Fighting the herd mentality: making use of news and statistics

We read more and remember more to achieve diverse goals. One of these goals: to be informed and avoid the herd mentality. To do that we may need to understand how to read between the lines: news and statistics. For more reading, you are welcome to check here, here, here, here, here, here, here, here, here, and here.

The basic tools of critical thinking

In various places, I address critical thinking in different ways. Here I want to focus on three critical skills: analysis, evaluation, and inference.

In this sense, analysis is the ability to label correctly various parts of argumentation, evaluation deals with validity and scope of each part of the argumentation, and inference is the ability to extend arguments to new cases and new areas.

Take for example reading. If we cannot analyze the parts of the text, we are not able to map them reasonably. Evaluation enables us to select very few critical words that define correctly long texts. Inference helps us build visual associations between diverse worlds of content.

Critical thinking when dealing with news and statistics

When dealing with news headlines and separate pieces of data, we need initially to map them with respect to our understanding of the issue. Effectively we grow our mindmaps by adding new branches or other hierarchical data.

When we evaluate the content, we need to pay attention to the cognitive biases of ourselves and the authors. We are very likely to accept information that resonates with the way we think, or with other information we find. This generates an echo chamber: many different resources sending similar messages. In this case, we need to focus on innovation: which new information is added by each source.

Statistics can be exceptionally challenging for our inference skills. We will tend to apply the statistics we discovered in very different situations. Each time we find divergence between the inferred statistics and the actual measurements we will search for the route cause. Typically this will imply that very different characteristics are measured: possibly there is a time shift or selection of the group that was measured.

For example, if we check people with symptoms of a disease, the results will be very different from sampling general population. If we make sampling in a certain locality, the results will be different from sampling the entire nation.

Be rational

One of my friends gave me strange advice: “When you try to evaluate the logical arguments, distance yourself from the situation. Try to visualize yourself as a robot, or as a historian reading things that happened thousands of years ago. This way you might avoid irrational feeling clouding your judgment”.

This is especially true in bubbles and in panic situations. Bubbles are often generated when people believe that bad things will not happen. Panic is usually caused when people do not see a viable solution in the visible future. Both situations are very risky and should be avoided.

We are predisposed to search for safety in numbers. This strategy may help find certain resources and avoid some dangers, but when it does not apply the result can be disastrous.

It makes sense to follow the trends in times of relative safety and boredom. Contrarians often win in murky waters. Simply to have a choice we should be able to think independently and this is difficult.

Look for the fakes

About 10% of the population will tend to criminal activities no matter what.  Possibly another 10% will always be just and true. Probably 80% can go either way. In times of crisis, the masses will tend to follow the more powerful and charismatic figures. Charismatic people should come with a warranty, as often charisma is a sign of narcissism, machiavellian lack of ethics or psychopathy.

If you are in states of extreme greed and fear, you are easy prey for cons selling fakes. When our daily routine is disrupted, we do not know what to believe. There will always be phishing attacks, sales of miracle products, and fake gurus. The situation gets worse when these products are recommended by our best friends and mentors whom we trust.

The search for the fake should be active. If something looks too good to be true, it usually is. Good performers hide their tricks from our eyes in plain sight. Read explanations of magic tricks to understand the mechanics used. A good magician often invents tricks that other magicians cannot understand without further explanation.

Miracle explanations are usually cons

There is a historical anecdote of Harry Houdini trying to communicate with his dead mother. None of the spiritualists of his time passed his scrutiny:  he was sufficiently wise to discover each trick. Other great thinkers of the time like Arthur Conan Doyle and Oliver Lodge did not uncover the same tricks and went to seances with their dead sons.

No matter how smart we are, we will discover things that defy our intelligence. We probably do not have enough knowledge to discover all the manipulations of reality. And our friends will have a desperate wish to believe: in conspiracies,  secret deals, and hidden knowledge. No matter how smart they are, they will be wrong from time to time.

Simple, stupid and boring explanations are usually more trustworthy than great ideas.

Statistics is all about sampling

Being good in math does not make you an expert in statistics. You may be a great engineer with PhD in random processes, and fail to notice the selection bias of certain research. The ways to manipulate statistics are very simple:

• Modify the conditions for the sampling, e.g. add test to apply for sampling.
• Infer from a small sample to a large dataset.
• Inverse the conditional probabilities (most terrorists are Muslims, but most Muslims are not terrorists).
• Add time bias (during COVID-19, death statistics lagged two weeks after official infection rates, which lagged two weeks after the actual infection rates).
• Unite options that cannot be united (during the Brexit referendum, the different ways to conduct the exit were united into one option).
• Researcher bias. Sometimes the researcher will mingle with the data to remove the outliers. This is often an honest approach rather than a smart way to manipulate the outcome.

You do not need math skills to understand bad statistics. Look for who was tested and who conducted the tests.

FOMO in times of crisis

When things are uncertain we always look for new inputs, more information, and further clarifications. Above some level, this fear of missing out starts to play tricks.

In one of the cabinet meetings, Benjamin Netanyahu presented his ministers with a video of Iranians dumping bodies during the COVID-19 crisis. The video was fake. It did not arrive from qualified sources and was not verified by the intelligence community. A very smart and sophisticated man took a very fake video and used it because the video corresponded to his view of events and his need for immediate actions.

The most recent news is the least likely to be tested and verified. By taking the news feeds as data sources, we not only lose our valuable time but also make very bad decisions.

Information is needed in times of uncertainty, and we are right to look for it. Simply, it is best to take information from verified sources after there was enough time to verify it.

Herd immunity

We could hope that the more people are exposed to a dangerous disease or bad information, the safer we are. The real situation is much more complex.

During the times of pandemics, the superspreaders carry the disease without exhibiting any symptoms. People who believe in crazy things are often very reasonable and successful.

Moreover, just as a disease can mutate, so can disinformation. A person who recovered from disease can get sick again with a mutation of the disease. If someone recovered from bogus beliefs, he is quite likely to catch some other bogus belief.

Digital detox

Popular subjects get our attention. If we want better chances to have independent ideas, we need some space to think. That means that occasionally we should turn out the new sources and simply analyze, evaluate and inference the information already available.

Then we can also detach ourselves from our current social support group and search diverse people for motivation and emotional support. Due to their diversity, these people will likely to challenge our ideas and suggest alternatives.