Freud has fallen out of favor, with his analogies of the workings of the mind to a steam engine (drives, pressures, releases, displacements), the dominant technology of his day–as the computer is to ours. However, the following paper [ Proc. Natl. Acad. Sci. vol. 113 pp. E616 – E6125 ’16 ] shows that we have an unconscious mind, and that it is mathematically sophisticated (although I don’t think the authors made this point).
[ Proc. Natl. Acad. Sci. vol. 113 pp. E616 – E625 ’16 ] The work used magnetoencephalography (MEG) https://en.wikipedia.org/wiki/Magnetoencephalography, to record brain activity in response to a series of tone pips. MEG is conceptually quite similar to the electrocardiogram (EKG) or the electroencephalogram (EEG), both measuring voltage differences between two electrodes over time. Well, a voltage difference causes an electric current to flow through a conductor, and the nice wet brain is nothing if not that. Anyone who has studied how an electric motor works, knows that an electric current produces a magnetic field, which is what the MEG measures. The great advantage of MEG is that it is temporally precise, and changes can be measured in milliSeconds.
So what did they do? They presented tone pips to an unspecified number of subjects. The relation of one pip to another could either be completely random (RAND) or part of a repeating pattern — say pip pip pip silence silence pip pip pip silence silence (PAT). In one series of experiments, subjects were asked to press a button as soon as the pip sequence went from random to patterned (RAND –> PAT), all this while the MEG was being recorded. In another, they were asked to do this for PAT –> RAND. The subjects were as good as something called the the mathematical ideal observer of the variable order Markov model. It only took one or two random pips after a patterned sequence to notice it and press the button. They could also pick up that a pattern was formed midway through the second repetition of a pattern.
The MEG showed abrupt changes at either transition (RAND –> PAT or PAT –> RAND). The work didn’t stop with just sequences of just one tone. They could use an ‘alphabet of tones’. The subjects could pick up when the number of tones in the alphabet changed, again with MEG values to match. So they had an independent signal from the MEG show that the brain picked up the transition without requiring any cooperation from the subjects. All very nice, but anyone who likes music can do this.
Then the subjects were then asked to perform the n-back task, in which the subject is presented with a sequence of stimuli; and the task consists of indicating when the current stimulus matches the one from n steps earlier in the sequence. Tricky isn’t it? Certainly, something that requires concentration. The load factor n can be adjusted to make the task more or less difficult. You’ve got to hold the sequence just presented in your head so the n-back task is a test of working memory.
Drum roll —
If the tone pips were presented while the subjects were doing the n-back task, the MEG still picked up RAND –> PAT and PAT –> RAND transitions, something the subjects weren’t consciously trying to do.
We know the brain does all sorts of things unconsciously — e.g. breathing. But they are pretty simple. The tests here are conceptually subtle. Your unconscious brain picks up statistical regularities and irregularities without your consciously trying. Who knows what else it does — maybe Freud was right.
Why should this be useful? Well, you’d want to know if a predator is sneaking up on you. The same work should be done with animals performing a task they’ve been trained to do.