Research

Data-sharing video library aids developmental studies

Two pre-adolescent girls engaged in social and gross motor free play on a public playground. The girls were instructed to show the camera operators the various ways in which they play for developmental research purposes. Credit: K. Adolph, DatabraryAll Rights Reserved.

UNIVERSITY PARK, PA -- The first large-scale, open data-sharing video library is expanding at a rapid pace, providing developmental researchers at Penn State and across the world unprecedented access to data in a rich, new way.

Called Databrary, the Web-based video-data library has grown to include data from more than 270 investigators from 166 institutions. The Databrary team, lead by Rick Gilmore, associate professor of psychology and faculty researcher with the Huck Institutes of the Life Sciences at Penn State, and Karen Adolph, professor of psychology and neural science at New York University, are excited by the data sharing systems’ growth in the just over two years since it was established.

According to Gilmore, there are many challenges in sharing data, especially in the developmental sciences. “Developmental science data can be hard to find and access as it may lack citable and searchable identifiers, and there is no consensus about who owns or should control research data. At the same time there are other significant factors, including technical, ethical, transparency, cultural and conceptual issues associated with the use of data and available datasets. When data is published, access is often restricted.”

Databrary addresses many of these challenges and allows researchers access to an extensive library of video data. According to Gilmore, Databrary focuses on video because it captures the complexity and richness of behavior unlike any other data collection tool and provides unique opportunities for reuse. “Video data can give researchers who study behavior in the laboratory, home, or classroom a uniquely valuable source of information.”

Databrary also enables researchers to curate data as they collect it, openly sharing video data and related information about the studies while they occur. "Researchers and graduate students can build on this work, test competing hypotheses and reuse data in ways unimagined by the original researcher," Gilmore explained. “It also allows access to bigger data so we can learn more about the fascinating processes we all share,” said Gilmore.

To further address how address how Databrary can help researchers, Databrary and the Penn State Child Study Center are sponsoring a workshop on March 25 at 10:30 a.m. in 127 Moore Building on the University Park campus. It is the first in a series of workshops to be offered at various locations across the country.

Users will be shown how to tag their studies based on subjects, how to use tasks and displays, and how to utilize study participant data, such as ethnicity, gender and age, to build upon findings. In addition to Databrary, the workshop will also review Datavyu, a free, open-source software tool that researchers can use to score, explore, and analyze video recordings.

Because videos contain faces and voices, Databrary limits full access to authorized researchers who have signed a written agreement to keep the identities of people depicted in stored recordings confidential. People depicted in recordings must give written permission for their information to be shared.

Databrary is sponsored by the National Science Foundation and the National Institutes of Health and is housed at NYU. The project is a part of big data science initiatives by Gilmore and other researchers currently underway at Penn State. Gilmore questions what makes data ‘big’ and what implications the size and complexity of datasets have for developmental sciences in an article recently published in Wiley Interdisciplinary Reviews: Cognitive Science.

“In developmental sciences, discoveries have the potential to improve health and maximize human achievement,” said Gilmore. “Big data can be invaluable to developmental researchers in advancing new ways to collect, manage, store, share and enable others to reuse data. In the future, I’d like to see all data accessible in some way, and allow other researchers decide how valuable data sets are by building upon them.”

Gilmore’s work is being supported by the National Science Foundation, the Eunice Kennedy Shriver National Institute of Child Health and Human Development, the Society for Research in Child Development and the Social Science Research Institute at Penn State.

Last Updated March 3, 2016

Contact