Reliable generalizations about human behavior are hard to come by, and much sought after. We want to understand ourselves as much as anyone else, and the notion of broad, easy-to-understand categories of personality promise to simplify things. Categories – personality types – suggest that understanding our motives, biases, and behaviors, can be less messy and probabilistic. Often enough, the tools we use to place ourselves in these categories come with the imprimatur of science.
The stamp of scientific is never more strongly emphasized than when the personality categorization is based on neuroscience. Thus, we’ve learned for years that we are either right or left brained, even as the meanings associated with right and left brained personality shift like dunes in the desert. Many readers will already know that such broad generalizations have not amounted to much in applied settings, even though they strike us as interesting.
Now Stephen Kosslyn and Wayne Miller have unveiled a new set of generalizations with their Theory of Cognitive Modes, courtesy of the Wall Street Journal, and their new book Top Brain, Bottom Brain: Surprising Insights Into How You Think. According to their theory, people are biased toward different modes of thinking, modes that are organized primarily around the top brain, which isoccupied more with formulating plans and integrating all manner of neural information, and the bottom brain, which is more dependent upon current sensory information.
Grounding their theory in “decades of unimpeachable research,” Kosslyn and Miller quite reasonably suggest that all people work with all parts of their brain more ore less all the time. This forms at least part of the basis for their criticism of categorization rubrics that have come before, such as the whole right-brain/left-brain dichotomy we’ve all grown used to. Also reasonably, they note that people rely on these different brain systems to varying degrees: this simple (and itself unimpeachable) truism forms the foundation of their own typological scheme. Depending on how your particular brain is organized – or rather, depending on how you answer a series of questions written by Kosslyn and Miller – you can be said either to operate primarily in Mover Mode, Perceiver Mode, Stimulator Mode or Adaptor Mode.
I took the test myself and learned that I operate mainly in Situational Perceiver Mode, meaning that I have a tendency to rely on bottom-mode thinking. More specifically, I “tend to interpret what is experienced, put it in context, and understand its implications.”
So, have Kosslyn and Miller cracked the personality code?
No, they haven’t.
Measures of personality can be quite stable across time (meaning that a person who scores high in something now will also score high in it in months or years from now). Yet such measures predict surprisingly little about actual performance – even the really good measures developed by psychologists over the past 50 years or so. Questionnaires like the popular Myers-Briggs are unlikely to yield anything of value beyond material for water cooler conversation. Despite what you may have heard from human resources consultants, such measures of personality simply do not work, if by work you mean increasing the performance of a company, team or particular employee. Worse, they may waste precious time and money.
Forty-five years ago, psychologist Walter Mischel published a critique of personality research in psychology that is as devastating today as it was then. His critique was simply to note that decades worth of careful empirical research had shown that personality types are not consistent across contexts. The problem of validity – of predicting behaviors of interest based on a test – has plagued personality researchers, but the reasons for the problem are now better understood.
It isn’t that people do not have stable biases or tendencies that can impact their behavior; they do. It’s just that those biases are much more context-dependent than popular notions of personality acknowledge.
I have observed this problem in my own research. More than a decade ago I sought to measure links between a certain pattern of activity in the brain’s prefrontal cortex and a personality style called negative affectivity, which refers to the degree to which you respond negatively to stressful events. At the time, the standard method for measuring neural patterns was based largely on something called the “resting task.” In this task, brain activity was recorded while people sat doing nothing. The idea behind this measurement approach was that when people sat still and were instructed to avoid engaging in any kind of task, you could then get a fix on their trait levels of brain activity. The theory was that “resting” or baseline activity would give you a strong idea of who each person was, in a general sense, across situations and contexts.
I found that measuring resting brain activity was not very helpful, however. Data collected by this method often didn’t replicate across laboratories, and just as often didn’t predict the things it was supposed to. On the other hand, I found, as Mischel would have predicted, that when brain activity was measured in specific contexts, it was both more reliable as a measure and more valid – it did a better job of predicting important outcomes.
If you think about it, this all makes sense. Humans are highly adaptive creatures, and our brains reflect this adaptability at almost every level of organization. Nowhere is this more in evidence than in the prefrontal cortex, which may have evolved to increase behavioral flexibility in a world of frequently shifting contexts. Indeed, it may be this capacity for flexibility that has allowed humans to live virtually everywhere on the planet; humans may have conquered the world in part because we are not locked into the kinds of fixed responses that other animals need to survive in specific environments.
But this is all a little abstract, so let’s break it down with an example. I’ll use myself. On standard tests of the “Big Five” personality dimensions, I score very high in the trait of Extraversion. This means that I like stimulating environments that include other people. Even a moment of introspection suggests that this is basically right. I’m never happier than on the National Mall in Washington during the 4th of July, with all the crowds and commotion.
But Mischel’s work, and my own, predicts that I might not be extraverted in all situations. And that’s absolutely right, too. For example, I get really very nervous at dance parties. If I’m at a dance party, my preference is to find remote corner where I can chat with somebody one-to-one. This is classic introvert behavior.
Over the years, Mischel and many other researchers have observed similar contextual tendencies in most people. People do indeed have personality profiles, the evidence suggests, but they are best understood as “if-then” statements, like, “If on the National Mall on the 4th of July, Jim behaves like a classic extravert. But if at a dance party, Jim behaves like a classic introvert.” That’s bad news for generalizations about individuals that are overly broad. Such generalizations just don’t hold up, empirically. Most commercially popular personality and hiring tools boast of being “statistically valid.” But dig a little further, into published or (often enough) unpublished reports, and you find what Mischel did: the predictive power across contexts is very, very small.
Luckily there is some good news, too. The “if-then” statements that constitute a person’s personality appear to be quite stable across time. This means that I’m likely to always be extroverted at the National Mall on the 4th of July and introverted at dance parties.
The critical thing is to assess personality in terms of “if-then” statements, like “if in a negotiation, Sarah works in mover mode; if at the office on an ordinary day, she works in adaptor mode.”
As far as I know, exactly no one is attempting to do this on a broad scale in applied settings. Until they do, we’ll continue to have lots of fun comparing our personality types and styles and so forth at the water-cooler – including Kosslyn and Miller’s. But they won’t predict much more than would a daily horoscope.