Perspective

Finding the right angle is easier said than done. Precisely because so much research has been published in recent years with fantastic visuals (Murienne et al., 2008; Ziegler et al., 2010). These are almost always scans; you still see an illustration occasionally, especially in paleontology (Witmer & Ridgely, 2008). In addition, despite the fact that I have been in the world of scientific visualization for more than twenty years, I wonder if I can see everything. I think so, but how can I appreciate the bias, which I must undeniably have?

For now, I think I just know what’s going on and that, as soon as I notice that I see developments that I haven’t noticed before, I adjust my focus. But the fact that there are so many fantastic visuals makes my work in this area not easy to explain. As a result, a grant application earlier this year was rejected: it was peer reviewed and we, I, could not make it clear why what I am going to do is both daring and new. In addition, visualizations are still viewed differently than text (Falletta et al., 2026). My research is also intended to change that, so for now it’s a bit of a chicken-and-egg story.

I will have to practice writing. Not only to make what I do really clear for once, in science and to the wider public (Erdt et al., 2017), but also because writing helps with thinking.

So I write.

I will try to explain what I am going to do and why. First of all, those incredibly beautiful visuals. Nowadays, researchers can make (CT) scans of all kinds of anatomical structures fairly easily (Lauridsen et al., 2011; Gray et al., 2025). And there are an awful lot of applications, which are increasingly accessible, in which you can distill something that looks good from those scans (Irschick et al., 2022; Plum & Labonte, 2021). Scans are also useful because you want to know something about the degree of variation in a structure, how that structure lies in a specific animal and what it looks like for that specific animal (Cardini & O’Higgins, 2026; Salazar‐Fernández et al., 2026).

Scans are incredibly useful for such studies and have therefore given a boost to specific research, for example in model organisms such as the zebrafish or the chicken embryo (Teame et al., 2019; Smallridge et al., 2025). But it is still raw material, focused on a (small) part of an animal, and it is research that means nothing if you don’t know the basics. And what is that basis? Comparative anatomy. As a biologist, someone who studies a certain animal species, you need to know how the anatomy and physiology of such an animal works (Diogo & Molnar, 2016).

You can do that by wrestling through books, looking at photos and scans of anatomical structures, but an animal in itself is a system in which everything interlocks (Ma’ayan, 2017; Nartallo-Kaluarachchi et al., 2025). Where you have to know what is normal and what is not. Of course, everyone gets a basis of biology at school. So that’s where you have to start. Does the way in which we now teach anatomy ensure that students really understand how the different solutions to life, because that is what you teach with comparative anatomy, came about, how the form is related to the function and how, if you know the field well in terms of content, you can already reason very well what the anatomy of a hitherto unknown animal looks like, for example? (Bhattacharjee et al., 2022; Cortese & Frascio, 2025).

And that’s where I’m now trying to make a difference with the anatomical 3D models of animals. Because if you want to teach the basics well, you first have to reduce topics to the core. And with anatomy, that means that you have to make sure that a model has the most common forms of structures, that individual differences do not initially occur in a model, that there is no noise in it and that a model is clear (Erolin, 2025). And that all models are constructed in the same way, because you want to use them for comparative anatomy (Mitsuhashi et al., 2009; Ng et al., 2009).

Creating such models relies on a lot of reference data. The scans we talked about earlier are part of that data, but they need to be used together with all kinds of other data to ensure that there is a consistent and standardized whole (Tarasov, 2019).

Comparative anatomy education as I envision it can therefore be compared to how architecture has almost completely transitioned from 2D CAD to 3D BIM (Building Information Modeling). Of course, construction drawings are still delivered, but they are a derivative of the BIM used for the construction site and in legal documents, not something that is used for design and decision-making. Ultimately, such models will be transformed into digital twins that allow the building, once built, to be monitored and operated down to the capillaries (Yang et al. 2024).

This is what I’m going to do with the anatomical 3D models of animals: make them more than just a model. That you can use it to perform simulations, that you can use it in education, but also in actual research (Armeni et al., 2022; Kumar et al., 2026). That the methods I use are standardized and usable by others. And that, where possible, I am not reinventing the wheel.

references