Peekbank
The open database
of
infant word recognition
Peekbank aggregates looking-while-listening data from 44 datasets into one uniform, tabular schema. Plan a study, replicate an effect, or reanalyse across the whole literature from a single query.
“Where's the ?”
What is Peekbank?
Various language development studies use the “Looking-While-Listening” task: A child is shown two items, a prompt names one of them, and the child’s resulting gaze is recorded.
Members of the Peekbank team manually curate these heterogeneous datasets and bring them into one unified format. Automated validation and standardized reviews ensure high data quality.
Once in Peekbank, the cleaned up data is freely accessible to anyone to use in their research. The labs supplying the original data get cited in research outputs.
Want to work with Peekbank?
Access the data
All datasets are free to use in your own research. Our R package loads them in a few lines, in the same format across every study.
Contribute data
We convert your data into the Peekbank format and distribute it via the database. Whenever your dataset is used through Peekbank, your original paper gets cited.
Science
We wanted to know what a few thousand infants could tell us about their language development and best practices for studying language development. Here is what we have found out:
Peekbank is built and maintained by a team of developmental researchers. Meet the team →