Worry, Not Bodily Tension, Enlarges the Brain’s Error Signal

A tenth of a second after a mistake

Error-related negativity is the name for a burst of electrical activity produced by the front-middle of the brain within a tenth of a second of a mistake, before the person has any conscious awareness of having made one. It is recorded at the scalp with EEG while someone works through a fast, dull, repetitive task. A bigger burst is a bigger alarm, and the question a group of researchers took up in 2013 was who carries the bigger one.

Their answer, pooled across 31 studies, was that the alarm grows with anxiety, at a correlation of 0.25. The paper is a meta-analysis by Jason Moser, Tim Moran, Hans Schroder and Brent Donnellan at Michigan State University, with Nick Yeung at the University of Oxford, published in Frontiers in Human Neuroscience. Its authors call that pooled figure small-to-medium, and it averages two kinds of anxiety that behave nothing alike.

Anxiety was split in two. Anxious apprehension is worry: verbal, forward-looking, a rehearsal of what might go wrong. Anxious arousal is what the paper calls somatic tension and physiological hyperarousal, the body’s answer to a threat already in the room. Sorted that way, ten studies of apprehension produced a correlation of 0.35 while every other measure of anxiety produced 0.09, a gap whose confidence interval excluded zero. The abstract puts apprehension’s effect at threefold.

Twelve of the studies measured the identical brain response on trials the person answered correctly, a signal called the correct-response negativity. There the correlation was 0.063, with a confidence interval that includes zero, meaning no reliable association with anxiety however anxiety was defined. The enlargement is confined to the moment of getting something wrong. A worried brain, on this evidence, treats a correct answer much as any other brain does.

The pattern also survived the removal of every clinical sample. Among healthy volunteers alone, apprehension reached 0.301 and the remaining anxiety measures 0.101, which places the effect inside the ordinary range of temperament.

A name for what you have watched

The alarm has a target. It fires on error, and fast enough that the mistake registers in the brain before it registers in the mind.

A child answers, and something changes in the face or the shoulders before anyone has said a word about the answer being wrong. The same child is unbothered by the difficulty of the next question and unbothered by the room, and then another wrong answer arrives and the sequence repeats.

The vocabulary for that is anxious apprehension, the worrying kind of anxiety, and the meta-analysis attaches the enlarged error response to that kind in particular. A jumpy body and a worrying mind are different things in this literature, and only the second tracked the alarm.

This is a literature of adults. Of the 37 studies, 19 used healthy adult volunteers and 16 used people with diagnosed anxiety disorders, while two used healthy children, and age was never tested as a factor. Any bearing on an eight-year-old is an inference drawn from adult data, and a parent is entitled to hold it loosely. The underlying studies were also small, with a median of 40 participants and the smallest at 18, and the division between worry and bodily tension was built from what the questionnaires contained rather than measured in the participants directly.

Nor can any of it say which way the arrow points. Whether worry enlarges the alarm, the alarm feeds the worry, or something else produces both, a meta-analysis of correlations has no means of telling. It tested no treatment and followed no one over time, and the mechanism the authors propose is a framework offered to explain the finding.

What is left is a description, and a good description does work of its own. A child who is vigilant at errors and untroubled through the rest of the hour is showing a vigilance with an object. That object has a location in the brain and a measured size behind it, which is a considerable improvement on being called sensitive.

The next time your child registers a wrong answer before anyone in the room has spoken, you will be watching something with a name, and with a target no wider than the mistake.