Echo Chambers and Variances.

The term “echo chamber” is an unduly harsh characterization of what happens when a “conventional wisdom” emerges.  Conventional wisdom is, invariably, an accurate description of the reality within a set of limits:  in the classic beauty contest game, for example, the “usual” standards of beauty are known sufficiently widely that people who want to strategically anticipate what others think, regardless of what they really think, can guess them.  The problem is that, usually, what happens is that they overthink what the “regular people” think and focus on the “conventional wisdom” too narrowly.  The consequence, as seen in the NPR story linked above, is that the first-order strategic thinkers (that is, people who are trying to anticipate what others think) wind up with an unduly small variance.


The problem, then, is not that the “conventional wisdom” is systematically wrong.  But reliance on it systematically underestimates the extent to which it is wrong.  The wisdom gained by statistics is not that you can necessarily predict the right answers, but that you can assess how wrong your answers are–because EVERY answer will be wrong, at least in social sciences.  The systematic reduction in variance from a widespread adherence to the conventional wisdom, then, subverts this wisdom, by becoming overconfident in the power of the conventional wisdom well beyond where it might be applicable.  In a sense, this is the real problem that arises from the echo chamber.  Variance is how we learn about things, by seeing the patterns in which the “reality” differs from what we expect it to be.  Without the variance, we cannot learn so much.

In a sense, a set of signs generated from deliberate choices and behavior by people are more problematic than the exogenously generated signals:  when astronomers were able to measure the patterns in sunspots more precisely, the gap between sunspot patterns and market patterns was revealed, and the equilibrium broke down quickly.  Sunspots, after all, do not deliberately choose their patterns.  But people go to churches on Sundays, and again on Chrismases and Easters, and they have calendars to make sure that they follow these patterns.  If you can define Easters more precisely, people will change their behavior to accommodate this more precisely defined definition of Easter–in a way, they did exactly that when the Gregorian calendar was introduced during the Rennaissance.  Ironically, this is even more the case because religious holidays, after all, are a human invention, something humans have agreed to as a coordination device, and not a force of nature independent of human choice.

Plenty of “human knowledge” falls in the same realm as religious holidays:  most things are what they are because we have agreed that those are what they are.  For example, nobody “really” understands the full complexity of evolutionary biology, and only a relative handful of real experts can really explain what we do understand and what we don’t at the level of molecular biology and below.  But the conventional wisdom version of evolution and the opposition thereto bear only superficial resemblance to the real science.  It is a construct that serves to coordinate the opposing camps, a creed whose recitation identifies them as partisans of one side or the other and something that is true for no other reason that their adherents agree that it must be true.  To the degree that politics, markets, and every other facet of human interactions are products of these consensuses, simply poring through the data makes it difficult to understand what might take if and when–if ever–the patterns might break, although they do go far to explain why cargo cults persist long beyond the cargo’s disappearance.


One thought on “Echo Chambers and Variances.

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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s