The representation of bibliometric datasets and citation analytics are a cornerstone of contemporary research assessment. However, this data can also be represented sonically and listened to. Deploying metrics sonification, Lutz Bornmann and Christian Ganser explore the phenomenon of hyperprolific authors, who publish a research paper every five days, and what listening to their output metrics can tell us about ‘research productivity’.
Visualisations have become an ubiquitous way of representing the findings of empirical research. Bibliometrics, the study of scientific outputs through data such as citations, is no exception, the Leiden Ranking and the Excellence Mapping tools being good examples. However, since the 1990s, researchers have begun to use the sonification of data as a method to complement visualisations. Engaging several human senses and their unique perceptual strengths, proponents of sonification argue it can increase the understanding of underlying data and processes.
Zanella and coauthors define data sonification as: “A technique for representing information and data using non-speech audio”. Sonification can be considered a type of sound design where the sounds produced are tied to the data. One of the most common sonification methods is called parameter mapping. In this case, the sound characteristics (pitch, volume or timbre) are tied to the data”. Overviews of sonification projects have been published by Dubus and Bresin (2013), Zanella and co-authors (2022), and Bornmann (2024). Many examples of data sonification can be found in the Data Sonification Archive.
So, what might the sonification of bibliometric data sound like and can it help build better understandings of research? This is exactly what was set out to find in a forthcoming paper in Scientometrics published by Bornmann and Haegner. Both authors used sonification in an attempt to demonstrate its value as an analytical and presentational tool for bibliometric information. The authors applied sonification to mark a sad occasion: By using bibliometric data from the career of Loet Leydesdorff, a giant in the field of scientometrics, who passed away in 2023. You can listen to the track on SoundCloud:
Here, we present another example of metrics sonification based on the publication output of hyperprolific authors. In a recent article for Nature magazine, Ioannidis, Klavans, and Boyack present the results of a study in which they identified authors in the Scopus database, who published a paper every five days. They define these authors as hyperprolific authors. Ioannidis, Klavans, and Boyack found around 9000 such authors, subsequently emailing 265 of them asking how exactly they had achieved this extraordinary level of productivity.
Comprehending what it means to publish a paper every five days and putting it into context can be difficult to convey in words. This is an instance when hearing this data can create a stronger impression than simply seeing a number or visualization. Our track shows the visualization and sonification of a (fictitious) scientist, who publishes 73 papers in one year (or every five days) and two other scientists who publish 10 and 3 papers in the same period.
The track follows the design of another track published by Buchanan, Huang, and Pearce. These authors visualised and sonified the firepower of weapons used by a gunman who killed 58 people and wounded at least 500 in Las Vegas in 2017.
The sonification includes visualized text and sound elements. Text at the beginning explains why the publication data has been sonified and text in-between serves as section guide for the viewer/listener. Three sounds have been used for the sonification: (1) A bass loop in the background across the whole track is intended to evoke the association: The viewer receives an insight in the “machine room of science” (with the presented publication process of three scientists). (2) A synthesizer sound in the track signals that it follows other visualized and sonified data than before. The bass and synthesizer loops change their pitch across the track: With increasing numbers of papers per year and scientist, the pitch increases. (3) The most important element in the track is a swoosh sound. Here, we wanted to convey the sound of the process of sending a manuscript from its author to the interested reader. The swoosh sound changes during the track: If only a few papers are visualized over the months in the year, the swoosh is clearly audible. However, this sound becomes distorted and lost for the scientist with 72 papers in one year associating the rapid (and possibly unhealthy) publication of so many papers.
Publication output is probably the strongest predictor of academic success. The publications of scientists follow a lognormal productivity distribution; the law of scientific productivity means that “the number of scientists producing N papers is proportional to 1/N2”. Since only 1% of scientists are able to publish one paper per year, only a small fraction of scientists is able to publish constantly. This core set of scientists is responsible for around 40% of all papers published and around 90% of the highly cited papers.
Throughout the twentieth century, scientists have consistently published on average around two papers annually. Over the past 15 years, this number has slightly increased, with the “average” scientist authoring or co-authoring approximately 2.5 papers per year. This rise in individual productivity can be attributed to increased collaboration, as scientists become co-authors on a larger number of papers. The reason for the great differences between very productive (73 papers) and “average” (3 papers) scientists probably lies in the effect of the Anna Karenina principle: Scientists have to manage many factors that are interrelated to be productive (high motivation, creativity, reputable professional position, early significant papers in reputable journals etc.). Failure occurs if any factor is absent, with the nature of the failure specific to the combination of existing and missing factors.
Our track therefore compares and contrasts a fictitious hyperprolific and an average author. We should have in mind that the number for the “average” author is probably an over-estimation of the real “average”, as calculations based on Scopus or the Web of Science database only consider publishing scientists and non-publishing scientists do not appear in the databases.
Careful readers and listeners will have noticed the intermediate author of ten papers per year in our track. Here again is the output of Loet Leydesdorff. Our search in the core collection of the Web of Science database shows that as of January 2025, Loet Leydesdorff published 428 papers between 1980 and 2023. Numbers can make clear that there are great differences between the three scientists, the combination of visualization and sonification can underline these differences and illustrate the differences more clearly. Loet Leydesdorff’s output provides an important note of individual difference, great scientists are not always prolific authors. However, the ineffable impact of this body of work can perhaps never truly be captured in this way.
The content generated on this blog is for information purposes only. This Article gives the views and opinions of the authors and does not reflect the views and opinions of the Impact of Social Science blog (the blog), nor of the London School of Economics and Political Science. Please review our comments policy if you have any concerns on posting a comment below.
Image Credit: guys_who_shoot on Shutterstock.
