Search
Browse By Day
Browse By Time
Browse By Panel
Browse By Session Type
Browse By Topic Area
Search Tips
Register for SRCD21
Personal Schedule
Change Preferences / Time Zone
Sign In
X (Twitter)
We live in a world that is dynamic and rapidly changing. Oftentimes, we rely on those that are younger than us to help us adapt. For example, we help our older colleagues and parents figure out how to use Zoom. Younger learners are often better at noticing irrelevant features (Plebanek & Sloutsky, 2017) and perform better on certain types of causal reasoning tasks than older learners (Gopnik et al., 2017). The current set of studies test two hypotheses: 1) that children have a bias towards exploration, and 2) that this exploratory bias can result in better learning.
Study 1 included 48 children ages 3-to-8. We created a gambling machine tailored to children. Children began each trial by depositing a coin into a machine with two large buttons. When a coin was inserted into the machine, the buttons glowed, indicating the beginning of a trial. Each button corresponded to one of two reward bins (e.g., capsules with stickers inside of them) each with a unique, hidden payoff distribution. After the child pressed a button, a clear ball popped out of the top of the corresponding side and rolled down a series of tracks on the side of the machine. Each ball contained 0, 1, or 2 stickers. Half of the balls on the blue side contained 1 sticker; and the other half contained 0 stickers (average payout of 0.5). On the green side, half the balls contained 1 sticker and the other half contained 2 (average payout was 1.5). We found that most children tend to continue to switch between each side for the entirety of the task, suggesting that children have a bias towards exploration.
In Study 2, 74 6-to-12-year-old children and 134 adults were shown four animated monsters, each associated with some reward (of 1, 2, 3, and 6 stars). Participants were instructed to click on the different monsters to collect stars. After participants click on a character, the screen reveals how many stars they won. After a certain number of stars are collected, participants were rewarded with a sticker. For the first 40 trials, the reward values remain the same. For trials 40-81, the monster that was giving 1 star starts giving 8 stars. We found that children were more likely than adults to keep clicking different monsters, which meant that when the values changed, children were much more likely to detect the change than adults were. Due to their continued sampling, we found that children were better able to identify the 2- and 3-star monsters significantly better than adults.
Taken together, these two studies demonstrate that children have a bias towards exploration. The tendency to explore at high rates can lead to the discovery of surprising information, which is useful when learning about the intricacies of a changing world. Thus, a higher level of exploration during childhood can be rational and leads children to discover information that adults miss.
Emily Sarah Sumner, University of California - Berkeley
Presenting Author
Amy Li, University of New South Wales, Australia
Non-Presenting Author
Andrew Perfors
Non-Presenting Author
Brett Hayes, University of New South Wales, Australia
Non-Presenting Author
Danielle Navarro, University of New South Wales, Australia
Non-Presenting Author
Barbara W Sarnecka, University of California - Irvine
Non-Presenting Author
Celeste Kidd, University of California - Berkeley
Non-Presenting Author