Gopnik Lab Summer 2024 Internship
Thank you for your interest. The application deadline has now passed, and applications are now under review.
This 8-week internship will take place in Berkeley, CA from Monday, June 3 –Sunday, July 28 . We are willing to accommodate a slightly later start date if you are on the quarter system or if your school’s semester ends after May 31st. Interns will be expected to work roughly 30 hours per week. Weekends will occasionally be required. We welcome applications from both UC Berkeley students and those from outside universities. We encourage applicants to apply for outside funding, but we will be able to provide a travel stipend of $2,400 to those who are not able to secure funding to offset living costs. We are unable to provide housing. Please note that we are unable to sponsor visas for international applicants.
Research interns will work with graduate students on multiple ongoing research projects and with the lab manager on administrative tasks. Interns will be responsible for coding data, recruiting and testing child and/or adult participants, and reading and discussing relevant theoretical and empirical papers. There will be weekly reading groups and lab meetings, as well as a professional development series for those interested in careers in psychology. Research interns should be comfortable working independently and managing their time effectively.
Application Requirements and Criteria
Please only apply if you are willing to make a commitment to work in our lab:
* ~30 hours per week for 8 weeks
* Start date may be flexible to accommodate students whose semesters conclude later than June 1
* Must be an undergraduate student OR recent graduate. We do not accept high school students or graduate students.
* International students may apply, but only if they have authorization to work/study in the US and are enrolled at a US-based institution. We are unable to sponsor visas for interns.
*Interns must be fully vaccinated and able to pass a background check to conduct research at our partner museums
Requirements for
All Applicants
All research in the Gopnik Cognitive Development and Learning Lab is broadly focused on children's development of cause and effect reasoning and how they learn from and about other people. We are looking for dedicated and motivated undergraduate students interested in pursuing a graduate degree in developmental psychology or a related field. Interns will work closely with a graduate student assisting them on all aspects of the research process. Interns will help with experimental and stimuli design; recruiting participants 3 - 14 years old and adults; and collecting, organizing, coding, and analyzing data. Interns will meet regularly with their mentors to discuss the theoretical motivations of the studies they are working on as well as the findings of other empirical papers both related to the studies in the lab and important to the field in general.
About Our Program
Qualifications for Applicants
* Must be excited about Cognitive Development research!
* Organized, self-motivated, independent, and hard working
* Prior experience with children (we welcome both formal, informal, and non-traditional experience with kids!)
* Comfort acting silly around children and professional when speaking to parents
* Prior research experience is not required
* Artistic, mechanical, electrical engineering or programming experience is not necessary, but would be great!
Please note that proof of full vaccination and ability to pass a background check are required for all successful applicants.
Please do not contact us about the status of your application. If you are selected for an interview, you will receive an email.
All applicants will be informed of their status at the end of the application review period.
Tentative Project Descriptions
Perception in Humans and Artificial Intelligence
We humans are remarkable learners and explorers. How do we perceive and explore objects in our everyday lives? How do we choose between information and reward in our exploration? And how do we scaffold our learning to reach a challenging goal that is not immediately attainable? These are questions I attempt to tackle in my research. While I employ computational approaches, my research is fundamentally driven by motivations in cognitive science, so it is okay if you don’t have a computational background! I am recruiting research assistants of all backgrounds to be involved in one or all of these projects, depending on where your interests and skills lie!
Project #1: A Developmental Psychology Inspired Visual Turning Test In a collaboration with Meta and Berkeley AI Research, we propose to design a visual Turning test that is inspired by developmental psychology to probe object reasoning in vision-language models (e.g., CLIP, ALBEF, Flamingo, FLAVA), as there is currently NO comprehensive standard to compare AI performance to that of humans. We examine how children develop various visuo-linguistic capacities (e.g., mental rotation, object categorization, causal reasoning) across age, and design relevant tasks that can help AI communities evaluate their models more robustly with human data as support.
Project #2: To Observe or to Bet? Choices between Information, Reward & Control Childhood experiences are variable, and environmental factors such as predictability and control may have significant impacts on a child’s developmental outcomes and decision- making abilities. In our experiment, we study how much children weigh learning information versus getting reward depending on how predictable the environment is and how much of the environment they can control.
Project #3: Automated Causal Curriculum Learning In a collaboration with DeepMind, we invite children, adults and reinforcement learning agents to play a fun but challenging video game without telling them what the rules of the game are. We examine how they actively explore and design their own curriculum to solve the game (i.e., what causal inferences to make about the rules, which “sub-tasks” to try before attempting the difficult game again). This study allows us to understand better how we come to achieve complex tasks that are difficult to solve right off the bat in our everyday lives. The comparison with RL agents may provide us more insights into the computational strategies humans actually use for their exploration.
Supervisor: Eunice Yiu, Graduate Student
Exploration and Learning in Adults, Children, and Artificial Intelligence
Project Description:
Through a number of projects, I will be studying how children, adults, and AI explore and learn in their environments. In the first project, I am investigating how children explore and exploit within a social context. How does the presence of another person affect exploration? Specifically, through multiple caregiving projects, I ask how does the presence of a caregiver influence risk taking, exploration, and learning? The second project focuses on automated curriculum learning. How do humans and reinforcement learning agents scaffold their learning to achieve a difficult goal they weren't able to achieve at first? We will use video games of varying difficulty to investigate how artificial agents and humans build a curriculum in order to succeed in a goal. An overarching question in line with other projects looks at how different environmental cues (for example safety, controllability, empowerment, volatility, etc) influence exploration and learning. Projects will use multiple methods including behavioral experiments, computational modeling, large language models, cross-cultural experiments, etc. What I am looking for: interns that are passionate about research, artificial intelligence, cognitive science, and cognitive development.Supervisor: Annya Dahmani, Graduate Student
Tissue Box Task
Project Description:
These studies seek to understand the factors that influence children's decisions to explore or exploit. The basic paradigm is a fun and quick game that involves hiding stickers inside of wooden boxes and tracking which boxes children choose to open.
Supervisor: Dr. Dorsa Amir, Post-Doctoral Researcher
Social Causal Difference-Making
Project Description:
We know that individual characteristics influence how individuals are treated and perceived. It’s not a secret; walking down the street, selecting which stranger to sit beside on a chock-full BART train, making allies on the first day at a new school — individual characteristics, past experience, and personal preference all play into the social dynamics of everyday situations.
In this line of research, we are looking into how and when kids are able to determine what characteristics of a social actor make a difference in how they are treated and perceived (i.e. skin color may have stronger social causal implications than what color necklace one is wearing) with the hopes of projecting onto everyday social categories such as race and gender.
Supervisor: Rose Reagan, Laboratory Manager