In today’s academic world, the pressure to publish and amass citations has created a culture of quantity over quality — the rise in retracted papers and irreproducible results underscores the deep flaws in our current system.

At the heart of these issues are broken incentives for academic success. In order to progress in their careers and attract funding dollars for their research, academics are optimizing for publication-based metrics. However, as Goodhart’s Law states: “When a measure becomes a public target, it ceases to be a good measure.”

That is why, instead of relying on citation-based success metrics, the scientific community needs to embrace more transparent, community-governed reward systems that incentivize healthy research behaviors rather than superficial, incomplete measures of quality.

Decentralized science (DeSci) and blockchain-based scientific publications are key to making this happen. To that end, scientific reputation should be determined by a decentralized community of scientific colleagues rather than reliance on the validation of higher institutions.

The current state of academic publishing

Most researchers today rely on bibliometrics (publication-based metrics) as a proxy for their research impact and consequent career success. This approach is embodied by the impact factor (IF), which measures the importance of a scientific journal by calculating the average number of citations received by its articles over a two-year period.

At first glance, this approach makes sense — assuming that widely cited papers are higher quality. But the sad truth is citation-based metrics are easily and often gamed in a myriad of ways: tit for tat citation agreements, incentivizing a prominent academic to co-author a report for clout, splitting one significant study into multiple publications — as just some examples.

IF (and other similar metrics) cannot then predict the success of individual papers or account for the multifaceted impacts of research. And researchers who want to advance their careers or secure funding are actively incentivized to optimize their outputs based on these metrics. In short, today’s research incentives are a reflection of Goodhart’s law.

The consequences of these broken incentives are widely felt.

Academia is currently in the midst of a “replication crisis,” with studies showing that approximately 50-85% of all biomedical research cannot be independently replicated. Even high-profile studies, such as a landmark 2006 Nature paper on Alzheimer’s, often contain doctored figures, underscoring the deep flaws in our current system. On top of that, traditional funding mechanisms unintentionally incentivize the production of irreproducible research. In 2009, The Lancet published an article estimating that 85% of all research funding ($200 billion annually) is wasted due to an overreliance on bibliometrics.

The solution: Focusing on behavioral targets, not publication metrics

Unless the underlying motivations driving academic publishing themselves are structurally improved, the consequences of this perverse incentive structure will continue to cascade.

Fortunately, Goodhart’s law also reveals that targets for human behavior can in fact be used as a barometer of quality — so long as the ones determining what “quality” means include a wide range of scientific stakeholders rather than ivory tower institutions.

Read more from our opinion section: We need to decentralize science

This is where Web3-enabled scientific communities come in. Blockchain’s decentralized governance mechanisms empower the scientific community to democratically refine incentive structures. The borderless, censorship-resistant flow of value and ideas makes it easier to distribute resources and collaborate both directly and globally.

What should tomorrow’s research incentives look like?

There are many types of DeSci incentive systems. But every successful ecosystem must be evidence-based, governed by the scientific community and intentionally crafted to incentivize healthy research behaviors.

To begin, we must incentivize open access, which ensures that scientific knowledge is freely available to everyone, breaking down barriers to information and fostering a more inclusive research environment. By providing financial incentives and recognition for open access publications, we can encourage more scientists to share their findings with the world.

To promote open data, we should also create targets that reward researchers for making their data publicly available. When researchers share raw data alongside their publications, it allows the greater scientific community to verify results, conduct meta-analyses and build on existing work.

And finally, we should encourage preregistration on open research platforms so scientists and academics actually lay out their hypotheses and methods of their intended study before they even begin. This will help reduce bias, increase transparency and help prevent fraudulent practices such as p-hacking (when non-meaningful connections are highlighted in data anyways). Creating a culture that values and rewards preregistration can improve the integrity of scientific research and increase the reproducibility of results.

Embracing the wisdom of the crowd

To date, academic publishers have monopolized scientific reputation — but the fact is, better alternatives already exist. Instead of a handful of traditional gatekeepers, we need to give the broader scientific community the power to decide which research is impactful and relevant. And the best way to do so in a truly trustless, transparent manner is by leveraging blockchain technology’s decentralized, immutable nature.

The scientific community is beginning to understand the need to adopt, or at least begin experimenting widely with, a new generation of Web3-enabled incentive systems.

Today’s leading DeSci projects are already putting much of this into practice: incentivizing open access, open data and preregistrations through token rewards, and allowing a global community of researchers to determine each others’ reputation scores across various subdomains. These efforts shift researchers away from superficial metrics and towards a transparent, adaptable reputation system that values genuine scientific impact beyond the esoteric realm of citation-based quality measures.