Epistemic noise

From Noise by Fischer Black. He is speaking about financial markets but I suspect the observation is pertinent to any complex system. This was written in 1986, towards the end of a ten or fifteen year cycle of deregulating financial markets. OECD countries loosened exchange controls, capital market controls, deregulated banks and savings & loans, deregulated real estate markets, deregulated brokerage pricing, revoked Glass-Steagall, etc. As the financial system became more responsive to the market, it served to reduce strategic risk but increased tactical risks. And it certainly increased noise in the system. Without pricing regulation, it became more and more difficult to intelligently price in risk at the very time things were becoming riskier. Black was writing this piece at the height of the financial deregulatory process and therefore was very attuned to noise in the system.

From the abstract.

The effects of noise on the world, and on our views of the world, are profound. Noise in the sense of a large number of small events is often a causal factor much more powerful than a small number of large events can be. Noise makes trading in financial markets possible, and thus allows us to observe prices for financial assets. Noise causes markets to be somewhat inefficient, but often prevents us from taking advantage of inefficiencies. Noise in the form of uncertainty about future tastes and technology by sector causes business cycles, and makes them highly resistant to improvement through government intervention. Noise in the form of expectations that need not follow rational rules causes inflation to be what it is, at least in the absence of a gold standard or fixed exchange rates. Noise in the form of uncertainty about what relative prices would be with other exchange rates makes us think incorrectly that changes in exchange rates or inflation rates cause changes in trade or investment flows or economic activity. Most generally, noise makes it very difficult to test either practical or academic theories about the way that financial or economic markets work. We are forced to act largely in the dark.

I think that Fischer's insight about noise in the financial markets can be applied to any system which becomes overwhelmed by noise. For example, the systems of education, communication and sifting fact from opinion. Richard Feynman was talking about some of these issues way back in 1974 when he gave a Caltech commencement speech on Cargo Cult Science. Since then, the situation was first exacerbated by a loss of cultural literacy and then by postmodernism which cultivated a habit of deliberate unknowing (group over individual, no facts only opinions, no hierarchy of truth, etc.) and overall a falling away from the culture of inquiry and progress which was the Age of Enlightenment.

From 2000 to 2010 there occurred two technology events which further exacerbated the dive into undifferentiated noise. The emergence of the internet (global connectedness and access to information) and the emergence of the smart phone, a device which enables constant connection to the internet.

With the internet and the emergence of the smart phone, we shifted from communication which was dominated by analog and symmetrical patterns (conversation and person-to-person knowledge sharing) to digital and asymmetrical patterns (personal seeking of information without filter or quality control). We can now access volumes of information inconceivable before, we can do so at times convenient to ourselves, and simultaneously, we have diverted time from dialogue and conversation to straightforward absorption. We read or we listen. We don't discuss as much. Or at least it seems to me.

And there is a real question as to what passes for reading.

Click to enlarge.

What are the implications of these changing patterns of communication where we spend less time in conversation and low intensity discussion and more time in strident declarations (from the media intent on gaining eyeballs through loud and extraordinary claims?)

The hypothesis is that with the advent of the internet and smart phones, people are 1) spending a greater portion of their time simply acquiring information and less time analyzing, or critically thinking about the information, 2) people are spending less time in structured and civil conversation and thereby are losing the habit and capacity for civil discourse, 3) people are experiencing an increasing noise to signal ratio because noise is cheap to produce and signals are expensive, 4) people are experiencing more cognitive false positives whereby they think they recognize a pattern in the information but it is really a coincidental pattern in the random noise, 5) people are more likely to experience noise cascades where coincidental correlations in the noise are widely believed to reflect some truth when in fact there is no such pattern, 6) with the loss of conversation, people lose a mechanism for normalizing beliefs and for calibrating facts from opinions, and 7) people become more exposed to noise amplification whereby some interest group is able to garner magnified influence owing to noise manipulation.

A consequence, if the foregoing is true, is that reality becomes increasingly difficult to discern, surviving discourse becomes more inconsequential and faddish, and people become more polarized in their opinions because they are less able to sort the wheat (signal) from the chaff (noise).

An elaboration on these suppositions.

We are spending more time gathering information and less time testing information - Pew documents a lot of this. More time on devices and less in conversation. At the same time, we are also seeing a low rate of knowledge acquisition in university, a period of time when one would expect knowledge acquisition to be at the highest.

We lose practice in the art of conversation and reading other people - A conversation or dialogue has historically been the fulcrum around education and knowledge generation (Socratic dialogue, the American Lyceum Movement, storytelling, etc.) We hone our skills in logic and evidence, in language and rhetoric, in reading body language and facial expression. We learn that human interaction is a complex dance requiring a lot of tolerance, play and ambiguity. When we no longer engage in such two-way conversations, we lose those skills.

We lose practice critically considering incoming data since it is now limitless. - The more information, the more critical it becomes to exercise filtering techniques. However, filtering and judging knowledge credibility are cognitively taxing and it is not uncommon for people to simply lapse into acceptance of all information regardless of quality.

We lose the benefit of a better noise to signal ratio. - Noise is cheap and signal expensive. Anytime there is an increase in amount, noise will increase at a faster rate than signal. Essentially this is a communication version of Gresham's law.

In economics, Gresham's law is a monetary principle stating that "bad money drives out good". For example, if there are two forms of commodity money in circulation, which are accepted by law as having similar face value, the more valuable commodity will disappear from circulation

The epistemic equivalent would be:

Gresham's epistemic law is an epistemic principle stating that "bad information drives out good". For example, if there are two forms of information in circulation, which appear to have similar face value, the more valuable information will disappear from circulation

We experience more false positives in pattern recognition owing to noise. - The more noise there is, the more opportunities there are to impute a (false) pattern which does not actually exist.

We experience greater exposure to noise cascades. - The fear of electromagnetic radiation, originally centered on the health effects of powerlines and later focused on the possible cancer causing potential of cell phones, serves as a good example of such a cascade.

We lose the normalization process when rare data is cogitated internally and then calibrated with others. - If we spend less time thinking critically, we become more subject to cognitive pollution and if we are at the same time also spending less time discussing things with others, when we could at least leverage a mutual critical capacity, then we become more susceptible to unsubstantiated opinion. This is further exacerbated, as we are seeing currently, if the idea gains traction that exposure to unfamiliar ideas is a form of assault.

We experience greater exposure to noise amplification with potentially catastrophic consequences. - The University of Missouri is an example, among many, of this problem. Students from the bottom ten percent of academic performance made social justice demands that are inimical with free speech, fair assessment of performance, with the law, and with the constitution. In doing so they forced the resignation of administrators, assaulted fellow students, and closed down any discussion about issues. As a consequence of the negative publicity attendant to the university's capitulation to baseless demands, the other 95% of students suffer as enrollment falls, budget deficits open up, services are curtailed, and tuition is increased.

Noise, AKA unsupported opinions, no platforming, fake news, etc. has a far greater epistemic consequence than we account for. It is not clear that there is a technological solution when sometimes the technology is the source of the no-platforming.

This would appear to be an instance where manners and cultural values regarding respect and free speech need to be the solution at the very time when all three of those norms are under attack.