The Math Trick of Neuron No. 12
Despite its romantic name, the “nightingale grasshopper” Chorthippus biguttulus has by no means a golden throat. Rather, the males of this species – as do all grasshopper species – rub their hind leg against the veins of the forewing, producing rasping or chirping sounds, the loud ‘syllables’ of which are interrupted by quiet ‘pauses’. “These calls serve to communicate with animals of the same species,” reports Andreas Herz, professor at the Bio Center of LMU Munich, and coordinator of the Bernstein Center for Computational Neuroscience Munich.
Grasshoppers even use their call as means to distinguish between different species. Naturally, this is a vital issue for females. “They have to make sure they pair up only with a male of the same species,” says Herz. “Any other outcome would literally be love’s labor lost.” But the females are faced with a problem: Grasshoppers cannot regulate their own body temperature. This means that, in the shade or on colder days, their biochemical processes slow down – which means the grasshoppers’ call, too, becomes longer and more drawn out.
The absolute length of a signal in the call can therefore hardly be a factor for distinguishing between species. The pause-to-syllable ratio, in contrast, does not vary with temperature. One would be forgiven for assuming that it would take a number of complex operations to calculate this ratio. The temporal lengths of the syllable and pause would first have to be measured and then divided, one by the other. Since syllables and pauses are offset in time, the result of the first time measurement would also have to be kept in memory until the second time measurement is complete. All together, this is a demanding task, which would appear to require at least a small network of nerves to solve.
“We were able to demonstrate, however, that this computational problem can be solved by a single nerve cell,” reports Herz. “In grasshoppers, we can even identify the responsible neuron: It is ‘ascending neuron No. 12’: This neuron responds to the onset of a call syllable with a burst-like discharge pattern, where the number of spikes within the burst increases linearly with the length of the preceding pause. Fast calls therefore lead to many bursts, each with few spikes, while slower calls lead to fewer bursts, each with many spikes.
If you then sum up the number of spikes over a fixed time window, you obtain an astonishing result: The total number remains the same. This reflects the syllable-to-pause ratio, and accordingly identifies the species of the male grasshopper – irrespective of the speed of its call. “It is certainly a fascinating angle to our work, to have discovered a simple trick that makes complex calculation completely unnecessary, and which might also be exploited by other animals,” states Herz. “The project undertaken with colleagues from Berlin and Göttingen is concluded, yet the name ‘Neuron No. 12’ alone implies there are at least eleven other similar neurons. And we would really like to know what computational problems those neurons solve." (suwe)
“Timescale-Invariant Representation of Acoustic Communication Signals by a Bursting Neuron“”
Felix Creutzig, Sandra Wohlgemuth, Andreas Stumpner, Jan Benda, Bernhard Ronacher, and Andreas V. M. Herz,
The Journal of Neuroscience, February 25, 2009, 29(8):2575-2580;