Tag: representativeness heuristic

Fort Hood: That No Such Attack Ever Occurs Again

Colleagues and correspondents have kindly shared their understandable discomfort with my conclusion in recent posts that the Fort Hood shooting was nearly impossible to discover in advance, and thus prevent.

The one ray of hope I can offer is that the shooting itself makes such things more foreseeable, putting the military community and investigators on notice prospectively that this kind of thing can happen. No formal policy change can do more than the Fort Hood shooting itself to ferret out inchoate incidents like it in the future. Belief that the Fort Hood shooting was easily preventable, though, is 20/20 hindsight.

I first read How We Know What Just Isn’t So: The Fallibility of Human Reason in Everyday Life to get a handle on how it became so plausible after the September 11, 2001 attacks that terrorists might next use chemical, biological, and nuclear weapons. Recall that their weapons of choice for the World Trade Center and Pentagon attacks were box cutters. How did we proceed to the assumption that nuclear terrorism was next?

One explanation is the “representativeness heuristic,” a mental shortcut people use to organize the world around them. “According to this overarching belief, effects should resemble their causes, instances should resemble the categories of which they are members, and, more generally, like belongs with like.” (page 133)

Big causes have big effects, so big effects come from big causes. … Right?

The 9/11 terrorists knocked down the World Trade Center and killed 3,000 people. Driven to match the huge effects of the those attacks to a sufficient cause, our common sense imported skills, knowledge, weapons, and organizational capability that terrorists do not in fact have. (Ongoing pressure worldwide will ensure that remains true.)

As to the 9/11 attacks, the representativeness heuristic lead us astray. I believe a similar mental error is at play in many people’s interpretation of the Fort Hood incident.

Though it’s not true, many maternity room nurses believe that more babies are born during a full moon than at other times. This is because of confirmation bias: They notice babies born during full moons and accumulate proof of the full-moon theory—but they fail to notice babies born at other times.

How We Know What Just Isn’t So has a chapter called “Too Much from Too Little: The Misrepresentation of Incomplete and Unrepresentative Data” that discusses not only the excessive impact of confirmatory information, but also the problem of hidden or absent data. We make many judgments in life without considering all the relevant data.

An extreme instance of this is Fort Hood, about which political leaders and millions of Americans are taking a few data points—one or two things occurring—and concluding from them that all instances of these things result in a shooting or other violence like we saw at Fort Hood. But, as I said with regard to Nidal Hasan’s contacts with a jihadi in Yemen, the relevant data includes thousands of times when such things happen. Because they were offshore communications with a jihadi, investigators appropriately examined the messages and found them lacking signs of intended violence.

The other major indictment is that Hasan’s Islamist rantings should have been a dead giveaway of violence to come. Political correctness drove colleagues to turn a blind eye to Hasan, ”permitting” the Fort Hood shooting to happen, this argument maintains.

There probably was some “political correctness” involved. I can think of no community more likely to withhold judgment of others than psychologists and psychiatrists, who are privy to the strange and dangerous thoughts of their patients day after day after day.

Note again the full range of relevant evidence, though: Thousands of times daily across the country, mental health professionals and social workers hear people’s violent thoughts—not just political rantings—which only rarely materialize into violence. In the military, it’s harder to guess at a number, but certainly thousands of times per year, service members discuss violence against other service members and political opinions that are odd or controversial, including Islamist political views. Very rarely—tragically when it does—this results in actual harm to men and women in uniform.

Nidal Hasan may have been fit for expulsion from the military. He may have been kept in by some form of political correctness or opportunistic bureaucratic burden-shifting once it was clear he was leaving Walter Reed for Fort Hood.

But only operation of the post hoc ergo propter hoc fallacy allows the conclusion that his expulsion from the military would have averted the tragedy. Because it followed in time, the shooting appears to be a result of his continued military service or his looming deployment to Afghanistan. But it is not so obvious that his discharge from the service would have caused him to go limp, take a job at a convenience store, and live a happy life.

Had he been pushed out of the military, it’s quite plausible that his resentments would have grown, his contacts with jihadis would have increased, his planning would have been more strategic, and so on. It is simple assumption that expelling Hasan from the military would have averted so many deaths and collective national pain, just like it is simple assumption that it wouldn’t have.

As I discussed in a recent podcast, information always points to what happened next when you look at it after the fact. Data does not point so clearly to any conclusion when you observe it in real time along with all the other then-relevant data.

The Fort Hood shooting was a tragic and regrettable incident, but correctable security failure is not easily shown. The idea that the shooting was predictable is fueled by a small array of common perception problems and errors in logic. These errors have now inspired a hearing in the Senate Homeland Security and Governmental Affairs Committee later this week. The committee will try to find security lapses and seek after conditions in which ”no such attack ever occurs again.”

Politicians can promise the public that every tragedy can be averted, but soldiers know better than most that tragedy and loss do happen. At the memorial service for the Fort Hood victims, Lt. General Cone captured that reality, and the spirit in which we must accept it, saying to victim’s families, ”The Fort Hood community shares your sorrow as we move forward together in a spirit of resiliency.”