The Matilda effect \ The stories outside the frame
We like to think of science as a mirror—an objective, edge-to-edge reflection of the world as it truly is. But mirrors are built by human hands, and the frames we put around them determine what stays in the reflection and what falls into the margins.
For decades, we’ve treated the gaps in that reflection as isolated and tragic accidents. A few missing names. A few “historical oversights.”
The Matilda Effect.
Sounds a bit like a secret code, doesn’t it? The kind of term exchanged in small circles by people who know something the rest of the world missed. Which is slightly ironic, considering the pattern behind it is neither small nor particularly hidden. Somehow, even after naming it, we still manage to look right past it.
So what is the Matilda effect? What is it hiding?
If you want to find out how deep this rabbit hole really goes, you’ll find some answers ahead. But fair warning: once the frame becomes visible, it becomes difficult to look at the reflection the same way again. Once you see it, you can’t unsee it.
If you’ve already heard of the Matilda Effect, that’s a good sign. If you haven’t, nothing to be ashamed of. Truly. That’s kind of the point.
Let’s start with a definition. And we’re checking where everyone used to go before AI took over: Wikipedia. According to it, “the Matilda effect is a bias against acknowledging the achievements of women scientists and inventors, whose work is consequently attributed to their male colleagues.” But we’re a science communication magazine, we can’t just stop at Wikipedia, can we? Let’s dig a bit deeper.
According to Arslan et al., “The “Matilda Effect” is a generic term for bias against contributions from female scientists with their achievements credited to male colleagues.”1 The term was introduced in 1993 by historian of science Margaret W. Rossiter, but Rossiter wasn’t inventing the phenomenon. She was naming a pattern that had been repeating itself for centuries.
As she herself noted, “not only have those unrecognized in their own time generally remained so, but others that were well-known in their day have since been obliterated from history, either by laziness or inertia.”2
The term builds on the Matthew effect, a term coined by sociologist Robert Merton used to describe how recognition in science tends to accumulate to those who already have it. Scientists with a reputation tend to receive more credit, more visibility, and more citations than those without it.
Or, put differently: the rich get richer, scientifically speaking.
In theory, this dynamic has nothing to do with gender. It’s about reputation.
But Rossiter noticed something curious while studying the history of science: women scientists seemed to fall on the losing side of this equation with striking regularity. Her conclusion was simple. The Matthew effect had a gendered twin.
And so the Matilda effect term was born. But... who was Matilda?
The inspiration behind the name was Matilda Joslyn Gage, a 19th-century suffragist, abolitionist, and writer who had already noticed something strange about the way society treated women’s ideas.
In 1870, Gage published an essay titled “Woman as an Inventor”. She argued that women’s intellectual contributions were often ignored, dismissed, or quietly absorbed into male reputations. Social norms made it difficult for them to claim recognition.
“If women have ideas they are taught to repress them as improper for their sex, and the genius which does them and their sex honor is deemed a matter to be hidden from light”, Gage wrote.3
Gage suspected that history had already erased many inventions and discoveries made by women. “We can well imagine of how much inventive power and reputation women have been robbed in ages past.”3 And even when women did innovate, she argued, social constraints often forced them to work quietly or indirectly: “Women have not dared to exercise their faculties except in certain directions unless in a covert manner.”3
More than a century later, Rossiter realized that Gage had essentially described a pattern that science was still reproducing.
Rosalind Franklin produced the X-ray diffraction image—Photo 51—that was key to identifying the DNA double helix. Her data were shared without her direct involvement, and while the Nobel Prize was later awarded to Watson, Crick, and Wilkins, her contribution remained largely unrecognized for many years.
_
Lise Meitner helped explain nuclear fission, but the 1944 Nobel Prize in Chemistry was awarded only to her collaborator Otto Hahn.
_
Jocelyn Bell Burnell first observed the signals that led to the discovery of pulsars during her PhD. The Nobel Prize later recognized her supervisor Antony Hewish and colleague Martin Ryle.
_
Some of these stories, like the ones with the images above, are now widely known. They’ve become emblematic, almost comforting in a way, as if we can point at them and say: “Look, we fixed it.” We know better now. But they also create a slightly comforting illusion: the idea that these were isolated historical injustices. A few unfortunate cases from a different era, when science was dominated by old boys’ clubs. Because, surely that was a problem of the past... Right?
Well, no.
A large analysis of research teams found that “women in research teams are significantly less likely than men to be credited with authorship” and that “gender differences are well-documented: women both publish and patent less than men”.4
Not dramatically less. Just consistently less. Quietly. Statistically. The kind of difference that accumulates over time—shaping careers, funding opportunities, promotions, and eventually the stories we tell about discoveries. Because authorship isn’t just a line on a paper. It’s a ticket into the historical record.
The famous examples we discuss today were recovered because historians found letters, notebooks, and archives. Someone did the detective work. And that raises some slightly more uncomfortable questions.
How many others were never documented?
How many contributions were absorbed into someone else’s reputation?
How many women never even entered the lab because they lacked references, mentors, representation?
And, as in every good movie, the plot thickens. Because even the Matilda conversation can be a bit reductive.
At first glance, it seems straightforward: women’s contributions are overlooked or reassigned. A problem of credit. Of visibility.
But stay with me for a moment. Because the Matilda effect doesn’t just describe a phenomenon, it can also work as a lens. A way of noticing patterns that don’t always come with a name.
So let’s start with what we know. Conversations about inequality in science have long focused on the gap between women and men.
Important? Absolutely.Complete? Not quite.
As researchers point out, “the lack of intersectional approaches has led to focusing on gender disparities while ignoring or underplaying the importance of other systems of oppression based on, for example, sexuality, gender identities beyond the gender binary, race/ethnicity, disability, and social class.”5
Women of color, for instance, often experience what scholars describe as a “double bind” of racism and sexism. Surveys in the physical sciences show that women, non-binary, and gender-nonconforming researchers report higher levels of discomfort due to “the interlocking oppression of heteronormativity, gender-based stereotypes, and sexism.”5
Now let’s zoom out a little further. At this point, someone might wonder, why does this matter? I don’t see how this can affect science. After all, science is objective. Right?
Well... yes, at the level of methods and principles, it strives to be.
But the production of scientific knowledge depends on human decisions: what questions are worth asking, which projects receive funding, which results are published, and whose expertise is taken seriously?
In other words: who participates in science can shape what science studies.
Still feels a bit abstract? Let’s look at where this shows up in practice. In medicine, for decades, clinical research relied predominantly on male subjects, implicitly treating the male body as the default. The result? Gaps in understanding how diseases affect women, from symptoms to treatment responses.
In medicine, what counts as typical has often meant male. What we recognize as a classic heart attack—crushing chest pain, numb arm—was largely defined using male patients. While women can present the same symptoms, they often also show nausea, fatigue, or jaw pain. By labeling these as atypical, the system creates a structural bias that leaves women significantly more likely to be misdiagnosed.6,7
This was reinforced by decades of underrepresentation of women in clinical research, only beginning to shift in the early 1990s. Let that sink in. Until the late ‘90s, women were not part of clinical research.
Pain pathways were mapped largely in male rodents using microglial cells. Yet, we now know females often rely on different immune mechanisms, specifically T-cell pathways. These pathways remained “invisible” for years, and by the time this was understood, many treatments had already been built on the male model, helping to explain why some drugs are less effective or carry higher risks of side effects for women.8,9
Technologies can carry similar blind spots: pulse oximeters, widely used during COVID-19, were shown to be less accurate in patients with darker skin due to how light interacts with melanin.10 Early facial recognition systems showed higher error rates for women and people with darker skin because the datasets used to train them were mostly composed of white male faces.
Not because the science was flawed in principle. But because the system deciding what to study was incomplete.
And beyond Western science, entire knowledge systems have been dismissed for centuries. Indigenous communities developed sophisticated ecological and environmental knowledge long before these ideas were formalized in academic research—yet they were often labelled anecdotal or unscientific.
So perhaps the question is not whether science is objective. It’s whether the systems that organize science are quite as neutral as we like to imagine.
(And if you’re curious about this, there happens to be an interview in the next pages covering exactly this. Just saying.)
’
Let’s take a little breathing pause here and do a quick mood check. Feeling overwhelmed? A bit hopeless? I did while writing this. And again while reading it back.
So let’s bring in some good news, because is exists.
The International Day of Women and Girls in Science exists to challenge a very specific image of who gets to be seen as a scientist. March 8 exists as a reminder that access to education, careers, or voting rights was never simply a given. Symbolic days matter. Mentorship programs, diversity initiatives, and institutional reforms are slowly shifting the landscape. These are gradually making it easier for more people to enter and remain in scientific spaces.
We have come a long way. Period.
There’s just some follow-up “buts...”
We don’t step outside the system just because we’re aware of it. We still carry biases. We still miss things. I did while writing this. You probably did while reading it.
So, is there a solution? What’s the conclusion here? Well, I’m no oracle. It would be great if I could give you some clear guidelines, a neat conclusion to this article that now feels a bit everywhere and nowhere. There isn’t one, I’m afraid. This topic is too layered. So instead of forcing one, let’s go back to the starting point: the Matilda Effect.
Now you know what it is. That’s already a step. There are important initiatives, like The Matilda Project, doing the work of recovering names, correcting records, giving credit where it’s long overdue.
Here, though, the Matilda Effect works as a lens. A way of noticing how certain contributions, perspectives, and knowledge systems become easier to ignore than others. Once you put on the ‘Matilda glasses,’ it’s hard to unsee the patterns. You might start by asking yourself:
Whose voices do I amplify?
Which narratives do I reach for first?
Which stories have I never thought to question?
And finally:
Which Matildas am I missing right now?
More matildas
There isn’t nearly enough space here to do justice to all the names that should be part of this story. These are only a few names recovered from the margins of the mirror, a small glimpse into the many women whose contributions were overlooked. If you want to keep looking beyond the frame, The Matilda Project is a very good place to start.
REFERENCES
Arslan, J., Azimi, S., Sami, L., Ajili, F., Domingues, D., Akpinar, D., Christiansen, L. V., Zujovic, V., & Benke, K. K. (2024). Know their name: An anthology of great women in the shadows [Preprint]. arXiv. https://doi.org/10.48550/arXiv.2411.08038
Rossiter, M. W. (1993). The Matthew Matilda Effect in science. Social Studies of Science, 23(2), 325–341. https://doi.org/10.1177/030631293023002004
Gage, M. J. (1883, May). Woman as an inventor. The North American Review, 136(318), 478–489
Ross, M. B., Glennon, B. M., Murciano-Goroff, R., Berkes, E. G., Weinberg, B. A., & Lane, J. I. (2022). Women are credited less in science than men. Nature, 608(7921), 135–145. https://doi.org/10.1038/s41586-022-04966-w
Reggiani, M., Gagnon, J. D., & Lunn, R. J. (2024). A holistic understanding of inclusion in STEM: Systemic challenges and support for women and LGBT+ academics and PhD students. Science Education, 108(6), 1637–1669. https://doi.org/10.1002/sce.21899
Sederholm Lawesson, S., Isaksson, R., Thylén, I., Ericsson, M., Ängerud, K., & Swahn, E. (2018). Gender differences in symptom presentation of ST-elevation myocardial infarction – An observational multicenter survey study. International Journal of Cardiology, 264, 7–11. https://doi.org/10.1016/j.ijcard.2018.03.084
Van der Ende, M. Y., Juarez-Orozco, L. E., Waardenburg, I., Lipsic, E., Schurer, R. A. J., Van der Werf, H. W., Benjamin, E. J., Van Veldhuisen, D. J., Snieder, H., & Van der Harst, P. (2020). Sex-based differences in unrecognized myocardial infarction. Journal of the American Heart Association, 9(13), e015519. https://doi.org/10.1161/JAHA.119.015519
Mapplebeck, J. C. S., Beggs, S., & Salter, M. W. (2016). Sex differences in pain: A tale of two immune cells. Pain, 157, S2–S6. https://doi.org/10.1097/j.pain.0000000000000389
Mogil, J. S., Parisien, M., Esfahani, S. J., & Diatchenko, L. (2024). Sex differences in mechanisms of pain hypersensitivity. Neuroscience & Biobehavioral Reviews, 163, 105749. https://doi.org/10.1016/j.neubiorev.2024.105749
Sudat, S. E. K., Wesson, P., Rhoads, K. F., Brown, S., Aboelata, N., Pressman, A. R., Mani, A., & Azar, K. M. J. (2023). Racial disparities in pulse oximeter device inaccuracy and estimated clinical impact on COVID-19 treatment course. American Journal of Epidemiology, 192(5), 703–713. https://doi.org/10.1093/aje/kwac164
Comments