It is tempting – if that’s the right word for it – to see the brain as a big, ugly ball of spoiled wet spaghetti. But even if you don’t find it gross looking, you still probably think of it as a ball.
If you could use an analogy, instead think of brain as a river – a long, complex river, with many tributaries and a complex system of channels. Now think of that river as being folded around itself in three dimensions rather than lying flat and merely curving side to side, the way real rivers do. Like, in a sense, an immensely long spaghetti noodle that you picked up and balled up. And now think of the physical structure of that river as something more specific than merely a river – think of the physical brain as the banks of the river. Because that’s in many ways what the brain is – an unspeakably long and complex set of riverbanks.
Here’s the payoff: the water flowing through the wending, weaving banks is information. Each time a neuron fires, it metaphorically moves a drop of water just a little farther down the river.
The metaphor is used, in one way or another, by most neurscientists, because rivers have a crucial quality that separates them from lakes and ponds: the water moves from one place to another. And that’s what information does in the brain – it flows from ‘upstream’ regions to ‘downstream’ regions. Like rain falling on a mountain, perceptions fall into the brain through the various senses, and these can regarded as the farthest upstream reaches of the brain’s informational river. These perceptions are the first bits of information to enter the brain, and as they float and are pushed (and sometimes are dammed up or drained off or even eddied back upstream) the brain continues to mold them, like water picking up bits of dirt and clay and temperature changes as it flows through varied terrain. And at the far other end of the brain, like water draining into the vast and unbounded ocean, the information emerges from the brain in the form of movement.
This analogy turns out to be rather more subtle than you might think. For example, pick a spot on your favorite river – mine is where Route 49 crosses the Yuba River, just outside of Grass Valley, California. If you imagine standing there, you’ll quickly see that on one side is the upstream site and the other is the downstream site. But if you jump in and swim a mile downstream, used to be ‘downstream’ is now ‘upstream’. Which is to say, upstream and downstream are relative rather than absolute terms. Neuroscientists therefore use these terms when discussing a specific brain region to talk about where the region is receiving information from (the upstream sources) and where it is in turn sending information to (the downstream targets).
Now the analogy gets complicated because you have to imagine something about rivers that doesn’t really occur in nature very often at all, but which occurs all the time in the brain. Luckily, you just have to imagine one thing. Here goes: imagine a second river, just as long and just as complicated as the first, plopped down on top of the first river, but flowing in the opposite direction.
What is this river doing? In a word: feedback. It is carrying – or feeding – information from downstream back to some upstream points. This allows the upstream points to ‘know’ what effect they’ve had downstream.
To use an example, imagine a chemical company upstream is dumping toxic waste into the river, but that it takes the fish swimming in it 10 minutes to die. A fisherman 10 minutes downstream would be the first person to see all the dead fish floating in the river. If he fished them out of the river and then placed them into the opposite-flow river, then 10 minutes later the chemical company would see a bunch of dead fish floating by. This would be a form of feedback – they’d be learning what happened as a results of their chemical dumping. And could then choose whether to do anything about it.
This kind of recurrent feedback, in which downstream regions feed back information to upstream regions, happens constantly in the brain, and allows for self-tuning.
It’s important to remember, however, that since downstream and upstream are relative terms – relative to where the information is coming from – when the fisherman sends the fish to the chemical company, neuroscientists will say that the chemical plant is downstream from him! That’s because the information is traveling ‘down’ from the fisherman to the company, even though, relative to the original perceptual mountain on which the information first rained, the information is moving upstream.
A particularly important way in which this kind of feedback occurs is through something called efference copy, otherwise known as corollary discharge, which we’ll get to another time.
Bottom-Up versus Top-Down
The whole downstream/upstream business is confusing enough. To make matters worse, neuroscientists often use a different analogy – that of a bureaucracy – to describe the very same regions that they call upstream and downstream. More on this in our next post.