People who know me know this about me: I sometimes let my humour shine in my classes and lectures. Not because I see myself as an entertainer, but because I have learned that a well-placed joke can defuse tension, bring attention back, or help an abstract point land just a little better. I deliberately invest time in doing that. At the same time, over the years, I have seen more than enough examples of people who probably should not do this. Student teachers or lecturers who overplay their own awkwardness. Speakers who seem to think that a random joke automatically creates a connection. Or worse: humour that comes at someone else’s expense, a student, or sometimes an entire group.

That discomfort does not just live in my gut feeling. There is now solid experimental research that sharply underpins this intuitive distinction between humour that helps and humour that harms learning. In a recent study published in Learning and Instruction, Sonja Bieg and colleagues examined how different types of teacher humour affect teaching quality, student motivation, and students’ emotions. They did not rely on retrospective questionnaires about favourite or disliked teachers, but used a tightly controlled experimental design. The same lesson. The same teacher. And the same content. The only thing that varied was the humour. Or more precisely: the type of humour.

The researchers distinguished four forms.

• Humour that is explicitly connected to the subject matter.
• Humour that is unrelated to the content, such as anecdotes or jokes on the side.
• Self-deprecating humour, where the teacher makes themselves the target.
• Aggressive humour, where the joke is at the expense of students.

There was also a control condition with no humour.

What did they find? Not all humour is equal. And more importantly, not all humour is harmless or effective.

Humour directly related to the lesson content consistently shows positive effects. Both experienced teachers and students rate these lessons as more interesting. The teacher–student relationship improves. Students experience more enjoyment, higher intrinsic motivation, and less boredom. Crucially, this kind of humour does not cost instructional time. On the contrary, it is associated with higher perceived time-on-task. The joke supports learning instead of interrupting it.

That aligns closely with what many teachers recognise when humour works. An absurd comparison to clarify a concept. A playful exaggeration that highlights the core idea. Humour as a didactic tool, not as a stand-up routine.

Humour that is unrelated to the content is more ambiguous. I am personally more cautious with this kind of humour, because it can easily distract. Think of what we sometimes call seductive details. Students may find the teacher more likeable and the lesson slightly more enjoyable, but cognitively, it does not add much. There are no clear gains in clarity or motivation. Among experts, this type of humour even raises doubts about the lesson’s overall interest. It is neither destructive nor reliably helpful. A joke for the sake of a joke can just as easily distract as connect.

Self-deprecating humour may be the most treacherous type. In theory, it sounds appealing: the teacher who does not take themselves too seriously. In practice, the effects are limited. It can be disarming at times, but it does not produce clear gains in motivation, interest, or clarity. In some cases, it even undermines perceived focus and time-on-task. Modesty, on its own, is not a didactic strategy. This is something I need to keep an eye on myself.

And then there is aggressive humour. The findings here are unambiguous. Students experience more anger and anxiety, less motivation, and a poorer relationship with the teacher. Both teachers and students rate these lessons systematically more negatively. This is not a matter of “some students can handle it”. The effects are consistent and clear. Humour that humiliates undermines learning.

Perhaps the most interesting contribution of this study is not that humour sometimes works and sometimes does not. We already suspected that. What this research shows very clearly is why humour works when it works. The positive effects do not run directly from joke to motivation. They are mediated by teaching quality. By the teacher–student relationship. By perceived interest and clarity. Humour should not be a random add-on. At best, it amplifies what is already happening didactically.

That also explains why some people are better off not using humour in their teaching. Not because humour itself is dangerous, but because it becomes detached from learning. Or worse, because it is used to display power, discharge tension, or mask insecurity.

Humour in education, then, is not a personality trait. It is a pedagogical choice. And as with so many options in teaching, the question is not whether you do it, but how and why. A good joke can open a lesson. A bad one can close it.

One thing I learned early on when I started teaching is confirmed once again by this research: you do not need to be funny to be a good teacher. But if you are, make sure your humour serves learning. The rest can safely be left out.