Decreasing readability in scientific papers over time

“Reporting science clearly and accurately is a fundamental part of the scientific process, facilitating both the dissemination of knowledge and reproducibility of results.” In this way, Plavén-Sigray et al. introduce us to their preprint in which they analyzed readability in over 700,000 abstracts of scientific articles published in 122 biomedical journals from 1881-2015. Using two different metrics that account for complexity of words and sentences, they found that readability of scientific papers is decreasing steadily over time. This decline is positively correlated with complexity of the rest of the manuscript, showing this trend exceeds abstracts and reflects the whole texts. Moreover, complexity over time is also associated with increasing number of co-authors (although that doesn’t fully explain the trend) and use of in-group science-specific words (jargon). Decreasing readability is a problem when science needs to be explained to non-specialists (digested facts and scientific results can be misinterpreted). This problem even affects specialists, as previously the number of citations has been negatively associated with increasing complexity of language use in papers. So, are we making science less accessible even for ourselves, scientists? (Summary by Gaby Auge) bioRxiv 10.1101/119370

UPDATE Sept 5 2017: This paper has now been published in eLIFE https://elifesciences.org/articles/27725