Neuron Signal Shapes May Reveal Hidden Brain Information

What more could be concealed within the electrical language of the brain’s cells?

Neuroscientists commonly log neuronal activity as a sequence of straightforward occurrences: a cell either emits an electrical pulse or it does not. The timing and frequency of those pulses can reveal a great deal about the brain.

Yet a new preprint study suggests this method could miss an important detail.

The millisecond-long electrical flashes generated by neurons are not necessarily alike. Their forms appear to provide hints about what a cell has experienced recently.

In other words, an electrical signal may hold more information than the binary marker typically used to represent it.

University of California, San Diego neuroscientist Bradley Voytek told ScienceAlert that these electrical pulses, called action potentials, appear to be simple, consistent spikes only from a distant view.

The research has not yet undergone peer review, and its authors advise caution until it does. Even so, it contributes to an increasingly detailed picture of how much information an individual neural signal may hold.

"An action potential only looks like a 'spike' when you are zoomed out, but when you zoom in on it, and look at its full waveform, which is only about a millisecond long, the shape of that voltage trace tells you something about what's happened recently to that neuron," Voytek said.

"When we turn an action potential into a binary event, we throw away information about its recent input."

Action potential shapes reveal recent electrical input

The team initially studied recordings from two mouse neurons in a brain region involved in vision. As the cells received different electrical currents, features of their signals shifted, including their height and the speed at which the voltage declined.

The currents consisted of a constant input, one that rose gradually, and a variable input called pink noise. Recordings made within the cells by other researchers showed subtle alterations in every electrical pulse.

For one neuron, a machine-learning model used signal shape alone to tell apart three forms of electrical stimulation with 78.4 per cent accuracy. The recordings from the second neuron were too limited to carry out the same analysis.

Signal shape also indicated the strength and variability of the incoming current. Some indications extended as far back as half a second before an electrical pulse - far longer than the pulse itself, which lasts around a millisecond.

The researchers next assessed brain recordings from rats. Rather than investigating cells supplied with a particular electrical current, they examined neurons that were active independently.

Of the 40 neurons assessed, 30 produced signals that separated into more than one clearly distinct shape group.

For a small number of neurons, the variation was large enough that signals from one cell could appear as different from one another as signals from different neurons. In most cells, however, the differences were more subtle.

Neuron signal shapes and nearby brain activity

The researchers also matched signal shape against electrical activity surrounding each cell. In most neurons, the shape predicted the strength of a local electrical signal more effectively than the interval between pulses alone.

This connection was statistically detectable in 87 to 90 per cent of cells, though its predictive strength was generally limited.

In the cell showing the strongest association, pulse shape explained about 64 per cent of the variation in one measurement of nearby electrical activity. However, this finding came from a selected cell, and the majority of cells had far weaker associations.

Do neurons, then, alter the form of their signals to transmit a concealed message to each other?

That remains the unresolved part of the question.

"Our data can't directly address whether the spike shape is communicating information to other neurons, unfortunately," Voytek said.

What the research does establish is that signal shape is connected to the electrical input a cell has recently received, as well as activity in the surrounding area.

Voytek cites earlier research showing that altering the width of an electrical pulse can change the amount of calcium entering a neuron, potentially affecting what it relays to other cells. Whether this is occurring in the present case still requires testing.

Limits for brain recordings and neuron identification

The recordings from experiments involving different electrical currents were taken from only two cells, and both datasets had originally been gathered for other reasons. As a result, it is not yet clear how common these patterns may be across other brain regions or neuron types.

The findings also have another implication.

Scientists can use signal shapes in brain recordings to determine how many distinct neurons they are detecting. If one cell is capable of producing signals with varying shapes, that work may sometimes be more difficult than previously assumed.

The key question is still unanswered. Recording the moment when a neuron generates a signal may mean overlooking information contained in its shape. The form of the electrical spark could offer another clue to the brain’s electrical language.

A preprint of the study is available on bioRxiv.

This article was fact-checked by Rebecca Dyer and edited by Rebecca Dyer. While we take pride in our process, we are only human. If you notice an error, please let us know.