Chatting with living neurons is like tuning an old radio: it's not enough to find the right frequency: you've got to aim the antenna too. Scientists sent images through electrodes and discovered: neurons ignore a monotonous signal, but short bursts, flashes akin to the impulses of pulsars, are understood with up to 95.6% accuracy. Such a code, with high entropy (a measure of unpredictability), carries far more information than a steady drone. However, just tuning the 'frequency' isn't enough. The spatial arrangement of contacts proved equally important: touching the wrong spot on the electrode array weakened the signal, as if the antenna were pointing away from the station. Most astonishing: neurons that had never encountered light recognized visual patterns, as if an artificial vision had awakened within them. Future biohybrid devices will demand spectroscopic precision (tuning across the frequency spectrum) and a move away from standard models of encoding.
The pistol shrimp hunts by snapping its claw to create a shock wave — an acoustic burst. So it is with neurons: short salvos of impulses cut through the 'noise' better than a monotonous drone.
It's reminiscent of William Gibson's 'Neuromancer,' where biological networks became processors; today, that's lab reality.