Why Harmony Pleases the Brain
The key to pleasant music may be that it pleases our neurons. A new model suggests that harmonious musical intervals trigger a rhythmically consistent firing pattern in certain auditory neurons, and that sweet sounds carry more information than harsh ones.
Since the time of the ancient Greeks, we have known that two tones whose frequencies were related by a simple ratio like 2:1 (an octave) or 3:2 (a perfect fifth) produce the most pleasing, or consonant, musical intervals. This effect doesn’t depend on musical training – infants and even monkeys can hear the difference. But it was unclear whether consonant chords are easier on the ears because of the way the sound waves combine in the air, or the way our brains convert them to electrical impulses. A new mathematical model presents a strong case for the brain.
“We have found that the reason for this difference is somewhere at the level of neurons,” says Yuriy Ushakov at the N. I. Lobachevsky State University of Nizhniy Novgorod in Russia.
Ushakov and colleagues considered a simple mathematical model of the way sound travels from the ear to the brain. In their model, two sensory neurons react to different tones. Each sends an electrical signal to a third neuron, called an interneuron, which sends a final signal to the brain. The model’s interneuron fires when it receives input from either or both sensory neurons.
However, the signals from the sensory neurons arrive at the same time if the tone is consonant, and so the interneuron still fires just once, then waits until it “recharges” before it fires again. The result is a regular train of pulses.
By contrast, the signals from dissonant tones arrive at different times and so generate an irregularly spaced train of pulses in the interneuron.
The researchers took their analysis one step further, and calculated the amount of information each signal carries. In the terms of information theory, a random signal carries very little information; a signal with a discernable pattern carries more. So naturally, the consonant notes carry more information than dissonant ones. They then used this to calculate the information content of the pulse trains generated by consonant and dissonant tones.
That makes the model experimentally testable. Neurophysiologists can study living neurons to see if they find the same information content in pulse trains.
“To me the beauty of their work is that they have an analytical technique to calculate the intervals between the firing times, which is a highly non-trivial problem,” says André Longtin of the University of Ottawa in Canada. “It remains to be seen whether this work, and in particular the info measure, can be picked up by neurophysiologists.”
Journal reference: Physical Review Letters, DOI 10.1103/PhysRevLett.107.108103