2025

Video Call improves Japanese English learners’ speaking skills
Video Call improves Japanese English learners’ speaking skills

Audrey Kittredge, Crystal Lee, Xiangying Jiang

Duolingo Research Report 2025

This is the first study to demonstrate the actual learning value of the AI-enabled Video Call with Lily feature: where Duolingo users can practice speaking with Lily (powered by Open AI GPT). Using a randomized control trial, we found that Duolingo users who engaged with the Video Call with Lily feature at least 2x a day for 30 days improved their speaking abilities significantly more compared to users who did not use this feature.

Video Call improves Japanese English learners’ speaking skills

Audrey Kittredge, Crystal Lee, Xiangying Jiang

Duolingo Research Report 2025

This is the first study to demonstrate the actual learning value of the AI-enabled Video Call with Lily feature: where Duolingo users can practice speaking with Lily (powered by Open AI GPT). Using a randomized control trial, we found that Duolingo users who engaged with the Video Call with Lily feature at least 2x a day for 30 days improved their speaking abilities significantly more compared to users who did not use this feature.

2023

The dynamic functions of social cues during children's word learning
The dynamic functions of social cues during children's word learning

Crystal Lee, Casey Lew-Williams

Infant and Child Development 2022

Photo by Pineapple Supply Co. on Unsplash. Please put a tldr (too-long-didnt-read, 1~2 sentences) of your publication here. It is not recommended to put the actual abstract here because it is usually too long to fit in. $\LaTeX$ is supported. $a=b+c$.

The dynamic functions of social cues during children's word learning

Crystal Lee, Casey Lew-Williams

Infant and Child Development 2022

Photo by Pineapple Supply Co. on Unsplash. Please put a tldr (too-long-didnt-read, 1~2 sentences) of your publication here. It is not recommended to put the actual abstract here because it is usually too long to fit in. $\LaTeX$ is supported. $a=b+c$.

How executive functioning, sentence processing, and vocabulary are related at 3 years of age
How executive functioning, sentence processing, and vocabulary are related at 3 years of age

Crystal Lee, Andrew Jessop, Amy Bidgood, Michelle S. Peter, Julian M. Pine, Caroline F. Rowland, Samantha Durrant

Journal of Experimental Psychology 2022

Photo by Pineapple Supply Co. on Unsplash. Please put a tldr (too-long-didnt-read, 1~2 sentences) of your publication here. It is not recommended to put the actual abstract here because it is usually too long to fit in. $\LaTeX$ is supported. $a=b+c$.

How executive functioning, sentence processing, and vocabulary are related at 3 years of age

Crystal Lee, Andrew Jessop, Amy Bidgood, Michelle S. Peter, Julian M. Pine, Caroline F. Rowland, Samantha Durrant

Journal of Experimental Psychology 2022

Photo by Pineapple Supply Co. on Unsplash. Please put a tldr (too-long-didnt-read, 1~2 sentences) of your publication here. It is not recommended to put the actual abstract here because it is usually too long to fit in. $\LaTeX$ is supported. $a=b+c$.

2022

Speech and social cues combine at discourse boundaries to promote word learning
Speech and social cues combine at discourse boundaries to promote word learning

Crystal Lee, Casey Lew-Williams

Cognitive Development 2022 Spotlight

Photo by Pineapple Supply Co. on Unsplash. Please put a tldr (too-long-didnt-read, 1~2 sentences) of your publication here. It is not recommended to put the actual abstract here because it is usually too long to fit in. $\LaTeX$ is supported. $a=b+c$.

Speech and social cues combine at discourse boundaries to promote word learning

Crystal Lee, Casey Lew-Williams

Cognitive Development 2022 Spotlight

Photo by Pineapple Supply Co. on Unsplash. Please put a tldr (too-long-didnt-read, 1~2 sentences) of your publication here. It is not recommended to put the actual abstract here because it is usually too long to fit in. $\LaTeX$ is supported. $a=b+c$.

Accessibility factors that lead to good-enough language production
Accessibility factors that lead to good-enough language production

Crystal Lee, Casey Lew-Williams, Adele Goldberg

Proceedings of the Annual Meeting of the Cognitive Science Society 2022

In a series of experiments, I determined three variables the influence words we say: word frequency, priming, and time pressure. By identifying these factors, I demonstrate that speech production is fundamentally a "good-enough" process, where what humans say is pushed and pulled by multiple dimensions.

Accessibility factors that lead to good-enough language production

Crystal Lee, Casey Lew-Williams, Adele Goldberg

Proceedings of the Annual Meeting of the Cognitive Science Society 2022

In a series of experiments, I determined three variables the influence words we say: word frequency, priming, and time pressure. By identifying these factors, I demonstrate that speech production is fundamentally a "good-enough" process, where what humans say is pushed and pulled by multiple dimensions.

2021

Action anticipation based on an agent's epistemic state in toddlers and adults
Action anticipation based on an agent's epistemic state in toddlers and adults

ManyBabies, Crystal Lee

Child development 2021

Cover image is a photo by Thomas Renaud on Unsplash. The abstract of the publication is meant to be a TLDR (very brief summary with 1~2 sentences) of your paper.

Action anticipation based on an agent's epistemic state in toddlers and adults

ManyBabies, Crystal Lee

Child development 2021

Cover image is a photo by Thomas Renaud on Unsplash. The abstract of the publication is meant to be a TLDR (very brief summary with 1~2 sentences) of your paper.

Learning maximum absolute meaning through reasoning about speaker intentions
Learning maximum absolute meaning through reasoning about speaker intentions

Crystal Lee, Chigusa Kurumada

Language Learning 2021

Cover image is a photo by Thomas Renaud on Unsplash. The abstract of the publication is meant to be a TLDR (very brief summary with 1~2 sentences) of your paper.

Learning maximum absolute meaning through reasoning about speaker intentions

Crystal Lee, Chigusa Kurumada

Language Learning 2021

Cover image is a photo by Thomas Renaud on Unsplash. The abstract of the publication is meant to be a TLDR (very brief summary with 1~2 sentences) of your paper.

2019

Augmenting communication between hearing parents and deaf children
Augmenting communication between hearing parents and deaf children

Ashley Tenesaca, Jung Yun Oh, Crystal Lee, Wanyin Hu, Zhen Bai

2019 IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) 2019

Over 90% of deaf children are born to hearing parents who are not fluence in sign language. This mismatch can lead to delayed exposure, and ultimately childrens' delayed language acquisition. In this team, I acted as a SME expert in the development of a visual augmentation prototype that allows hearing parents to just-in-time ASL to their deaf children.

Augmenting communication between hearing parents and deaf children

Ashley Tenesaca, Jung Yun Oh, Crystal Lee, Wanyin Hu, Zhen Bai

2019 IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) 2019

Over 90% of deaf children are born to hearing parents who are not fluence in sign language. This mismatch can lead to delayed exposure, and ultimately childrens' delayed language acquisition. In this team, I acted as a SME expert in the development of a visual augmentation prototype that allows hearing parents to just-in-time ASL to their deaf children.