Neuro: E/I Microcircuit

This scenario simulates a small network with excitatory (E) and inhibitory (I) neurons. It’s designed to show how inhibition can stabilize activity and how networks can exhibit emergent patterns you don’t see in a single neuron.

What you’re simulating

Excitatory neurons tend to amplify activity.
Inhibitory neurons counterbalance that activity.
Together, they can form a regime that is stable, oscillatory, or runaway depending on parameters.

Run it (web UI)

Open Examples
Choose Neuro → E/I Microcircuit
Click Run

Run it (local SimUI)

From the B‑Simulant library repo:

pip install "bsim[ui]"
python examples/neuro_simui_demo.py --mode circuit --port 8765

Open http://localhost:8765/ui/.

What results to expect

Raster plots: which neurons spike and when (E vs I may differ).
Population rate: how overall activity rises, falls, or stabilizes.
Voltage traces: sample neurons show how spikes are generated.

Parameter presets (and what they mean)

Goal	Steps	dt (s)	Expected result	What it means
Quick preview	1,500	0.0001	Some activity + a rough rate curve.	Useful to confirm the circuit is active, not to judge stability.
Standard run	3,000	0.0001	Sustained spiking with a bounded population rate.	A balanced regime: inhibition prevents runaway excitation.
Longer run	20,000	0.0001	Stable, oscillatory, or drifting regimes become obvious.	Long runs reveal whether “balance” holds or slowly fails.
Faster (less detail)	3,000	0.0002	Activity may look noisier; timing less precise.	Larger dt is faster but can blur timing-dependent effects.

How to interpret what you see

If activity dies out, inhibition or insufficient drive may be too strong.
If activity explodes, inhibition may be too weak (or excitation too strong).
If activity settles into a rhythm, you may be seeing a stable balance or oscillatory regime.

This example is a good mental model for “balance” in real circuits: inhibition isn’t just stopping spikes, it can shape and stabilize computation.

Neuro: Single Neuron