When entrepreneurs get out of the building to go and talk to their customers, they accumulate data. People get excited about the idea or they don’t, they ask good questions or misunderstand what the entrepreneur is saying. They cut the conversation short or keep it going past the allotted time. They brush the entrepreneur off or offer to sign up. Through many customer interactions, entrepreneurs identify patterns and iterate their MVPs until they are able to validate the product idea.

Or anyway, that’s what they think they’re doing.

In a famous 1980 study, the psychologists Robert Kleck and Angelo Strenta used makeup to apply disfiguring scars to their subject’s faces.* The subjects were told that the point of the study was to understand how people reacted to negatively-valued physical characteristics. They were to have conversations and report back about how they had been perceived and treated. Before going out to hold the conversations, the psychologists applied a moisturizer to the scars so that the makeup wouldn’t crack. Or that’s what they claimed to do. Actually, they removed the scars entirely.

The subjects went out, held their conversations, and reported back that the (nonexistent) scars caused people to treat them very poorly. Their conversation partners were tense and patronizing. The partners had been videotaped, so next, the subjects were shown the tape and asked to point out instances where the other person was reacting to the scar. None of them had any trouble pointing out moments when the partner had looked away or otherwise reacted badly.**

Human beings have a remarkable ability to recognize patterns and find causes for the phenomena around them. This ability is located in the left hemisphere, mostly in the left lateral orbitofrontal cortex and ventrolateral prefrontal cortex, and is commonly called the “interpreter module.” It is automatic, unconscious and ingrained, and it operates in parallel to perception, not subsequent to it. We don’t first notice and then make sense of things – making sense of and noticing goes on simultaneously.***

This system is prone to failure. The flip side of being good at finding patterns is not being good at finding that there is no pattern. In the Kleck/Strenta experiment, the conversation partners did not exhibit a pattern of avoidance behavior in reaction to imaginary scars. Instead, the subjects organized random or default behavior that they observed into a pattern. For example, they interpreted the partner looking away as averting their eyes. But people often look away during a conversation; turns out it doesn’t mean anything.

It’s easy for entrepreneurs to think that they are accumulating data and validating an idea or product. More than easy: during a conversation with a potential customer, it’s impossible to banish the sense that that’s happening. The interpreter module doesn’t have an off switch. But entrepreneurs who depend on it mistake random activity and indifference for data and validation.

What’s the alternative? From the point of view of the Kleck/Strenta subjects, the null hypothesis (how people act in conversations with others who do not have scars) and the alternate hypothesis (specifically how they might react to the scar) were both undetermined. Without them, we’re at sea, merrily interpreting whatever floats by.

*Perceptions of the impact of negatively valued physical characteristics on social interaction. By Kleck, Robert E.; Strenta, Angelo, Journal of Personality and Social Psychology, Vol 39(5), Nov 1980, 861-873.

**There’s an alternate interpretation to the outcome of the Kleck/Strenta experiment: Maybe conversation partners really did react badly, because the subjects, thinking they had the scars, acted differently. Both this team and others did multiple follow up experiments both to confirm the results and to determine if this was the case. It wasn’t.

** *Michael Gazziniga coined the phrase “interpreter module,” he’s the pioneer researcher in this area of brain function