PsyFi Search


Wednesday, 15 June 2011

Fooled By Fluency

Q: How many animals of each kind did Moses take on the Ark?

Mere Exposure is Not Enough

In 1968 Robert Zajonc identified an odd behavioral bias, known as the mere exposure effect. This predicts that people will prefer items that they’re familiar with over those which they’re not – and that repeatedly exposing people to something will increase their preference for it. Cue a thousand feeble advertising campaigns.

There’s more than a hint of suspicion that behind mere exposure lies a more fundamental bias, one that spills over into all sorts of apparently irrational behaviour, some connected to money and others not. The idea is that familiarity breeds fluency, or ease of neural processing. So if you’re thinking that Moses had an Ark, that you should name your company q@l&tee or that your favorite whacky font type doesn't affect people's financial I.Q. then think again.

Disorted Fluency

There's a huge variety of research that shows how ease of processing information – fluency – distorts judgements. So, for instance, rhyming aphorisms seem more true than non-rhyming aphorisms. Hence a stitch in time saves nine but too many cooks don’t spoil the broth. And, of course, sell in May and go away feels right, while selling your losers and running your winners is rather harder to follow.

Other examples are that more easily imagined travel destinations are preferred to harder to imagine ones, that statements written in an easier to read font engender confidence and text that’s easier to process is believed to written by a more intelligent author. There’s a long list of these fluency judgements in this paper by Alter and Oppenheimer, Uniting the Tribes of Fluency to Form a Metacognitive Nation. With a complicated title like that you'd think those guys are really dumb, but more of them later.

Money Illusions

These fluency effects spill over into all sorts of areas. For instance, in Money: A Bias for the Whole Mishra, Mishra and Nayakankuppam showed that:
“Greater value is perceived for money in the form of a whole (large denomination) than for equivalent amounts of money in parts (smaller denominations), resulting in a lower inclination to spend with the whole”
In essence the same amount of money presented in the form of small denomination notes or larger denominations led to a change in spending intentions: people were much more reluctant to spend higher denomination notes. The researchers suggest that this is a fluency effect: it’s easier to attribute the real value to a larger note, rather than a bunch of loose change, so people are less likely to spend it.

This bias for the whole is not undisputed. Raghubir and Srivastava have proposed a denomination effect which is related to self-control: the use of larger denomination notes as a precommitment device to stop spending. However, Alter and Oppenheimer (them again) have also shown that people will preferentially believe that an amount of money in a familiar form has greater purchasing power than the same amount of money in an unfamiliar form: so a standard $1 bill was preferred to an unusual $1 coin, and so on. All of which leads the researchers to conclude that when governments start mucking about with currencies they may be causing more harm than they realise: there’s nothing like changing your currency for slowing down the economy.

The Moses Illusion

The classic example of how our innate bias for fluency betrays us is the Moses Illusion, quoted at the head of this article. Most people will answer “two” to the question, despite knowing that Moses never so much as got his feet wet in the Red Sea, and definitely never around recruiting for Noah’s Arc. Song and Schwarz used this trick to analyse how we process familiar and unfamiliar information. Their conclusion was that:
“Low processing fluency improves performance when one’s spontaneous answer is wrong, but impairs performance when one’s spontaneous answer is correct.”
So basically, if you actually know the correct answer you’re more likely to get it wrong if it’s presented in a non-fluent form, but more likely to get it right if you don’t immediately know the answer. In both cases the inability to use fluency as a cue forces us to process information more deeply, but also means we end up doubting our own instincts when we’re right. Basically, new information presented in a familiar form isn’t likely to cause us to change our minds, even when the content should, while information presented in an unfamiliar way may cause us to revise our opinions, even when there’s no reason to do so.

Value in a Name

If familiarity is linked to our perceptions of value and if these perceptions of value are underpinned by fluency then you might predict some peculiar impacts on the stock market. You might, if you were particularly mad, suggest that stocks with names that are easier to remember and pronounce, thus promoting fluency, would be preferred to stocks with odd and difficult names. Of course, this is a stupid idea because we all know stocks are valued based on fundamentals like earnings, assets and relative valuation against other companies.

And, of course, it turns out that this stupid prediction appears to be true. Here's Alter and Oppenheimer again (who do get around, don't they?):
“Three studies investigated the impact of the psychological principle of fluency (that people tend to prefer easily processed information) on short-term share price movements. In both a laboratory study and two analyses of naturalistic real-world stock market data, fluently named stocks robustly outperformed stocks with disfluent names in the short term.”
Of all the bizarre behavioural biases we’ve yet examined this has got to be one of the oddest, related to the behaviors we saw in The Halo Effect: What's In A Company Name?, yet it points to some fundamental behaviours that we probably ought to be attempting to consciously overcome if we want to be successful investors. The standard explanation for fluency effects is that we’re attempting to avoid thinking hard and are looking for simple cues to make decisions – which is a perfectly valid thing to do in much of life, but is probably not that wise when it comes to money.

Financial IQ

To check this, the incredibly industrious Alter and Oppenheimer and colleagues plied some innocent participants with a variation of the Cognitive Reflection Test we discussed in What’s Your Financial IQ? They varied the test by presenting it in two forms: an easy to read font and a hard to read font. The results were fascinating: 65% of the people in the hard to read font sample got all the questions correct, compared to only 10% in the easy to read sample. Which appears to prove that font type determines financial IQ, an odd result if there ever was one.

More practically, it seems that the difficult to read font forced people to process the questions analytically, because they couldn’t easily take their cues from fluency and familiarity concepts. In essence the experiment enforced the type of analytical and careful thinking that marks out people who naturally do well at the CRT, and who are more likely to make smarter investors.

All of which reinforces the idea that investment is a process best done slowly and rationally. Mere familiarity is not equivalent to proper analysis and just because information is easy to process doesn’t mean that we’ll come to the correct answer. Even worse, though, if bad news is presented in a way that looks familiar our fluency cues may override the detailed processing that we need to do: all of which suggests we ought to look at everything with suspicion. Of course, as most news ends up as tomorrow’s packing case lining, the alternative might be to do nothing.

Related articles: The Halo Effect: What's In A Company Name?, What’s Your Financial IQ?, The Language of Lucre


  1. thanks for the insight I walked right into the Moses question it really proves your point

  2. outstanding writing backed by disparate by relevant research. Interesting that despite your very fluent writing, the argument was heard loud and clear. Now think that my financial IQ has been raised. Well done.

  3. I’m really liking your blog since finding it on Abnormal Returns. I’ve never studied psychology, so a lot of this is new to me.

    I have to wonder about the denomination effect. People of a certain age remember when breaking a $100 bill in a retail store was a difficult thing to do. Merchants just didn’t keep that amount of money in their registers. It wasn’t that long ago, if you wanted to break a $100 bill, you had to go to a bank.

    Also, it’s much easier to have a $100 bill fall out of your pocket unnoticed than five $20’s. As for $1 coins, it’s common knowledge that merchants have nowhere to put them. All of this leads to foreknowledge of certain forms and denominations as being impractical for daily use. It seems to me that money’s role as an ease to commerce was neglected in the study.

  4. Hi Ron

    It's a fair point: many of these studies lack a sense of being grounded in the real-world. In fact there's evidence to show that some of the results of behavioral studies in the laboratory don't translate to real life at all.

    Probably the guy that does the best research in this area is John List, who I've written about a number of times: the blog's google search bar will thrown up the articles (assuming it's working), but his main point is that if your stick a person in an unreal situation they'll try and make sense of it. That leads to odd results - and odd experiments, which is what you're (correctly) pointing out. Still, it keeps us honest, so to speak :)

  5. Genesis 6:20 says "Two of every kind of bird, of every kind of animal and of every kind of creature that moves along the ground will come to you to be kept alive".

    BUT Genesis 7:2 says "Take with you seven of every kind of clean animal, a male and its mate, and two of every kind of unclean animal, a male and its mate"

    Nothing in investing is easy, even when you KNOW that that you're smarter than everyone else ;-)