinct population of auditory cortex neurons that are not responsive to passive sounds or to the expected sound but that encode prediction errors. These prediction error neurons are abundant only in animals with a learned motor-sensory expectation and encode one or two specific violations rather than a generic error signal.

Together, these findings reveal that cortical predictions about self-generated sounds have specificity in multiple simultaneous dimensions and that cortical prediction error neurons encode specific violations from expectation.

Join our Newsletter
I agree to have my personal information transferred to AWeber for Neuroscience Newsletter ( more information )
Sign up to receive our recent neuroscience headlines and summaries sent to your email once a day, totally free.
We hate spam and only use your email to contact you about newsletters. You can cancel your subscription any time.