Forgetting has its own engram—and the brain can tune it
A study in the dentate gyrus argues that what we call an “engram” may not be a single memory-supporting unit. Instead, the authors describe two intermingled neuronal ensembles: one that supports later recall and another that actively promotes forgetting. They report that shifting the balance between these ensembles via Rac1 signaling can bias behavior toward remembering or forgetting, and they link dysfunction of the forgetting-promoting ensemble to abnormal memory phenotypes in mouse models relevant to Alzheimer’s disease and autism.
Why it might matter to you:
If your work models memory stability over time, this “dual-ensemble” framing offers a concrete neural target for separating loss of information from impaired access to it. It also suggests specific molecular levers (e.g., Rac1-linked pathways) that could be incorporated into mechanistic hypotheses about when and why memories fade—or persist in maladaptive forms.
Dopamine doesn’t speed up the clock—it moves the finish line
Using optogenetics in mice performing a timed lever-press task (targeting ≥3 seconds), researchers tested how substantia nigra dopamine neurons shape interval timing. Exciting tyrosine hydroxylase–positive neurons shifted produced intervals later, while inhibiting them shifted timing earlier. A drift–diffusion timing model fit to behavior indicated these manipulations changed the decision threshold for ending the timed action—raising it with excitation and lowering it with inhibition—without altering the rate of temporal integration (“clock speed”).
Why it might matter to you:
If you are building systems-level accounts that relate neuromodulation to learning and memory, this result narrows dopamine’s timing role to a control parameter you can formalize as a bound/criterion rather than a pacemaker. It may also help interpret timing-related phenotypes in dopaminergic disease models by separating altered accumulation from altered decision policy.
Cells that don’t code proteins can still run the nucleus
A PNAS paper reports that a long noncoding RNA produced via intronic polyadenylation from a protein-coding locus can influence nucleolar integrity and function. The work highlights a route by which transcripts derived from within coding genes—yet not translated into protein—can exert regulatory effects inside the nucleus. The authors frame this as a reminder that “noncoding” output from coding regions can carry distinct biological roles rather than being mere byproducts of transcription.
Why it might matter to you:
For neuroscience-relevant gene regulation, this expands the space of candidate mechanisms: activity-dependent changes could alter not only protein abundance but also noncoding isoforms with organelle-level effects. It’s a prompt to consider whether transcript-processing choices (like intronic polyadenylation) could be coupled to long-timescale cellular state changes that influence plasticity.
Stay curious. Stay informed — with
Science Briefing.
Always double check the original article for accuracy.