Science and Technology Advance through Surprise
This work addresses how scientific and technological progress can be predicted and evaluated, offering insights for improving institutions like education and peer review, though it is incremental in applying existing models to new data.
The study analyzed tens of millions of research papers and patents to predict breakthrough discoveries, finding that unexpected combinations of content and context, such as solving problems from one field with methods from another, predict up to 50% of the likelihood of high citations and awards, with models achieving 95% AUC.
Breakthrough discoveries and inventions involve unexpected combinations of contents including problems, methods, and natural entities, and also diverse contexts such as journals, subfields, and conferences. Drawing on data from tens of millions of research papers, patents, and researchers, we construct models that predict next year's content and context combinations with an AUC of 95% based on embeddings constructed from high-dimensional stochastic block models, where the improbability of new combinations itself predicts up to 50% of the likelihood that they will gain outsized citations and major awards. Most of these breakthroughs occur when problems in one field are unexpectedly solved by researchers from a distant other. These findings demonstrate the critical role of surprise in advance, and enable evaluation of scientific institutions ranging from education and peer review to awards in supporting it.