Clean Run - October 2017 - 39

Cues, Jackpots,
and the
Curse of
Knowledge
© eileen anderson

By Eileen Anderson

I remember the moment I realized that
verbal cues didn't mean the same thing
to the dog that they did to me. It happened when a friend assigned a nonsensical verbal cue to a behavior she was
teaching her dog. The cue had no meaningful verbal connection to the behavior.
It didn't name or describe an aspect
of it. It wasn't a description in another
language. It wasn't even a joke. It was
just... random. It blew my mind because
it didn't mean anything.
In that moment I was able to finally
grasp something I had read: dogs don't
understand English or other human languages in the way we do. I realized that
verbal cues for them are probably just
sounds, without the subtext of meaning
they have for us. I got it then that a cue is
a signal, that it can be anything the dog's
senses can perceive and discriminate,
and that I had always overestimated my
dogs' understanding of English.
Not that I learned it once and for all. It
didn't stick. The consciousness of the
difference slips away time and time
again. I can't keep in my mind what the
dog doesn't know.
October 17 | Clean Run

The slipping away is normal. Some researchers have called this the curse of knowledge.
It's not that I'm so knowledgeable-it's that I don't have the ability to ignore things I
personally understand. I can't do it even when it would be in my interests to do so. For
instance when it could make me a better trainer with fairer expectations of my dog.

Beginnings of the Concept
Psychologists and economists are interested in the effects of differing knowledge between individuals, and also the effect of hindsight on our decision making after we
have learned something.
The curse of knowledge was named by Camerer, Loewenstein, and Weber in a 1989
paper called "The curse of knowledge in economic settings: An experimental analysis."
They were researching the effects on market transactions when individuals had different levels of knowledge. The experimental subjects were given information about the
earnings of actual companies. Then they had to perform actions in which they needed
to predict the judgments of others whom they knew didn't have that information.
Most of us would assume that in an economic transaction, having more information
about a product or company would always be better. But having more information does
not help us predict the behavior of someone who doesn't have that information. It impairs it. And sometimes we need that ability to predict. As the authors state:
"Better-informed agents are unable to ignore private information even when it is in
their interest to do so; more information is not always better... In predicting the judgments of others, agents are unable to ignore the additional information they possess."
(Camerer, Loewenstein, & Weber, 1989)

Tappers and Listeners
The curse of knowledge made it into the popular press after a Stanford University graduate student demonstrated it in a dramatic way in her dissertation (Newton, 1990). She
used pairs of volunteers. One person would tap out the rhythm of a well-known song
39

Table of Contents for the Digital Edition of Clean Run - October 2017