The Neural Signal and the Neural Noise

The quantitatively inclined will no doubt recognize my reference to the recent book by Nate Silver about the potential and perils of prediction. While not exactly a reference for high level predictive techniques in statistics, the book was a good introduction to the general reader from a bright guy who is best known for revealing the uselessness of political pundits during recent election cycles.

And accurate prediction is at the heart of the scientific method; it’s what sets that method apart from other ways of knowing about the world. From the movement of the stars to the constituents of atoms, the true test of any scientific hypothesis is not the elegance of its theory (though that is typically held in high regard as well) but its ability to make concrete (typically quantitative) and accurate predictions about events that have either not been observed or not yet happened at all.

But to paraphrase either Niels Bohr or Yogi Berra (or someone completely different), ‘prediction is difficult, especially about the future.’ No less so in neuroscience, with its famously squishy subject matter. Whether you stick an electrode into a neuron and measure its membrane potential or image the combined activity of billions of neurons (and glia, by the way) with an fMRI scanner, there is a lot of variability in the response that seems to persist no matter how meticulously you control the inputs to the system. The typical approach to deal with this problem is to do your experiments over and over again with the expectation that the “noise” in the system (whatever its source) will eventually average out. So, you present a stimulus to a single cell or a network or a whole brain, measure the result, and maybe on that experimental trial the response is a little stronger. You repeat the stimulus. On the next trial, despite keeping everything as precisely identical as you can, the response is a little weaker. Rinse and repeat.

After a while you’ll have enough examples of the response that you can average all these together and expect the ups and downs not associated with your stimulus to balance each other (overall). It’s exactly the same principle as taking the average of all the scores on a given test for all of the students in a class. You expect that the average will tell you something about the performance of the class as a whole independent of the individual background and daily drama of the particular students within the class.

This leads to one of the most important issues with separating the signal from the noise. The difference between the two is mostly dependent on what information you want to extract. It’s like being at a party and trying to watch something on TV. For you, all that chit-chat is noise, a distraction from what you are interested in, while for someone else at the party that damn TV is interfering with her efforts to follow a conversation. Given a set of data about student grades, a teacher may be interested in the variability that relates to teaching methods while a demographer might be interested in differences associated with socio-economic status and a policy-maker might be concerned with how differences in funding in different schools are reflected in achievement (Needless to say, any of these people would likely have at least some interest in the other sources of variability as well).

Still, there are some examples of noise that are not just “shit that doesn’t interest me.” Some of it is “shit I just can’t get my head around.” Imagine a pair of dice, for example. At the macro, everyday, craps table level, they are pretty much unpredictable (random), meaning that all the variability in each throw is unexplained (really no signal there at all, unless you believe you have a “system”). Still you can imagine that if you had enough information about the mass, precise shape, and molecular composition of the dice (and table), and enough control over the throw, that at least in principle you could predict the outcome.

Nonetheless, at the micro (or rather nano, or ato) level, sometimes it’s not even possible in principle to make fully accurate predictions. Quantum theory argues that the very small bits that make up our universe don’t behave in that nice Newtonian billiard ball regime we are so used to. The counter-intuitiveness of that fundamental, intrinsic, elephants-all-the-way-down, randomness famously led Einstein to protest that “God doesn’t play dice with the world.” In other words, he thought the indeterminacy of quantum physics reflected “shit I just can’t get my head around” rather than true randomness.

There is one other source of unpredictability: chaos. Chaotic behavior is a feature of some systems that, despite being essentially deterministic, are fundamentally unpredictable, except over very short time horizons. Without going too far into the details, the important point is that the unpredictability of chaotic systems comes not from intrinsic randomness, but from the fact that they can produce wildly erratic behavior from the most infinitesimal differences in starting points.

Coming back to neuroscience, it turns out that the sources of “noise” in the nervous system can be quite controversial (and with important consequences for computational theories). As I said above, variability between trials using the same stimulus, between different neurons, between different brains, subjects, or days of the week are all vexingly real in experimental neuroscience. Nonetheless, in many experiments it remains maddeningly unclear whether the variability comes from intrinsic randomness percolating up from the nano-scale fluctuations of individual molecules, from the vast number of unmeasured and uncontrolled variables in any network, or from more strictly defined chaotic dynamics. Kind of like elections. At least we don’t have worry about the variability caused by the Koch brothers.


~ by nucamb on November 22, 2014.

12 Responses to “The Neural Signal and the Neural Noise”

  1. I think at this point in Neuroscience most of the “noise” that we observe in the data are things we don’t understand or things that are probably not interesting for the micro-system we are trying to analyze.

    • I think that’s true since in most experiments we have so little control over the variables (all the spontaneous activity; thousands upon thousands of other inputs).

  2. This was nicely done. A funny thing about noise… My wife and I have been in various public places where the various languages spoken sounded like “noise”. But as she can speak several, and on occasion I can understand a few, we know it isn’t noise. Have you followed Gerstner’s work too? It struck me that, whether neurons, or glia, or sub or super threshold ripples or spikes, all that energy wouldn’t be there if it weren’t doing “something”. Do you think it mightn’t be noise at all? Nor “random”? When we listen for some one specific thing, suddenly all else becomes “noise” with respect to that. Maybe it’s simply what our ordered expectation leaves out — and we create the concept of noise?

    • So little is known about neural codes in many systems that it is hard to say what sort of side bands could be encoded in the “noise.” But I do think that computational neuroscience (grounded in behavior/ethology) helps provide a framework to ask those questions.

  3. I found this very thoughtful. I do think that neuroscientists make way too much of randomness, often positing it out loud as a generative mechanism but with a telltale affect*.

    “Well, there’s something unexplained here” has become “We modeled the unpredictable part of the input to our system with a stochastic process”, which in certain circles has turned into “The system is itself stochastic.” The differences between these things are huge, as you’ve detailed.

    I honestly don’t know where the third argument comes from, or how it became solidified as an entire camp in neuroscience. My guess is that someone made a rash decision after seeing a few patch clamp traces: “Well, individual ion channels seem stochastic, therefore neurons must be also.” But there really is a lot of trouble with assuming noise scales up. It’s something you have to prove.

    *I once saw a lecture on stochastic resonance of which I remember very little aside from the unsatisfying conclusions and the look on the speaker’s face at the end – he clearly got away with stealing all the cookies from the cookie jar.

    • I think it’s reasonable to ask what intrinsic randomness does exist in different systems, and whether it is computationally relevant (stochastic resonance) or computationally irrelevant to most tasks (for example, if the random inputs are invisible at the soma, or average out overall). But stochasticity can also come from unexpected places. In my PhD work I did simulations of the respiratory rhythm generator that showed that you could get a lot of trial by trial variability out of a sparsely connected network with little membrane noise.

      • I suppose my question is whether a source of unpredictability can really be claimed to be internal to the system, or whether it should be put aside as an external factor.

        If you can get a decent amount of neural-level variability from a small amount of membrane noise, does analysis of this sort of model tell you whether noise jostled the system out of an unstable equilibrium or some other “small” manifold in the space of all possible parameterizations/initial conditions of noiseless models? In other words, do trajectories like those you determined using noise exist in a related noiseless system with, say, a different choice of parameters?

      • Excellent question, and I don’t see any reason to prejudge the answer for any given system. Could have different answers in different models.

  4. A neuron is actually quite a noisy device. See for a summary of my Ph.D thesis, back in ’68, later observed in further detail by E. Neher,

    • Yes, I think there is some good evidence that some neurons can be noisy in some systems. We also know that in some systems neurons can repeat millisecond-scale spike timing patterns reliably across trials. So the question becomes one of evaluating the computational/functional/ethological consequences of neurons being noisy (or not).

  5. Koch Brothers? Really? Ever compared their contributions to those of labor unions?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: