Four Decades of Open Language Science: The CHILDES Project

The Child Language Data Exchange System (CHILDES), created by Brain MacWhinney and Catherine Snow in 1984, is one of the earliest Open Science and data sharing initiatives in child language development research, and probably in developmental psychology and the behavioral sciences more generally. It is the cornerstone of TalkBank––a repository of transcripts, audio, and video files of natural language samples. Here we highlight how the CHILDES Project served as a trailblazer for the language development research community by being the first initiative to introduce a Big Data approach, encouraging and facilitating crosslinguistic data collection and championing science collaboration through open access to data and analysis tools. We conclude with an outlook on the future of CHILDES and suggestions for where child language development researchers might turn their attention when collecting and donating observational data. Understanding the many paths to language will require expanding CHILDES to increase representation of culturally and neurally diverse populations, finding solutions to the challenge of promoting Open Science practices while safeguarding participant agency and privacy, and leveraging AI tools for automated transcription and data analysis.