The general election campaign begins officially today, so let's think about colonoscopies.

In the winter of 1994-95 doctors in Toronto performed colonoscopies on patients and later asked them to recall how uncomfortable the experience was (pdf). They found that those patients who had the tip of the colonoscopy left in their rectum for a few minutes after the procedure recalled the experience as less unpleasant than those who had the colonoscopy removed immediately. This is evidence of what Daniel Kahneman calls (pdf) duration neglect; our memories of experiences are "radically insensitive to duration."

Here's another story. Constantinos Antoniou and Christos Mavis have found a systematic bias in betting on men's tennis matches. Bookies offer more generous odds on the better player in five-set matches than they do in three-set ones. They fail to appreciate that quality is more likely to tell in longer matches, whereas randomness favours the lesser player in shorter ones.

And here's a third. People are massively wrong about many basic social facts - for example, they hugely over-estimate the number of immigrants in the UK, the amount of benefit fraud or spending on foreign aid.

These three points seem different. But they have something in common. They show us that people are bad at judging scale. They just don't get proportions right.

Which brings me to Diane Abbott's interview yesterday. Yes, it was awful. But it poses three questions:

- Is this a story about Ms Abbott, or about the general standard of MPs? Yes, she revealed a horrible lack of mental agility and feel for numbers. In this, though, she is not alone. Less than half of MPs can correctly answer a question as basic as "if you spin a coin twice, what is the probability of getting two heads?" As Janan Ganesh says, politics is not well-stocked with talented people.

- To what extent is an interview with a local radio station evidence of ability? I said yesterday that being good at interviews might be a sign of dangerous over-confidence. By the same token, being bad at interviews need not be evidence of a lack of ability. There's a big difference between keeping one's head when under pressure and being able to take good decisions at leisure in the comfort of one's desk when surrounded by civil servants and spads who can provide you with facts and logic.

- How great was Ms Abbott's mistake? Errors are ubiquitous in politics and inevitable. And by those standards, Ms Abbott's incoherence barely registers. Boris Johnson wasted £37m of taxpayers' money on a non-existent bridge; the last government wasted over £1bn on the cancer drugs fund; George Osborne cost the economy tens of billions in needless austerity; David Cameron called a needless referendum in the mistaken belief he could win it; Tony Blair took us to war on the basis of errors of judgment which should have been widely known. And so on and on.

All those errors were due to persistent and corrigble mistakes, rather than to heat of the moment confusion. And all had effects which were orders of magnitude worse than Ms Abbott's.

From this perspective, the fuss about that interview is yet another example of scale neglect. The same professional and amateur pundits who are getting hysterical about a "car crash interview" (these people love their clichés) also regard Johnson, Blair and Osborne as "credible" despite making much more expensive mistakes, without the excuse of momentary lapses.

But I wonder. Perhaps in thinking in terms of mistakes and costs I'm missing the point. What pundits want from politicians is slickness and confidence, and just don't care that this might well be a front for grotesque incompetence. In failing to provide this, Ms Abbott committed a solecism as grave as that of the man who wears brown shoes in the City or who tries to get into a gentleman's club without wearing a tie - something which is utterly unforgivable.