Artificial Intelligence (AI) plays a special role in cognitive science. Here at Johns Hopkins we concentrate on two areas with direct relevance for current AI systems: vision and language. Researchers in our department are interested not only in human-level performance on tasks, but also insight into the particular ways that humans do those tasks.
We pursue these questions at multiple levels:
- Abstract task characterization
- Mechanistic cognitive processes
- Brain representations e.g. as observed with neuroimaging
This work continues a tradition that includes:
- Newell and Simon Human Problem Solving 1972
- Rumelhart, McClelland and the PDP Research Group. Parallel Distributed Processing: explorations in the microstructure of cognition 1986
- Smolensky and Legendre The Harmonic Mind 2006
Faculty with AI-related interests
This is a non-exhaustive list of AI topics that faculty work on in the department, alongside their other research interests. (Note: DSAI refers to the Data Science and AI Institute.)
- Mick Bonner: deep neural network models of human vision, high-dimensional statistics, unsupervised learning algorithms. DSAI member
- John Hale: language models, parsers. DSAI member
- Jennifer Hu: language models, language understanding, interpretability, evaluation. DSAI member
- Leyla Isik: vision models, video understanding, multimodal models, inductive biases. DSAI member
- Annemarie Kocab: event recognition in video, linguistic sign classification
- Kyle Rawlins: language understanding, computational semantics, language models. DSAI member
- Margaret Renwick: forced alignment, classifiers for speech sounds, speaker diarization
- Colin Wilson: XAI (interpretable/explainable AI); multisensory language; pattern discovery
- Alan Yuille: vision-language-action models, interactive virtual worlds and analysis by synthesis, Bayesian models of cognition. DSAI member
The department offers a Computational Cognitive Science track within its PhD program.