By Alex Fradera
Whether we’re testing our mettle on a video game, on the golf course, or at the bowling alley, it’s good to have a realistic sense of our ability, so we attempt things that are feasible – and don’t accept unwise bets. But how accurate are we at judging ourselves in this way? In a new study in Neuron, researchers from Oxford University have shown that our sense of our own ability is coloured by the other players around us. Specifically, their findings suggest that when we’re competing with a strong player, we tend to downgrade our own ability. Conversely, when that player is on our team, we see ourselves as better than we really are.
Marco Wittmann and his colleagues asked 24 participants, mostly in their twenties, to play a series of short mini-games: for example, making judgments about the colours of shifting shapes, or estimating the time passing between flashing items. Participants played each mini-game simultaneously with two other “players” – actually research assistants who only pretended to play.
After each round, participants were given fake feedback on how they and the other players had just performed. They were encouraged to use this to make predictions on how many points they and the others would score on the next round of play. Their performance predictions were influenced by this feedback, which shows they were following the instructions.
In each round of the game, participants were paired with one of the other players and told either that they were a team (with success based on whether their combined performance reached a certain threshold), or that they were competing (with success dependent on their beating the other person by a high enough margin). In theory, this framing shouldn’t have affected their judgment of their individual performance because although their scores were combined or compared, their own performance was independent of the other player. But in practice, it did.
When teamed up with a previously strong player, participants rated it likely that their own performance would be especially strong in future. When pitted against a strong player, they rated their own future performance as more likely to be poor. A complementary pattern arose for ratings of the other player: participants who were good at the game thought a teammate would do better than his or her performance would suggest, and a competitor would do worse.
Wittmann’s team describe this as a “merging of estimates”. During collaboration, this works a little like an anchoring effect: mentally, you ride on the coat-tails of a teammate, assuming that you are sort-of-similar in ability. But when competing, you exaggerate difference: the better the opposition, the more conscious you are of your own deficiencies.
We’ve known for a while that collaborative contexts draw us to notice traits we have in common with others, but this is the first work to show that these contexts also influence our evaluation of a particular trait. So if you’re playing golf on Tiger Woods’ team, beware of taking side-bets on your putting game: your sense of what you can deliver is likely to be off.
(This work was brought to our attention through an excellent commentary in Trends in Cognitive Science.)