Timothy Behrens
Timothy Behrens trained in information engineering at Oxford, where he earned a doctorate in June 2004. He held subsequent research posts in computational biology and neuroscience and became a professor in July 2013. In a manuscript first submitted in 2021, he co-authored work relating transformers to models and neural representations of the hippocampal formation. The authors reported that transformers with recurrent position encodings could reproduce place- and grid-like representations. This work links his neuroscience research with artificial neural network models; the results are limited to the reported experiments.
- Now at
- University of Oxford and Wellcome Trust
- Works
- 347
Publication numbers from OpenAlex, as of 14 October 2026. Citations and h-index are hidden because this OpenAlex record may be incomplete or mix in other authors.
Timeline
-
2004
Information engineering at Oxford
Behrens studied information engineering at Oxford from 2000 to early 2004 and earned his doctorate in June 2004. He then worked as a research assistant at the FMRIB Centre.
-
2005–2013
Computational biology and neuroscience
He held an MRC fellowship in computational biology from 2005 to 2009. He also held research lecturer roles and became professor of Computational Neuroscience in July 2013.
-
2021–2022
Transformers and hippocampal models
With James Whittington and Joseph Warren, Behrens co-authored a paper relating transformers to models of the hippocampal formation. It was first submitted in December 2021 and revised in March 2022. The work connects an artificial neural network architecture with a neuroscience model.
-
2021–2022
Spatial representations in networks
The authors reported that transformers equipped with recurrent position encodings reproduced place- and grid-like spatial representations. They also compared performance with the neuroscience model.
Sources and credits
Photo: www.ellis.ox.ac.uk; creator not stated. License: Not stated by source. Source: ellis.ox.ac.uk. Cropped to a square and resized. Ask us to remove or replace this photo.