For any one of us, there is just one mistake: a violation of the plan, a failure of execution. We fail to catch an error in DNA replication or we choke on a pretzel that we intended to swallow or we step repeatedly into the same ice puddle. We might mistakenly think we have the facts in hand in celebrating a birthday or offering to help someone understand the Kibble-Zurek mechanism.
Mistakes are defined by the bad consequences of an intended act. Although you can make a mistake on a bet, you can’t make one on an honest coin toss. Evolution can’t really make a mistake because it is blind. There are random events—coin tosses—in mutation and in independent assortment.
We all make mistakes because every act is accompanied by unintended variation, often random error; mistakes are a part of learning; and doing one thing may be incompatible with something else we’re doing at the same time. That’s why we push the wrong button on the vending machine two percent of the time.
Mistakes of carelessness and ignorance usually find few victims, although a crowd isn’t needed to experience the pain. Errors in investment, a doctor’s office, or a courtroom provoke wider concern. Just a very few become classics.
Fortunately, biological and cognitive research has turned up ways to detect and correct for mistakes.
BIO: Fortunately, we monitor our actions in a couple of ways, to find out if our bodies end up in the expected position and whether we obtained the desired reward.
The neurons that catch our errors are found in the medial frontal cortex, which you would see if you pried the two cerebral hemispheres apart along the longitudinal fissure and examined the brain that lies just anterior to, or in front of, the corpus callosum. These neurons generate a signal called error-related negativity (or ERN) in the electroencephalogram (EEG) when we make an error.
The cerebellum matures quickly in the fetus. The frontal cortex matures very slowly into adulthood. Neither of these brain structures seems to detect mistakes made by infants and children, however. They do not monitor error size. and lack mature cerebral cortex, yet toddlers as young as 2 ½ years old generate a feedback-related negativity that resembles the ERN.
Another mechanism for detecting and correcting errors is our reward prediction system. It predicts the availability of reward and signals errors in obtaining it using dopamine-releasing neurons. Minimizing errors in obtaining rewards then appears as learning.
Of course we also monitor errors made by other people, and our pupils dilate each time we detect one. Pupillary dilation marks attentional arousal, along with some other changes in psychological condition.
PSYCHO: Errors in performance are common. If feedback warns us of a mistake we show the immediate (and automatic) reaction of post-error slowing. This presumably gives us time to re-evaluate and strategize for a new decision.
However, that’s not simple. As we repeat our mistakes, we face quite a challenge. Learning a habit is a hit-or-miss, error-prone process, then habits become stubbornly automatic, for better or worse, residing no longer in the cerebral cortex but mostly in the basal ganglia, which links skills to habits.
It’s remarkable when some particular context or goal trigger a long sequence of behavior. We routinize behavior that we want to repeat until about 45 percent of our activities (p. 27-28 of the journal, not your Adobe Reader) occur “on automatic”. When the context or goal changes becomes inappropriate, we start to make mistakes. In principle, at least, that’s when we start to pay attention.
Mistakes that are regarded as pathological support diagnoses that may also be mistaken. Although there are websites describing clumsiness as a type, I’m skeptical because the diagnostic category is sloppy.
The “clumsiest president” was, in fact, not clumsy. Kids with ADHD may be labeled clumsy when they are, instead–like Gerald Ford–sometimes inattentive. Other kids might be seen as clumsy because they are small. Clumsiness might also reflect Parkinson’s disease, multiple sclerosis, or a stroke.
Aside from performance mistakes, perceptual and cognitive errors are common as well, partly because so much of the environment is ambiguous. Although disambiguation can be fast—50 milliseconds—it is not necessarily correct.
Often we don’t realize that stimuli are ambiguous, as with The Dress or the Yanny-Laurel perplexity. For the dress, the widely accepted explanation is color constancy. For Yanny and Laurel, the sound is ambiguous, so we can’t process it in bottom-up fashion as we would with a donkey’s bray, for example. There’s nothing unique about “Yanny” or “Laurel”. Try “Brainstorm” vs. “Green Needle” for a duplicate experience.
At least in conversation, ambiguity helps us to communicate by making speech more efficient, surprisingly (research here). To remove the ambiguity we rely mostly on context. Without ambiguity we would have an even harder time communicating.
SOCIAL: Our way of reacting to mistakes betrays some cultural influence, inasmuch as both post-error slowing and rror-related negativity vary between collectivist and individualist cultures. This is perhaps a biological echo of anthropologist Ruth Benedict’s labels of “shame culture” and “guilt culture” applied to collectivist and individualist cultures, respectively.
Travelers become aware of their cultural offenses with the help of numerous guides to differences in customs around the world. Now that everyone is a virtual traveler through the Internet, corrections have also become global in origin.