Fly, monkey, human: brain-circuit studies converge on measuring and moving circuits
A fly model shows repair needs the right cells, a monkey platform records across regions, and human studies track aging, stimulation and anesthesia in circuit terms.
- In a preprint fly-steering model, losing 2,000 neurons at a sensorimotor site cut simulated displacement to 6.5% of intact levels. Self-reported m:…5688
- Replacing 100 cells with their original connections restored displacement to 62.5% of intact, compared with 10.1% for a random selection of 100 cells. Self-reported m:…5688
- A preprint platform called MIND enables stable multi-probe recordings across cortical regions and durations in rhesus macaques, designed for chronic, ambulatory, head-free recordings. Self-reported m:…7010
- Alpha-frequency 10 Hz repetitive TMS to the intraparietal sulcus specifically increases working memory precision in a preprint study. Self-reported m:…6862
- Under propofol anesthesia, stimulus-rate power and phase alignment fall in human auditory cortex and acoustic-envelope decoding accuracy drops, in a preprint study. Self-reported m:…5367
One question at five scales
Five preprints posted to bioRxiv take up the same question at different scales: how do brain circuits break, and what does it take to fix, stimulate or read them? The studies span a fly steering circuit, recordings in monkeys, aging in humans, noninvasive stimulation and anesthesia. Read together, they suggest the field is moving from mapping brain areas to testing circuits directly. m:…5688 m:…7010 m:…9127
The fly study makes the repair question concrete. In a connectome model of fly steering, focal neuronal loss of 2,000 neurons at a sensorimotor site reduced simulated displacement to 6.5% of intact levels. Replacing 100 cells with their original connections restored displacement to 62.5% of intact, compared with 10.1% for a random selection of 100 cells. The authors report that cell identity and type-dependent integration are required for functional recovery in the connectome-constrained model, which means replacement alone is not enough: the replacement has to be the right cells, wired the right way. m:…5688
Recording across regions in moving animals
The monkey study attacks the measurement side. A platform called MIND enables stable multi-probe recordings across cortical regions and durations in rhesus macaques, and the platform is designed to support chronic, ambulatory, head-free recordings in non-human primates. That design matters because it moves recording out of the restrained setup and into animals that move freely, which is where natural behavior happens. m:…7010
The same preprint introduces UnitAssay, a tool that enables automated curation of large-scale spike-sorting output and improves reliable single-unit identification relative to Kilosort 4 labels and standard SpikeInterface quality control. In practice, that is the unglamorous step that decides whether a multi-region recording yields trustworthy single neurons or a pile of sorting errors, and the preprint reports improvement on exactly that step. m:…7010
Aging, stimulation and anesthesia in humans
Two human studies track how circuits shift with age and intervention. Cerebello-basal ganglia functional connectivity is higher in early middle-aged adults compared with older adults, and connectivity correlates with Stroop task performance, with stronger associations in early middle-aged adults. Progesterone levels are positively associated with cerebello-basal ganglia functional connectivity in females, which points to hormonal state as one of the variables behind the age pattern rather than age alone. m:…9127
Stimulation, in the preprint on repetitive transcranial magnetic stimulation, works through frequency. Alpha-frequency 10 Hz stimulation to the intraparietal sulcus specifically increases working memory precision, and rhythmic stimulation at theta 6 Hz or alpha 10 Hz frequencies reduces response error, guesses and swaps in working memory tasks compared with arrhythmic controls. The contrast with arrhythmic stimulation is the key detail: it is the rhythm, not just the stimulation, that carries the effect. m:…6862
Anesthesia, in the human sEEG preprint, shows up as a loss of signal alignment rather than a simple switch-off. Propofol-based anesthesia reduces stimulus-rate power and phase alignment in human auditory cortex, the cortical distance-response relationship from the posteromedial Heschl's gyrus is significantly weakened under propofol anesthesia compared with wakefulness, and low-dimensional embeddings show lower acoustic-envelope decoding accuracy under anesthesia. The implication is that depth of anesthesia can be read from how well cortical signals stay aligned with sound, not just from whether they respond. m:…5367
Why the set matters together
The shared move across all five preprints is to treat a circuit as something to be perturbed and measured, not just described. The fly model perturbs by removing and replacing cells, the monkey platform measures across regions in freely moving animals, the stimulation study perturbs by frequency, and the anesthesia study measures the loss of alignment. For anyone building brain-computer interfaces or reading the neuroscience literature, that is the direction to watch: claims that come with a perturbation and a measurement, rather than a map alone. m:…5688 m:…7010 m:…6862 m:…5367
What is still unresolved
- All five studies are preprints and have not been peer-reviewed.
- Whether the fly-model repair rules generalize to mammalian circuits.
- Whether the stimulation and anesthesia measures hold outside the tested tasks and conditions.
Provenance by Synorb. Every claim links to its source. This story was written by an agent from the Manifests listed below. Read the methodology and limitations, or send a correction.
Free to reuse, including commercially. Credit “Hanging Context, with provenance by Synorb” and link to this story. Keep the cited Manifest ids. Reuse terms v1.