Why study baby rats?

Newborn rats, like human infants, spend a lot of time sleeping and moving — but their sleep looks different. During “REM” (rapid‑eye‑movement) sleep they make tiny, jerky twitches with their paws, whiskers, and tails. Those twitches might look random, but they send a burst of feedback to the brain that helps wire up motor circuits long before the animal can run or climb.

The brain areas we study

Primary Motor Cortex (M1)

In adults, M1 makes most of the movements you make every day.

In pups, it just listens to what the rest of the brain is doing.

We want to know when it stops listening, and starts doing.

The Red Nucleus (RN)

In adults, RN isn't thought to do much

In pups, it seems to do it all while M1 is just listening

Our most recent publication

What we did...

Tiny head‑posts, but freedom to move.
Pups (12–24 days old) were gently head‑fixed in an “air‑floating skateboard” cage, so they could walk, groom, or sleep while we recorded.

Recording spikes and video.
We measured individual neuron firing in M1 (and, later, RN) and filmed every twitch or wake movement with high‑speed cameras.

Sleep vs. wake, day by day.

We tracked how neural responses changed at postnatal days 12, 16, 20, 24.

What we found...

Twitches still matter at 3 weeks old.
Even at P24, about one‑quarter of M1 neurons fire for REM‑sleep twitches. These signals are somatotopically precise (forelimb neurons only fire for forelimb twitches).

M1 responses sharpen with age.
Between P12 and P24 those twitch bursts get shorter and earlier, showing the circuit is refining its timing.

RN fires before the paw moves; M1 mostly fires after.

In simultaneous recordings at P24, RN spikes often came ~20 ms before a twitch or a wake movement, while M1 spikes lagged behind. Translation: RN is still driving the action; M1 is mostly listening.

RN neurons prefer specific wake moves; M1 is more “whatever moves.”
Using machine learning (DeepLabCut), we saw certain RN cells activate only when the pup brought both paws to its face (grooming), but stayed quiet during walking. M1 neurons reacted to many directions and amplitudes without that tight tuning.

The Big Picture...

Learning by Listening (before weaning).

RN sends the commands; M1 records the outcome via sensory feedback from twitches and early wake moves.
Taking the Wheel (after ~P25).

As corticospinal pathways strengthen, M1 should develop the same movement‑specific coding we already see in RN, eventually taking over as the main “driver” of voluntary behavior.

Why it matters...

Understanding how early sleep shapes motor circuits could:

Explain developmental disorders where movement and sleep are both affected (e.g., cerebral palsy, autism).
Inspire new rehab strategies that use structured sleep or “micro‑movement” therapy for preterm infants.
Reveal general rules for how the brain turns feedback into control—a principle that spans robotics, AI, and learning sciences.

Stay tuned—our sleeping rats still have a lot to teach us!

What's next

Sneak peek at some experiments we're considering...

We recently discovered something surprising: in baby rats, the M1 doesn't always act like it's in charge. During REM sleep, it quietly inhibits the RN, a brainstem region that drives movements early in life. But during wakefulness, that connection seems to disappear. Why would M1 only inhibit RN during sleep? And how does that help the brain learn to move?

We’re especially interested in a special kind of RN cell called a PV+ neuron, which may be required for M1 and RN to talk to each other—turning movement signals on or off depending on the sleep state. These neurons are also sensitive to noradrenaline, a brain chemical that’s high during wake and low during REM sleep, which might explain the sleep-only connection.

Our next step is to find out how this connection develops—and what happens when it doesn’t. We'll study baby rats with a mutation in the Fmr1 gene, a known cause of Fragile X Syndrome, the most common inherited form of autism. These pups often have trouble with sleep and movement, just like many kids with autism. We'll record from M1 and RN in both typical and Fmr1 pups at different ages to see how sleep, movement, and brain activity change over time.

Finally, because both M1 and RN get strong signals from the cerebellum—a brain region that helps fine-tune movement—we're starting to investigate how it fits into the story. Does the cerebellum help M1 and RN talk to each other during sleep? We're trying to find out.

Stay tuned—our rats are teaching us that sleep isn’t just rest. It’s rehearsal.