Have you heard of Rashomon effect?
Contact me

Have you heard of Rashomon effect?

Tue, Apr 28, 2020

Read in 4 minutes

Let's talk about the Rashomon effect and different versions of one truth

I was always interested in versatile concepts.

Those that manage to navigate through several fields and still maintain the freshness and originality in each one of them.

I will give you an example.

There is something called the Rashomon effect, have you heard of it?

Well, you might say, I know there is a film.

Filmed in 1950’s (unbelievably 70 years ago) (and yes, it is in black and white), it was revolutionary and not for only one, but several reasons.

As a matter a fact it was so good that it made the Oscars consider giving the best foreign film award.

So why was everyone so obsessed with it?

It’s a story about two events: the murder of a husband and the rape of his wife.

The events are described through the eyes of the woodcutter, the bandit, the woman, and the spirit of the husband.

When they talk about it, everyone says their version of the truth, but each story is significantly different from others.

It is more complicated that saying they are lying, because none of them is lying only with a goal to protect their innocence.

In fact, everyone but the woodcutter implicate themselves as the murderer, making us believe that they really believe in what they are confessing.

In the end, the viewer is left to wonder: who was actually telling the truth?

This filmed forever changed the filmmaking and storytelling, but the idea that the truth is shaped by the viewers’ perceptions and their own understanding of the world found its place in many fields.

In computer science, Leo Breiman used the Rashomon effect to problems where many accurate, and yet different models can be used to describe the same data.

In economics, this idea is linked to so called ‘dispersion of knowledge’.

In sociology and psychology, Rashomon effect refers to people’s perceptions that can differ based on their individual experiences.

So, one concept, at least 3 different fields.

But what does this effect mean for people asking common business questions: why do customers only shop once or what predicts the new product uptake?

Where does this leave research and analytics?

This very question has been the backbone of a century-long argument between two paradigms in science: the positivist approach and interpretivist approach.

The positivists: statisticians, data scientists, anyone who cares about how much data we have, maintain that a true explanation of an event can be found and tested by scientific standards of verification.

This means that they might send out a survey and focus on the objective truth: if we have enough people that experience the phenomenon, we can objectively measure it and make conclusions.

The interpretativits, on the other hand, do not seek an objective truth, but claim that each person represents a universe on their own.

They might conduct unstructured interviews and aim to describe the subjective understanding of one’s truth.

However, the question remains, who is right?

The objective truth is surely right.

But a person that experiences emotions that are based on their perceptions are also right.

Because, can we tell someone what they feel is not real?

In our quest for truth there are some things we can do differently in order to not get trapped in one, narrow point of view.

Perhaps using different analytical lenses on one problem could be the solution.

Perhaps mixed methods, or combining the two approaches could give us better insights.

Can we change our perspective, turn the problem on its head?

Some argue it all depends on the nature of the question were asking.

Some argue we should reconsider the question itself.

But sometimes, the truth is greater than the sum of its parts or any of our partial understandings.

The Schrödinger’s cat is alive.

The Schrödinger’s cat is dead.

Sometimes it is both.

And sometimes… It doesn’t even exist.

Get awesome content in your inbox!