Why do people believe in God, ghosts, goblins, spirits, the afterlife and conspiracy theories? Two common threads running through these belief systems are what skeptic Michael Shermer in his insightful book “The Believing Brain” calls “patternicity” and “agenticity”. As the names indicate, patternicity refers to seeing meaningful patterns in meaningless noise. Agenticity refers to seeing mysterious but palpable causal ‘agents’, puppet masters who pull the strings and bring about unexplained phenomena. God is probably the perfect example of an agent.

Patternicity and agenticity can both be seen as primitive evolutionary features of our brain that have been molded into instinctive behaviors. They were important in a paleolithic environment where decisions often had to be made quickly and based on instinct. In a simple example cited by Shermer, consider an early hominid sauntering along somewhere in the African Savannah. He hears a rustle in the grass. Is it a predator or is it just the wind? If he assumes the former and it turns out to be the latter, no harm is done. But if he assumes it’s just the wind and lets down his guard and it turns out to be a predator, that’s it; he’s lunch and just got weeded out of the gene pool. The first mistake is what’s called a ‘Type 1’ or false-positive error; the second one is a ‘Type 2’ or a false-negative error. Humans seem more prone to committing false positive errors because the cost of (literally) living with those errors is often less than the cost of (literally) dying from the false negatives. Agenticity is in some sense subsumed by patternicity; in the case of the hominid, he might end up ascribing the noise in the grass to a predator (an ‘agent’) even if none exists. The important thing to realize is that we are largely the descendants of humans who made false-positive errors; natural selection ensured this perpetuation.

Before we move on it’s worth noting that assuring yourself a place in the genetic pool by committing a false positive error is not as failsafe as it sounds. Sometimes people can actually cause harm by erring on the side of caution; this is the kind of behavior that is enshrined in the Law of Unintended Consequences. For instance after 9/11, about a thousand people died because they thought it safer to drive across the country rather than fly. 9/11 did almost nothing to tarnish the safety record of flying, but those who feared airplane terrorism (the ‘pattern’) reacted with their gut and ended up doing their competitors’ gene pools a favor.

Yet for all this criticism of pattern detection, it goes without saying that patternicity and agenticity have been immensely useful in human development. In fact the hallmark of science is pattern detection in noise. Patternicity is also key for things like solving crimes and predicting where the economy is going. However scientists, detectives and economists are all well aware of how many times the pattern detection machine in their heads misfires or backfires. When it comes to non-scientific predictions the machine’s even worse. The ugly side of patternicity and agenticity is revealed in people’s belief in conspiracy theories. Those who think there was a giant conspiracy between the CIA, the FBI, the Mob, Castro and the executive branch of the government are confronted with the same facts that others are. Yet they connect the dots differently and elevate certain individuals and groups (‘agents’) to great significance. Patternicity connects the dots, agenticity sows belief. The tendency to connect dots and put certain agents on a pedestal is seen everywhere, from believing that vaccines cause autism to being convinced that climate change is a giant hoax orchestrated by thousands of scientists around the world.

Notwithstanding these all too common pathologies of the pattern detection machine, it’s satisfying to find a common, elegant evolutionary mechanism in our primitive brain that would be consistent with generally favoring false positives over false negatives. What I find interesting is that this behavior even seems to exist at the level of molecules.

I realized this when I was recently studying some proteins whose exclusive job is to metabolize and detoxify foreign molecules. These proteins can be seen as the gatekeepers of the cell. Throughout evolution we have been bathed in a sea of useful, useless and toxic chemicals. Our bodies need some mechanism for distinguishing the good molecules from the bad. To enable this living organisms have evolved several proteins which bind to these molecules and in most cases change their structure or simply eject them from the cell. The most important of these are called cytochrome P450 and P-Glycoprotein. Cytochrome P450 metabolizes drugs, nutrients, hormones, poisons; basically any molecular entities that living organisms encounter in a changing environment. P-Glycoprotein is a kind of vacuum cleaner that first sucks up molecules and then throws them out.

Cytochrome P450 and P-Glycoprotein are crucial for detoxifying our body and letting only ‘good’ molecules pass through. But like our early hominid they are imperfect and seem to often err on the side of caution, making false positive errors. This problem is routinely confronted by drug developers who are consternated to find molecules that may perform perfectly in killing cancer cells in test tubes, but that are immediately modified or ejected out of the cells by cytochrome P450 and P-Glycoprotein when administered to test subjects like rats or human beings. Finding a putative drug compound that will not be modified or rejected by cytochrome P450, P-Glycoprotein or any number of other gatekeeper proteins is one of the biggest challenges in early stage drug development.

And yet if we think about it, both cytochrome P450 and humans are doing the bidding of patternicity and agenticity. For a human as well as for a protein, generally speaking it’s much safer to make a false positive error than a false negative one. In case of cytochrome P450, it might be ok if it discards a useful nutrient or two along with dozens of toxic chemicals. But if it lets even two or three deadly compounds from, say, snail toxin or snake venom in, those might be the last compounds it encounters during the painfully short lifetime of its human owner. Now of course, at the beginning when cytochrome P450 was in the process of evolving it probably existed in many more forms than what it does today. Some of these forms committed false positive mistakes and others committed false negatives. But it’s clear from the ongoing discussion that just like the human hearing the rustle in the grass and mistaking it for the wind, proteins which committed false negative errors were declared persona non grata by natural selection and weeded out. Those making false positive mistakes lived another day to see another molecule ejected.

To me the observation of patternicity and agenticity at the level of human brains as well as individual proteins is a testament to the enormous power and elegance of evolution in molding living organisms across an incredible hierarchy of molecules, cells, organs, individuals and societies through common mechanisms. It occurs to me that if evolution had to pick favorite lines from poems, one of them would probably be “Two roads diverged on the way to life, and I took the one which made me commit a false positive error”.