- Evidence Snacks
- Posts
- Assume The Worst
Assume The Worst
Overcoming 'Expert-Induced Blindness'
Hey 👋
Hope your Jan’s chokka with joy. Today, another snack on thinking smarter in the classroom…
Big idea 🍉

As teachers, we are at constant risk of ‘expert-induced blindness’ (or ‘curse of knowledge’). This is a cognitive bias which makes it hard for us to set aside what we know and empathise with those who don’t. Answers come so easily to us—it’s hard to imagine anything otherwise.
This is especially common in schools, where the knowledge gap between teachers and students is substantial. It leads us to overestimate what students know and how readily they can access and use that knowledge.
(It’s also why newer teachers can sometimes be great mentors in certain respects, and why classroom peers can sometimes offer really helpful explanations)
Expert-induced blindness shows up in lots of ways:
When we explain, we can easily assume too much prior knowledge—especially technical terms or jargon.
When we pose questions, we don’t always give our students enough thinking time.
We underestimate how much rehearsal is needed to become fluent in even simple vocab or ideas.
We assume that students appreciate the value of what we’re teaching them, or why we’re doing it this way.
How can we reduce the negative impact of our expert-induced blindness? The most powerful approach may be to get into the habit of ‘assuming the worst’ when it comes to student understanding.
For instance, after explaining a concept, it’s tempting to assume students have grasped it, and proceed accordingly. Instead, it’s better to flip our mental script: assume that they haven’t, and then gather evidence to prove ourselves wrong, ideally in a rigorous and efficient way, using tools such as: whole-class finger-voting, mini-whiteboards, or heads-down-thumbs-up.
Assume less, assess more.
This shift reframes our practice. Assuming the worst makes us more responsive, rigorous, and ultimately more effective.
NOTE: While assuming the worst works well for student learning, the reverse is generally better when it comes to behaviour… to assume the best of intentions until proven otherwise.
🎓 For more, check out this study on efforts to try to reduce expert-induced blindness.
Summary
Expert-induced blindness makes it hard for us teachers to set aside what we know and empathise with those who don’t.
The best way to mitigate this bias is to adopt the position that ‘my students don’t know X until I prove to myself otherwise’.
Assuming the worst is best used in relation to student understanding—when it comes to behaviour, it’s better to assume the best.
Little updates 🥕
Study on how teacher feedback affects relationships in class → finds positive feedback makes kids more liked, while negative reduces popularity, especially when peers are empathetic.
Meta-analysis on metacognition interventions for young children → suggests they tend to improve self-regulation, academic outcomes, and executive functions, with long-term benefits for self-efficacy.
Paper examining school leader use of data → finds that benchmark assessments are typically underused, even though they correlate well with improved student outcomes, especially in schools with high staff trust.
Analysis of ChatGPT for lesson preparation → finds it reduces teacher workload by 31% without compromising lesson quality.
Support the snacks & get a smarter brain → Hello Snacks PRO
Laters potaters.
Peps 👊
PS. Tomorrow sees the launch of a secret project I’ve been working on for the last 18 months. Can’t wait to share with you—full access coming your way over the next couple of weeks.