Postdoctoral Research Fellow, Cognitive/Computational Neuroscience

Barnard College • New York City, NY

22d • $68,000 - $72,000 • Hybrid

About The Position

The Barnard Visual Cognition Lab in the Department of Psychology at Barnard College is seeking applicants for one postdoctoral research fellow for the 2026–2027 academic year. This is a one-year, full-time position with a possibility of renewal contingent on funding and performance. The fellowship is designed for an emerging scholar who wants deep, hands-on experience working at the intersections of human cognitive science and artificial intelligence to understand the time-course of naturalistic scene understanding. We are especially excited about applicants who enjoy: (1) working with large, real-world datasets, (2) combining visual processing and semantic modeling, and (3) translating theory questions about perception and meaning into concrete, testable analyses.

The Visual Cognition Lab studies how humans perceive, interpret, and navigate real-world scenes, linking visual information, semantic inference, and task demands to behavior and brain activity. Ongoing projects include:

Large-scale naturalistic image and video datasets (including curated “visual experience” style datasets; indoor/outdoor scenes, places, objects, and actions).
Multimodal scene descriptions and embeddings : human and LLM-generated descriptions across multiple task prompts (e.g., affordances, navigation, aesthetics, danger, multisensory inferences), and embedding-based targets (e.g., MPNet/Transformer sentence encoders).
Model–brain alignment using encoding/decoding with EEG time courses and/or fMRI (e.g., ridge regression, variance partitioning, RSA, representational geometry, temporal generalization).
Computational measures of visual information (e.g., image statistics/compressibility proxies, deep network features, object/scene representations).

The postdoctoral fellow will lead and co-lead projects that combine computational modeling, machine learning, and EEG to answer questions about scene understanding and neural representation. The fellow will work closely with the PI, collaborate with students, and contribute to manuscripts, conference submissions, and grant-related research aims. This is a 35-hour/week position with flexible scheduling; on-campus presence is encouraged for mentorship and collaboration, with hybrid arrangements possible depending on project needs. Barnard provides an intellectually vibrant environment with close ties to Columbia University and the broader NYC cognitive science community.

Requirements

• PhD by start date in Psychology, Neuroscience, Cognitive Science, Computer Science, Statistics, or a related field.
• Strong scientific computing skills in Python (NumPy/Pandas, reproducible pipelines).
• Demonstrated ability to run and interpret statistical analyses with appropriate validation (cross-validation, uncertainty, robustness checks).
• Evidence of research productivity (publications/preprints, conference papers, or equivalent).
• Commitment to inclusive mentorship and working respectfully in a diverse academic community.

Nice To Haves

• Experience with machine learning / deep learning (PyTorch; model training; GPU workflows).
• Experience with Transformers / text embeddings / multimodal modeling (e.g., Hugging Face ecosystem).
• Experience with EEG (MNE-Python) and encoding/decoding frameworks.
• Comfort working with large datasets.
• Strong data visualization and figure generation skills for publication.

Responsibilities

• Develop and maintain Python-based pipelines for large-scale data processing (images/video, text descriptions, embeddings, metadata).
• Train and evaluate models for representation learning and prediction (e.g., PyTorch, Transformers, CNN backbones, contrastive/embedding objectives).
• Perform rigorous statistical modeling of behavior and/or neural data.
• Conduct model-to-brain analyses for EEG (e.g., MNE-Python workflows; feature extraction; time-resolved encoding; representational similarity; temporal dynamics).
• Write clean, documented code; use version control (Git); build reproducible experiments.
• Prepare datasets and analysis outputs for publication and sharing (data dictionaries, provenance, basic QA/QC).
• Provide light-to-moderate mentorship to undergraduate/RA contributors (code review, research hygiene, analysis planning).
• Participate in lab meetings, research discussions, and departmental intellectual life.
• Lead/co-lead manuscripts and conference submissions (e.g., VSS/CCN), including figure generation and method writeups.

What This Job Offers

• Job Type: Full-time
• Education Level: Ph.D. or professional degree