Novelty, Disruption, and the Evolution of Scientific Impact

Since the 1950s, citation impact has been the dominant metric by which science is quantitatively evaluated. But research contributions play distinct roles in the unfolding drama of scientific debate, agreement and advance, and institutions may value different kinds of advances. Computational power, access to citation data and an array of modeling techniques have given rise to a widening portfolio of metrics to extract different signals regarding their contribution. Here we unpack the complex, temporally evolving relationship between citation impact alongside novelty and disruption, two emerging measures that capture the degree to which science not only influences, but transforms later work. Novelty captures how research draws upon unusual combinations of prior work. Disruption captures how research comes to eclipse the prior work on which it builds, becoming recognized as a new scientific direction. We demonstrate that: 1) novel papers disrupt existing theories and expand the scientific frontier; 2) novel papers are more likely to become "sleeping beauties" and accumulate citation impact over the long run; 3) novelty can be reformulated as distance in journal embedding spaces to map the moving frontier of science. The evolution of embedding spaces over time reveals how yesterday's novelty forms today's scientific conventions, which condition the novelty--and surprise--of tomorrow's breakthroughs.