No one can be right all the time – but it turns out we can get better at judging when we might be wrong.
Knowledge about the accuracy of our knowledge is an ability called metacognition. Requiring self-awareness and introspective judgements, it is used in many everyday decisions, such as those about our social lives, studying or finances. For instance, we might decide to invest in a new venture because it seems promising, but choose not to risk much money because we realise we might be wrong.
People naturally vary in their metacognitive abilities, and Stephen Fleming of University College London wondered if this was something they could be trained to improve on.
Advertisement
Fleming’s team found that a short period of training could boost metacognitive abilities, not only for the visual task used in training, but also in an unrelated skill involving memory. “It suggests a training protocol might have quite general benefits for your metacognition – that’s the result we are very excited by,” says Fleming.
Fleming and his colleagues recruited 61 people online. Each was presented with a series of paired images and asked to decide which one in each pair had the brightest lines shown against a dark background. Participants were also asked to rate on a scale of one to four their confidence in each decision.
About half of the people were in the “control” group and were simply told how well they did on the brightness question. The rest were in the “training” group and were given feedback on the accuracy of their confidence judgement. “They got points if they were right on the task and confident, but also if they were wrong and not confident,” says Fleming.
The volunteers earned money from their point score to encourage them to come back for training sessions spread over several days. After eight 20-minute sessions over a two-week period, the training group’s scores for metacognitive ability on the visual task rose from 0.74 to 0.84 on a scale of 0 to 1.0. Crucially, there were similar improvements in their confidence judgements on a different task, involving trying to remember words they had previously been shown. The control group’s scores didn’t improve on either measure.
The team is now repeating the experiment but scanning people’s brains during the tasks, to see if the changes in ability are linked with more activity in the parts of the brain known to be involved in metacognition. It’s unknown how long any effects would last.
Nick Yeung of the University of Oxford says it’s still unclear if the training would improve people’s decision-making generally. “A reductionist view is that they get good at using that scale. In everyday life people think about their decisions but don’t rate them on a one-to-four-point scale.”
“What we need to do next is to see if these benefits translate to improvements in metacognition in the real world,” says Fleming.
Other groups have found that metacognitive abilities can be improved by meditation, as well as by taking a and a .
Yeung says teachers are becoming interested in ways to improve students’ metacognitive abilities, so they can better judge whether they have mastered a topic or if they need to keep at it. Animals were once thought to lack metacognitive abilities, but recently dolphins and some monkeys have shown glimpses.
is due to appear in the Journal of Experimental Psychology: General
