Research

Factors that shape facial variation: More than what meets the eye

Shriver Lab partners with Disney Research for study on human diversity

The team of researchers working on the ADAPT4 study in Shriver Laboratory, left to right, back row: Mark Shriver, Stephen Beckerman and Jacob Vandever; front row: Rafael Osso, Sarita Greer, Yesenia Hill and Laura Perez. Credit: Penn StateCreative Commons

UNIVERSITY PARK, Pa. — Mark D. Shriver, Penn State professor of anthropology, wants to understand the factors that shape facial variation while accurately representing human diversity.

Shriver has long studied facial characteristics, skin pigmentation, voice pitch and other traits related to recent human evolution. He can expand his research even further now thanks to a partnership with Disney Research for the "Anthropometrics, DNA, and the Appearances and Perceptions of Traits 4" (ADAPT4) study, part of Penn State's Center for Human Evolution and Diversity’s ADAPT study series.

Unlike preceding studies, ADAPT4 involves collaboration with Disney Research to improve computer-generated imagery (CGI) and animation. Disney Research is a network of research labs that pursue scientific and technological innovation to advance the Walt Disney Company’s broad media and entertainment efforts.

“Many different groups are interested in variations in the face,” said Shriver, who heads the Shriver Anthropological Genomics Lab in the Department of Anthropology in Penn State’s College of the Liberal Arts. “It might seem like we are all very different from each other, but when you look a little closer you can see that the genes determining our appearance have evolved much faster than any other genes.”

Although genetic makeup largely determines what a person’s face looks like, environmental factors also influence facial appearance, according to Shriver.

"I contacted Disney because I was interested in having faces put onto other faces, and I thought it would be cool if we could switch people’s sex or ancestry group to see how they are perceived differently by others,” Shriver said. “Disney is interested in sampling a younger population to develop better computer graphics, and I realized collaborating with them would help us both accomplish our goals.”

Disney Research agreed, providing a 12-camera system called the Medusa, and a facial expression-reading camera called the Anyma, to collect data for the study. The samples collected will provide Disney with a standardized database of 3D facial videos and still images, voice recordings, full body photos, body mass index (BMI), biological sex, height and iris photos.

Assisting Shriver with the study is project coordinator Sarita Greer, a 2015 Penn State alumna with a degree in biological anthropology. On a monthly or bi-monthly basis, Greer securely sends the collected data to Disney and communicates research progress.

Major potential outcomes from this study include the ability to take perpetrators’ genetic information from crime scenes and recreate their faces, as well as advancing health research by identifying genes that directly correlate with certain birth defects, Greer said.

“The collaboration between Disney Research and Shriver Lab is exciting because it's an innovative and mutually beneficial pairing of evolutionary anthropology and applied science,” said Timothy Ryan, interim head of the Department of Anthropology. “Dr. Shriver's long-term work on the genetics underlying phenotypic variation provides important insight into human evolution, as well as other fields like forensic science.”

Volunteers needed for study

Shriver Lab is recruiting individuals ages 18 and older to participate in the study. Their goal is to sample 1,500 participants by March 2020. To date, 400 individuals have participated.

Study participants will receive copies of their facial scans in several formats and can create a keychain-sized 3D print of their face for free in Penn State’s Maker Commons. An ancestry report, BMI print out, hair form report, and iris photos are the other incentives provided to participants. For the second part of the study, MRI head scans will be performed for 10 percent of the collected sample.

Interested participants must complete the eligibility questionnaire on Shriver Lab’s website, which will provide an eligibility identification number, an opportunity to join the waiting list for future sampling sessions, and the opportunity to schedule a sampling time if none of the sessions work with a participant’s schedule. Any contact or self-reported genetic information is kept confidential.

“The hardest age range for us to reach is ages 25 and older, because those individuals are typically busy professionals,” Greer said. “In the past, we have been able to remotely sample participants, but for this study we have to bring them to us since the cameras are as big as rooms.”

Prior to the sampling session, participants need to be clean-shaven and wear minimal or no makeup. European Union citizens cannot be sampled due to new privacy laws that are restrictive to research. People who have adverse reactions to camera flashes or bright lights should not participate, but accommodations can be made for those with pacemakers.

Open sampling sessions will be held during Penn State football weekends this fall to accommodate large groups of alumni.

Each sampling session takes approximately 90 minutes. Lunch-hour sampling sessions are held from 12:10 to 12:50 p.m. every Tuesday and Wednesday during the summer.

For more information, contact Shriver Lab.

Last Updated July 16, 2019

Contact