An 'impact disparity' is emerging in science — only a few papers earn the largest share of citations. This is comparable to the income disparity in the United States, known as the 1% phenomenon, where 1% of the population earns a disproportionate 17.4% of total income (see go.nature.com/yajthu).

Credit: SOURCE: THOMSON REUTERS SCIENCE CITATION INDEX

The number of citations acquired by a paper is a proxy of its impact. We found that of all papers published in five leading journals in 1990, the most highly cited 1% in each collected around 17% of citations in 2010 (see 'The 1% effect'; data from Thomson Reuters Science Citation Index).

Changes over time in the citation share of the top 1% are evidence of endogenous shifts in underlying processes. These trends are particularly pronounced for citations of older papers. For example, the top 1% of 1990 papers collected only about 5% of citations in 1991.

This shift of attention over time towards the top 1% may reflect the fact that, although the number of research papers has exploded, the time scientists devote to reading them has not. Researchers increasingly rely on crowd sourcing to discover relevant work, a process that favours the leading papers at the expense of the remaining 99%.