PsyFi Search

Tuesday 21 August 2012

Birds, Bees and Decoy Effects


If there's one message I hope to get across here it's that we're our own worst enemy when it comes to making decisions if we're anywhere in the vicinity of money.  If there's a second one it's that if we don't damage ourselves with our behavioral biases then someone else will try and do it for us.  Nowhere is this more evident than with the fiendishly devilish decoy effect.

The decoy effect was made salient by Dan Airely in Predictably Irrational, presumably as a warning for us all about the dubious machinations of marketing departments.  In a world where no good deed goes unpunished it's promptly spawned a thousand websites excitedly describing how to fleece their customers.  You've got to love the entrepreneurial spirit even if the moral outlook is distinctly cloudy.

Orange Sherbet
“A person sitting in a restaurant looking at the dessert menu chooses crème brûlée over the tiramisu.  Upon learning that the special dessert of the day is an orange sherbet, he then changes his mind and orders the tiramisu”.
This comes from Dan Ariely and Thomas Wallsten's Seeking Subject Dominance in a Multidimensional Space: An Explanation of the Asymmetric Dominance Effect, and provides a neat demonstration of the issues posed by the decoy effect.  After all, if you normally prefer crème brûlée over tiramisu you wouldn’t expect to suddenly change your mind because someone introduces orange sherbet.  However, this is exactly what the the decoy effect predicts.  The introduction of a third, irrelevant option, the decoy, can cause us to reverse our preferences.

This breaks two of the basic rules of rational choice theory, the standard economic mantra about how we make choices: transitivity and regularity.  Transitivity predicts that if we prefer A to B and B to C then we’ll prefer A to C.  However, as  we're not dealing with matters of pure logic but the logic of grey matter it turns out that we regularly break this rule.

Regularity predicts that if we add a new option C then this shouldn’t change our preference for A over B, aka orange sherbet shouldn't cause us to stop liking crème brûlée and become rapt consumers of  tiramisu. Unfortunately adding new choices to the menu can cause to do exactly this.


Of course, the main alternative to rational choice theory is the behavioral concept of bounded rationality.  Rather than assuming that people are perfect maximizers this starts from the idea that we have limited processing power, and therefore make mistakes and generally compromise around our preferred options. 

Amos Tversky, one of the inventors of theories of bounded rationality, showed as far back as 1972 that if you introduce a new choice then it will take a disproportionate share of the market from similar existing choices, which has led lots of marketers to build careers designing dissimilar products.  As these dissimilar products need to fit within a branding scheme this often generates lots of cognitive dissonance, and a range of rather lame product marketing efforts.  This effect is known as the similarity hypothesis.

However, research into the similarity hypothesis has shown that it to can often be violated, and in the most peculiar fashion.  Researchers have demonstrated that if you add a new choice   the decoy  which is similar to an existing one, but which is less popular – it’s “asymmetrically dominated” by the existing choice – then this will increase the market share of the existing choices which are most similar to it.

This offers firms the opportunity to push more expensive options onto hapless consumers, and no doubt there are lots of them doing this having been helpfully made aware of the opportunity.  Whether that’s moral or not is an open question, but frankly exploiting our irrationality lies at the heart of most marketing and sales campaigns.


The outcomes of the decoy effect are decidedly odd.  As expressed by Joel Huber, John Payne and Christopher Pluto in Adding Asymmetrically Dominated Alternatives: Violations of Regularity and the Similarity Hypothesis:
“Managerially, the results lead to the counterintuitive conclusion that there are times when the profitability of a product line can be increased by adding a (dominated) alternative that virtually no one ever chooses. The effect of a dominated alternative is to draw attention to a more profitable item rather than to generate direct sales”.
What’s happening here?  Well, the suggested hypothesis is that the decoy is acting as a reference point – an anchor – and providing a separate frame against which the preferred option is being compared.  Because we can easily compare the preference against the decoy we can easily assign a value to it, while dissimilar options are less easy to quantify in this way.  Basically we analyse our choices based on the information that’s most easily brought to mind – and if we’re presented with a decoy this makes the comparison easier.

As usual, underlying this odd behavior is some interesting neurology.  A brain scan study by William Hedgecock and Aksha Rao discovered that introducing the decoy makes the brain’s decision processing more pleasurable – or, to be precise, it reduces negative emotions as compared to when we’re faced to with two roughly equal alternatives.  This is cognitive dissonance writ in neurons, and the effect of the decoy is to relieve the tension.  The researchers tag our unwillingness to face up to difficult decisions as trade-off aversion.

Birds and Bees

Although the evidence has been growing for decades that we don’t conform to rational choice models of decision making it’s now becoming evident that animals don’t either. For reasons that aren’t immediately apparent researchers have also plagued hummingbirds with decoy effect experiments and it turns out that they too make choices contingent upon the options available to them, see: Irrational Choices in Hummingbird Foraging Behavior.  It's suggested that this to is due to limited processing power and results which don’t matter much in the only evolutionary scoring system that’s important: survival. 

Similar findings have been made with grey jays and honeybees.  Of course, it’s not entirely clear how often ancestors of humans, birds and bees would have faced these choices, since the ascent of behavioral researchers and marketing departments is a relatively recent invention, so what we may have here is what is known as an emergent property of cognitive systems – an accidental by-product that wasn’t relevant in evolutionary terms.

Credit Cards and Cash

The power of the decoy effect appears to be significant.  In an unusual experiment Promothesh Chatterjee and Randall Rose looked at the effect of priming people for either cash or credit cards who were then presented with product choices.  The hypothesised that people primed with cash would then focus on the costs of products while those primed with credit cards would focus on benefits – and this is exactly what they found.

However, they then introduced decoy products.  In the cash option they introduced a decoy for the benefits choice, while in the credit card option they introduced a decoy for the costs choice.  This countered the initial priming, reducing the salience of the benefits choice for the credit card primed and of the cost choice for the cash primed.  The decoy effect is surprisingly powerful, because these types of priming effects are unconscious and usually hard to overcome.

Be Very Prepared

The evidence that the decoy effect can be found across species is a fascinating development, because it suggests that at least some of our behavioral biases are very, very fundamental.  After all, we can't accuse hummingbirds and honey bees of letting their irrational emotions get in the way of logical analysis.   The suggestion that it's a side-effect of simplifying cognitive processing, albeit at the cost of making less than optimal decisions, is intriguing, because it adds to the evidence that a good enough decision made quickly is often better than a perfect one made slowly, an idea we explored in Satisficing Stockpicking.

Unfortunately, translated into the modern world, especially where we're constantly dealing with the automated outputs of comparison websites, e-commerce sites and stock screeners, this exposes us to the dangers of demented e-marketers determined to push up their profits at our expense.  No one is going to stop them, it's simply capitalism in its rawest form, so a little knowledge may, for once, be a valuable thing.

The decoy effect added to The Big List of Behavioral Biases.

Related articles:

1 comment:

  1. Hmmmm, I was half expecting your amazon store to be displaying prices after that post...