A metric of knowledge as information compression reflects reproducibility predictions for biomedical experiments

Forecasting the reproducibility of research findings is one of the key challenges of metascience. Above-chance predictions have mainly been achieved by pooling the subjective ratings of experts, and how these predictions are formed remains to be understood. Here, we show that reproducibility forecas...

Full description

Saved in:
Bibliographic Details
Main Authors: Daniele Fanelli, Pedro Batista Tan, Olavo B. Amaral, Kleber Neves
Format: Article
Language:English
Published: The Royal Society 2025-07-01
Series:Royal Society Open Science
Subjects:
Online Access:https://royalsocietypublishing.org/doi/10.1098/rsos.241446
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849246257515593728
author Daniele Fanelli
Pedro Batista Tan
Olavo B. Amaral
Kleber Neves
author_facet Daniele Fanelli
Pedro Batista Tan
Olavo B. Amaral
Kleber Neves
author_sort Daniele Fanelli
collection DOAJ
description Forecasting the reproducibility of research findings is one of the key challenges of metascience. Above-chance predictions have mainly been achieved by pooling the subjective ratings of experts, and how these predictions are formed remains to be understood. Here, we show that reproducibility forecasts made for the Brazilian Reproducibility Initiative (BRI), a large-scale replication of experiments in the life sciences, are significantly correlated with K, a principled metric of knowledge as information compression. For each study in the BRI sample, we calculated K by dividing the effect size, measured in bits of Shannon entropy, by the descriptive length (a proxy of the complexity) of the study’s methodology, calculated as the optimal Shannon encoding of a conceptual graph representing the replication protocol. We found that experts’ predictions about reproducibility were statistically associated with K values and with the complexity of protocols. This relation was robust to controlling for study methodology and other possible confounding factors. These results suggest that expert raters partially judge the reproducibility of findings by assessing the ratio between the information yielded and the information required by a study, and they support the hypothesis that scientific knowledge may be understood and studied through the lenses of information compression.
format Article
id doaj-art-48e920dc40d44121ae85a2245a354b3f
institution Kabale University
issn 2054-5703
language English
publishDate 2025-07-01
publisher The Royal Society
record_format Article
series Royal Society Open Science
spelling doaj-art-48e920dc40d44121ae85a2245a354b3f2025-08-20T03:58:35ZengThe Royal SocietyRoyal Society Open Science2054-57032025-07-0112710.1098/rsos.241446A metric of knowledge as information compression reflects reproducibility predictions for biomedical experimentsDaniele Fanelli0Pedro Batista Tan1Olavo B. Amaral2Kleber Neves3Theoretical and Empirical METaknowledge (TEMET) lab, School of Social Sciences, Heriot-Watt University, Edinburgh, UKVrije Universiteit Amsterdam, Amsterdam, The NetherlandsFederal University of Rio de Janeiro, Rio de Janeiro, BrazilFederal University of Rio de Janeiro, Rio de Janeiro, BrazilForecasting the reproducibility of research findings is one of the key challenges of metascience. Above-chance predictions have mainly been achieved by pooling the subjective ratings of experts, and how these predictions are formed remains to be understood. Here, we show that reproducibility forecasts made for the Brazilian Reproducibility Initiative (BRI), a large-scale replication of experiments in the life sciences, are significantly correlated with K, a principled metric of knowledge as information compression. For each study in the BRI sample, we calculated K by dividing the effect size, measured in bits of Shannon entropy, by the descriptive length (a proxy of the complexity) of the study’s methodology, calculated as the optimal Shannon encoding of a conceptual graph representing the replication protocol. We found that experts’ predictions about reproducibility were statistically associated with K values and with the complexity of protocols. This relation was robust to controlling for study methodology and other possible confounding factors. These results suggest that expert raters partially judge the reproducibility of findings by assessing the ratio between the information yielded and the information required by a study, and they support the hypothesis that scientific knowledge may be understood and studied through the lenses of information compression.https://royalsocietypublishing.org/doi/10.1098/rsos.241446metasciencemetaresearchreproducibilitycomplexityinformation compressionphilosophy of science
spellingShingle Daniele Fanelli
Pedro Batista Tan
Olavo B. Amaral
Kleber Neves
A metric of knowledge as information compression reflects reproducibility predictions for biomedical experiments
Royal Society Open Science
metascience
metaresearch
reproducibility
complexity
information compression
philosophy of science
title A metric of knowledge as information compression reflects reproducibility predictions for biomedical experiments
title_full A metric of knowledge as information compression reflects reproducibility predictions for biomedical experiments
title_fullStr A metric of knowledge as information compression reflects reproducibility predictions for biomedical experiments
title_full_unstemmed A metric of knowledge as information compression reflects reproducibility predictions for biomedical experiments
title_short A metric of knowledge as information compression reflects reproducibility predictions for biomedical experiments
title_sort metric of knowledge as information compression reflects reproducibility predictions for biomedical experiments
topic metascience
metaresearch
reproducibility
complexity
information compression
philosophy of science
url https://royalsocietypublishing.org/doi/10.1098/rsos.241446
work_keys_str_mv AT danielefanelli ametricofknowledgeasinformationcompressionreflectsreproducibilitypredictionsforbiomedicalexperiments
AT pedrobatistatan ametricofknowledgeasinformationcompressionreflectsreproducibilitypredictionsforbiomedicalexperiments
AT olavobamaral ametricofknowledgeasinformationcompressionreflectsreproducibilitypredictionsforbiomedicalexperiments
AT kleberneves ametricofknowledgeasinformationcompressionreflectsreproducibilitypredictionsforbiomedicalexperiments
AT danielefanelli metricofknowledgeasinformationcompressionreflectsreproducibilitypredictionsforbiomedicalexperiments
AT pedrobatistatan metricofknowledgeasinformationcompressionreflectsreproducibilitypredictionsforbiomedicalexperiments
AT olavobamaral metricofknowledgeasinformationcompressionreflectsreproducibilitypredictionsforbiomedicalexperiments
AT kleberneves metricofknowledgeasinformationcompressionreflectsreproducibilitypredictionsforbiomedicalexperiments