Beyond the retina: cortex, behavior, and ultrasound
Explain how researchers connect stimulus, neural activity, and behavior.
Loading video…
## From retina to cortex and behavior
Retinal output travels through the thalamus to **visual cortex**, where signals from many ganglion cells are combined into more complex representations. A central goal is to connect three things:
stimulus → neural activity → behavior
**Naturalistic stimuli** and **virtual reality** let researchers present controlled but realistic visual worlds to an animal while recording neural activity and measuring **visual behavior** (for example, orienting toward a target). Linking a representation to behavior requires showing not only that neurons encode a feature, but that the animal's behavior depends on that encoding.
## Ultrasound as a perturbation tool
**Ultrasound** consists of mechanical pressure waves at frequencies above human hearing. **Focused ultrasound** concentrates that energy into a small region, potentially including deep tissue, without surgery to insert electrodes. In experimental preparations, ultrasound can modulate neural activity. One proposed contributor is **acoustic radiation force** — a small, sustained push on tissue that may act on mechanically sensitive ion channels — though the mechanisms are still being investigated and may differ across preparations.
The retina is a useful testbed: ultrasound can be delivered while ganglion-cell activity is recorded with **electrophysiology** or **optical imaging**, so its effect on a well-understood circuit can be measured precisely. This makes ultrasound a promising tool for **mechanistic investigation**, adding a new way to perturb circuits.
Ultrasound neurostimulation is an active research area. It is **not** an established clinical treatment for vision loss or other disorders, and safety, specificity, and mechanism remain open questions.
## The full arc
This primer has followed one chain of reasoning:
FUNCTION → COMPUTATION → MECHANISM → MODEL → PERTURBATION → BEHAVIOR
Each step asks for a different kind of evidence. Research in computational neuroscience moves back and forth along this chain — and the most convincing conclusions come from combining measurement, modeling, and perturbation.
## Further reading
- [Baccus Lab research overview](https://baccuslab.github.io/research/)
- [Baccus Lab publications](https://baccuslab.github.io/publications/)