The impact of online publishing

"I haven't browsed a table of content in ages; I find all my papers by Pubmed searches anyway". We have probably all heard this remark, which reflects a general trend as how online publishing has changed the way we retrieve scientific publications. In a study published today in Science, Evans ("Electronic Publication and the Narrowing of Science and Scholarship", Evans, 2008) presents data on citations patterns showing that the appearance of electronic publications has been accompanied by a decrease in the number of citations and a progressive restriction of citations to recent papers:

Collectively, the models presented illustrate that as journal archives came online, either through commercial vendors or freely, citation patterns shifted. As deeper backfiles became available, more recent articles were referenced; as more articles became available, fewer were cited and citations became more concentrated within fewer articles.

The interpretation offered is that online availability has driven citations to become more focused while less relevant articles are more easily filtered out. In addition, Evans argues that facile navigation through the network of hyperlinked citations may amplify the tendency to be influenced by other's choice when citing "reference" studies and thus accentuates the dominance of a restricted number of articles:

By enabling scientists to quickly reach and converge with prevailing opinion, electronic journals hasten scientific consensus. But haste may cost more than the subscription to an online archive: Findings and ideas that do not become consensus quickly will be forgotten quickly.

It is probably difficult to be sure that all sources of bias and confounding factors can be eliminated in this type of analysis. For example, on the Friendfeed discussion thread, LJ Jensen asks whether the sheer amount of published research could explain why scientist restrict their citation to the most recent literature. See also some additional discussion in the associated News & Views (Couzin, 2008)

In any case, the study highlights two complementary strategies in information retrieval: finding relevant papers by targeted searches versus staying informed on a broad range of topics by systematic browsing. In our Google-driven era, we may have the tendency to forget the importance of good old-fashioned 'table-of-content-skimming' to stimulate cross-disciplinary thinking, widen our horizon and cultivate scientific curiosity.

Perhaps it is a specificity of printed media to provide "poor indexing" and therefore enforce broad exposure to unrelated areas of research. On the other hand, some web technologies already help to browse through vast amounts of online publications (for example an RSS aggregator helps me to generate a daily literature survey; this can be further combined, for example here at Frienfeed, with other community-centered feeds; other aggregators highlight information by automatic clustering: Postgenomic and Scintilla). However, these tools remain imperfect and, in our reflection on the future of scientific publishing, we will need to find the right balance between the two strategies above and think of how the increasing efficiency of search engines can be complemented by means providing a continuous exposure to diversity.

[From The impact of online publishing]