Mind Matters Natural and Artificial Intelligence News and Analysis
Quitting smoking - male hand crushing cigarette
Quitting smoking - male hand crushing cigarette

Yellow Fingers Do Not Cause Lung Cancer

Neurosurgeon Michael Egnor and computer engineer Bob Marks look at the ways Big Data can mislead us into mistaking incidental events for causes

It’s easy to explain what “information” is if we don’t think much about it. But what if we ask a student, what does your term paper weigh? How much energy does it consume? More or less matter and energy than, say, lightning striking a tree?

Of course, the student will protest, “But that’s not the point! It’s my term paper.” Exactly. So information is very different from matter and energy. It means something.

Realizing that information is different from matter and energy can help us understand issues like the difference between the causes of a problem (causation) and circumstances that may be associated with the problem but do not cause it (correlation).

In last week’s podcast, “Robert J. Marks on information and AI, Part 1.” neurosurgeon Michael Egnor and computer engineer Robert J. Marks talked about that:

Their discussion of causes vs. related events (correlation) starts at 17:38. The Show Notes and transcript follow.

Here is a partial transcript:

Robert J. Marks (pictured): There’s a great website called Spurious Correlations where, in big data, you can get correlations between totally unrelated causes. They have a graph where, for example, people who died strangled in their bedsheets are compared to the cheese consumption of Wisconsin. And the curves lined right up.

They were totally correlated, but clearly there was no causation there.

Note: The top two items currently on the site are “US spending on science, space, and technology correlates with Suicides by hanging, strangulation and suffocation” and “Number people who drowned by falling into a swimming-pool correlates with Number of films Nicolas Cage appeared in.”

One hopes no one takes these correlations seriously. They are meant simply to teach that correlations need not involve any true relationship at all. They are accidental outcomes of big numbers.

Michael Egnor: All kinds of errors are made in medical research because it’s assumed that correlates imply causes, which they don’t necessarily.

Robert J. Marks: And this is interesting because most of the journals that accept papers based on statistics are only into the correlations. They require an r value of such and such, meaning that the data corresponds to a high degree of correlation. But the problem is, of course, we do have these spurious correlations. And this has led to the conclusion of one researcher that up to 90% of the papers published in the literature that are based on statistics are flawed.

Michael Egnor: I think it’s much higher than 90%. I think, that’s a real underestimate of the number of papers that are published.

Note: Eminent researcher John Ioannidis argued in 2010 that most medical research studies are wrong, on account of issues such as this one. It’s not a new problem; more of an under-recognized one.

Robert J. Marks: Today, coffee is good for you, coffee is bad for you. You hear all of these fleeting studies that are reported in the news as gospel and it can have a terrible effect. It can make you paranoid. And so, I try to ignore all of these, especially since Gary Smith pointed this out to me, yeah, it’s really terrible.

Note: Pomona College economics prof Gary Smith often writes about the perils of Big Data spewing out meaningless correlations that are taken seriously. Any very large sequence of pieces of information will yield apparent (but not genuine) patterns, precisely because it is random. Only systematic, intelligent investigation can determine if a pattern is a find or just a blip.

Michael Egnor: Machines can crunch numbers, and do things that don’t provide insight. But to get insight into what causes what, I think, requires a human being.

A good example is that smokers frequently have yellowing of their fingertips because they’re holding cigarettes. And they are also predisposed to get cancer. There’s no question that yellowing of your fingertips correlates with having lung cancer but it doesn’t mean that yellow fingertips give you lung cancer. Or that they cause lung cancer. You have to get the causal arrows right. Yellow fingertips and lung cancer are both caused by a common factor, which is smoking, but they don’t cause each other.

Robert J. Marks: I remember a similar story about the ice cream consumption and murder rate in New York City. The ice cream consumption rate would increase and then the murder rate would increase. They were both related to the rise in temperature.

Michael Egnor (pictured): When I was a resident in neurosurgery in Miami during the drug wars, we would get gunshot wounds to the head coming into the ER constantly, except these admissions would always stop when it rained. Miami has quite a bit of rain, so when we would have an hour or two of rain, the ER would just go completely quiet. Nobody would come in. And then the sun would come out, and people would shoot each other again. But people wouldn’t shoot each other during rain.

Robert J. Marks: So, good weather causes people to kill each other.

Michael Egnor: Yeah. Or that there’s something about being wet that protects you from gunshot wounds.

Or, alternatively, the Spurious Correlations website and John Ioannidis are right and we should listen to them before we buy into the Next Big (Spurious) Scare.

You may also enjoy: Information is the currency of life. But what IS it? How do we understand information in a universe that resists resolution into one single, simple system? The information in living things is directed toward purposes in a way that we don’t see in non-living things. The word for that is creativity.

Show Notes

  • 00:27 | Introducing Dr. Robert J. Marks
  • 01:14 | What is information?
  • 06:44 | Exact representations of data
  • 08:29 | A system with minimal information
  • 09:27 | Information in nature
  • 10:43 | Comparing biological information and information in non-living things
  • 11:29 | Creation of information
  • 12:50 | Will artificial intelligence ever be creative?
  • 17:38 | Correlation vs. causation

Additional Resources

Podcast Transcript Download

Mind Matters News

Breaking and noteworthy news from the exciting world of natural and artificial intelligence at MindMatters.ai.

Yellow Fingers Do Not Cause Lung Cancer