I have read several recommendations, e.g. this one to use different font families for titles (sans-serif), the text body (serif) and graphs (sans-serif). This is also what the default article document class in LaTex uses. I read that those fonts should be “complementary”. So this seems to be a widely accepted practice, unless it is taken too far.
But I wonder why? Afaik design should be as simple as possible, be consistent and clean, especially for formal, scientific papers. Why does the title not stand out enough by having a bigger font size or increasing the font weight? Shouldn’t the text in graphs be consistent with the one in the body of text?
So far I read only recommendations but no reasoning behind this decision? Have there been studies regarding this? Are there any design principles that outweigh the drive for simplicity? Or do complementary font types just look that much better that there is no discussion necessary?
I made some quick examples (the former looks better to me and has the same font):
Edit: I understand that a different font face increases the contrast between the title and the body of text, but as already written: Why is increasing the font size/weight not enough? Why sacrifice coherence if it can be avoided?
Using different fonts for headings & body text is very general advice, based on some assumptions, which are not necessarily true in every case:
- Many faces are designed for display use (more ornamented, work better at bigger sizes) or text (increased legibility and performance for long texts/small sizes while perhaps sacrificing uniqueness), so it makes sense to use a display face for titles and a text face for running text. There’s many faces that work well for both, depending on their use.
- Traditionally, in publications, you wanted to use a serif typeface for text and either a sans or a display serif for titles. Very related to the point above, but more historical than practical.
- You generally want to create contrast in your text between headings, running text, captions, formulas, etc. While having different typefaces is not the only way to do it, it is a pretty effective way to do it. you could also use spacing, use variants of the same type family (bold/regular/thin or condensed/expanded)
So as you see it’s not a hard fast rule, but if you understand the principles behind the rule of thumb, you can choose other options that allow you to do good typesetting regardless of if you use the same or different fonts.