On Thinking Like a Human

Intuition Algorithm download in progress

The computer processing analogy that we use for describing the way the mind works matches only the more rational side of human thought, and the senses most associated with them (vision and sound). Or at least that’s the way it used to be, before the AlphaGo defeat.

Other ways of thinking are more intuitive, and more in line with the primitive senses (smell). These types of thinking are becoming more important, because they are more analogous to what it means to be human today, as compared to a computer, that is. It turns out that humans are not rational enough to integrate with computers, i.e. robots; the problem with self-driving cars is that they’re too rational; they follow all the rules, all the time, and humans don’t.

What we bring to the table, as humans rather than algorithms, is the ability to make intuitive decisions based on accumulated experience. I’d like to point out that this accumulation is called memory, and although it is “stored” in the brain, it requires a body to get there in the first place. This is one of the main reasons why olfactory perception can be such fertile ground for research.

Basically, humans would rather guess than go through a bunch of bad options. Sometimes we’re wrong, but when we’re right, we just saved a whole lot of processing power.

In this study, a huge group of game players was tasked with solving a very complex problem. Instead of running every possible solution, which would take forever, the players search intuitively. And the study showed that this intuition-searching happened to be similar for each of the individual game players, hinting at a possible “intuition algorithm” for the future of computing. If investigating the artificial unconscious looks like a good bet, then we might see the half-primitive half-cognitive “language of smell” gain some attention soon.

Notes:

Are humans the new supercomputer? Team blurred the boundaries between man and machine

phys.org, May 016