Bias And How to Deal With It

The coverage of the Japanese reactor situation reminds me of the coverage of many other technical issues when they overlap with serious breaking news stories. I wrote a little on this subject a few years ago, talking about the Merck/Vioxx business, but I wanted to expand on it.

I’m not going to rant on about the popular press not understanding this or that scientific or technical issue. There are more systemic problems with the way that news is reported, and in the way that we take it in. I’m not sure of what to do about them other than to be aware of them, but that’s an important step right there.

The first of these is narrative bias. Reporters like to relay stories (and the rest of us like to hear stories) that have a progression. They have a beginning, a middle, and an end, the way our most popular novels and movies do. Something starts, something happens, something ends. Real life sometimes conforms to this template, but sometimes it doesn’t. For example, some situations don’t start, so much as they suddenly get noticed after they’ve been there all along. And some don’t end, so much as they just stop having attention paid to them.

Another narrative-bias problem is the tendency to assign participants in any event to recognizable categories: good guys and bad guys, for starters. Moving to finer distinctions, there’s Plucky Young X, Suffering Y, Salt-of-the-Earth Z, along with Untrustworthy Spokesman A, Obfuscating B, Crusading C, and the whole crowd. Mentally, we tend to assign people to such categories, especially if we don’t know them personally, and it makes it easier for reporters, too. It’s a team effort. The problem is, of course, that not everyone fits into a recognizable category, and many others overlap in ways that a simple narrative structure won’t accommodate. Most real people are capable (more or less simultaneously) of great and venal actions, of heroism and cowardice, of altuism and selfishness.

Even when events are progressing in some sort of recognizable way, they’re seldom doing that at the tempo that we’d like. This is the problem of temporal bias. They’re especially unlikely to do that at the tempo that various news organizations would like. A cable news network would like to have something new to report every fifteen or twenty minutes; a newspaper would like something every day. But events happen when they happen, which means that in the absence of anything new to report or talk about, a tremendous amount of wind is generated to make it appear as if something is actually going on.

Our sense of history reinforces this bias. We compress and even out timelines. Look, say, at the start of World War II. Yep, Hitler invades Poland. Then he invades France. Dunkirk, Rotterdam, Battle of Britain, here we go. But there was a big gap in there, the so-called “Phoney War“, where nothing much happened (at least, not compared to the way things started happening afterwards). We sort of edit that out, mentally, but it was a long period to the people living it at the time. A 24-hour news outlet would have had a rough time of it.

As an aside, a large, complex, and relatively well-documented event such as the Second World War (and the common knowledge that people have about it) furnishes all sorts of illustrations of the various forms of cognitive bias. Not so many people these days, unless they’re history buffs, are aware of lacunae such as the Phoney War, out-of-the-spotlight actions such as the Battle of Madagascar, roads-not-taken such as the shipload of mustard gas that sank at Bari, or tragic mistakes such as the Cap Arcona incident. These and many other parts of the record have been sanded down or paved over, not by any conspiracy, but by natural human tendencies.

I find, getting back to the Japanese situation, that I’m getting more useful information from blogs and even the Wikipedia pages on the Fukushima incidents than I’m getting from primary news sources. Those tend to have jumbled timelines, unclear sourcing, and all sorts of overlap and garble. Reading the efforts of various other people who are trying to make sense of it all (and checking them against each other) is so far providing me with more useful information. My television is turned off.

Derek Lowe, PhD, received his doctorate in organic chemistry from Duke University and pursued post-doctorate research in Germany on a Humboldt Fellowship. He’s worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer’s, diabetes, osteoporosis and other diseases. He comments about drug discovery and the pharma industry at In the Pipeline.

1 reply »