Skip to content
Buy Spikeling units

Synapses and inputs

This experiment introduces synaptic integration using Spikeling. You will connect a presynaptic unit to a postsynaptic unit (or emulate the same setup in software) and observe how spikes from one neuron produce input currents and Vm changes in another neuron.

Spikeling simplifies synaptic physiology into a teaching-friendly model:

  • presynaptic spikes arrive as discrete events
  • each spike generates an exponentially decaying synaptic current
  • synapse gain can be positive (excitatory) or negative (inhibitory)

Background reading: Concepts → Synapses and networks.

Learning goals

By the end of this experiment, students should be able to:

  • explain how presynaptic spikes create postsynaptic input currents
  • distinguish excitatory vs inhibitory synapses by sign and effect on Vm
  • demonstrate temporal summation (integration) of repeated synaptic events
  • predict how synapse gain and decay shape postsynaptic spiking
  • build simple network motifs (feedforward excitation, inhibition, E/I balance)

What you need

Hardware route

  • Two Spikeling units (Unit A presynaptic, Unit B postsynaptic)
  • A TRS cable to connect:
  • Axon output (Unit A) → Synapse input (Unit B)
  • Spikeling GUI for monitoring and recording

Emulator route

  • One computer running the GUI
  • Emulator mode with Auxiliary Neuron 1 (and optionally Auxiliary Neuron 2)

Recommended signals to display on the postsynaptic unit: - Vm - Synapse 1 input current / Synapse 2 input current (if available) - Total input current (Itot) (optional but very helpful)

Part A — Wiring (hardware)

  1. Connect Unit A and Unit B to the GUI (or select one to monitor at a time depending on your setup).
  2. Patch a TRS cable:
  3. Axon output (Unit A) → Synapse input 1 (Unit B)

Ground reference

If you see confusing behaviour, confirm that both units share a consistent ground reference through the cable conventions used in your setup (see: Controls and I/O).

Part B — Create a presynaptic spike train

Presynaptic (Unit A)

  1. Choose a neuron mode that spikes reliably.
  2. Use injected current (hardware knob or Patch clamp slider) to produce steady spiking.
  3. Optionally add a low level of noise for realism (but keep it minimal at first).

Target

A stable, regular presynaptic spike train makes postsynaptic effects easier to interpret.

Part C — Excitatory synapse (positive gain)

Postsynaptic (Unit B)

  1. Ensure Unit B is initially below threshold (silent) using its current injection / patch clamp.
  2. Set Synapse 1 gain to a positive value.
  3. Observe Vm on Unit B.

What to look for

  • Each presynaptic spike produces a brief depolarising drive
  • With repeated spikes, depolarisations can summate
  • If strong enough, the synaptic drive can recruit postsynaptic spiking

Teaching tip

Start with Unit B just below threshold. Then increase synapse gain slightly until synaptic input reliably triggers spikes. This makes “synapses can drive firing” very intuitive.

Part D — Inhibitory synapse (negative gain)

  1. Keep the presynaptic spike train the same.
  2. Set Synapse 1 gain to a negative value.
  3. Observe postsynaptic Vm and spiking.

What to look for

  • Each presynaptic spike produces a hyperpolarising effect or a reduction in spiking probability
  • Inhibition can:
  • suppress spiking entirely
  • delay spikes
  • create spike skipping (missed spikes)

Part E — Synaptic summation and time constants (decay)

Concept: synapses are filters in time. If the synaptic current decays slowly, inputs integrate more strongly.

Steps

  1. Set Unit B below threshold.
  2. Keep synapse gain fixed (moderate excitatory).
  3. Increase presynaptic firing rate (increase injection in Unit A).
  4. Observe whether postsynaptic depolarisation builds up.

Expected observations

  • At low presynaptic rate: isolated postsynaptic responses
  • At higher presynaptic rate: responses overlap and summate
  • Summation can push Vm over threshold and trigger postsynaptic spikes

If your build exposes synapse decay parameters in the GUI: - faster decay → sharper, briefer postsynaptic effects (less summation) - slower decay → more integration (more summation)

Part F — Two synapses: coincidence and competition

Spikeling supports two synaptic inputs, which enables simple network motifs.

Motif 1: coincidence detection (two excitatory synapses)

  1. Connect Unit A → Synapse 1 on Unit B.
  2. Connect a second presynaptic unit (Unit C) → Synapse 2 on Unit B (or use emulator Auxiliary Neuron 2).
  3. Set both synapses excitatory at moderate gain.
  4. Adjust presynaptic timing/rates so that spikes sometimes coincide.

Observation: coincident excitatory input can trigger postsynaptic spikes when either input alone cannot.

Motif 2: E/I balance (excitation + inhibition)

  1. Set Synapse 1 as excitatory (positive gain).
  2. Set Synapse 2 as inhibitory (negative gain).
  3. Compare postsynaptic output under different E/I ratios.

Observation: inhibition can veto or shape excitation, controlling spike timing and probability.

Emulator version (no hardware needed)

In emulator mode, reproduce the same logic:

  1. Start emulator and configure the main neuron (postsynaptic).
  2. Open Auxiliary Neuron 1 and drive it to spike (Patch clamp).
  3. Enable coupling (toggle Synapse main neuron).
  4. Use Synapse gain (1 or 2) to set:
  5. excitatory (positive)
  6. inhibitory (negative)
  7. Repeat with a second auxiliary neuron if available to demonstrate E/I balance.

What to measure (simple, robust metrics)

Choose one metric depending on your course level:

  • Postsynaptic spike probability vs synapse gain
  • Postsynaptic firing rate vs presynaptic firing rate
  • Latency from presynaptic spike to postsynaptic spike (when recruited)
  • Summation index: Vm deflection after a burst vs a single presynaptic spike

Discussion prompts

  • Why do synapses need a time course (why not instantaneous)?
  • What changes when synapse sign is inverted?
  • How does presynaptic firing rate interact with synaptic decay to control integration?
  • In what situations might inhibition be more important than excitation?