Why Low-Stakes Conversation Teaches More Than High-Stakes Tests
In game theory, there is a term that sounds like an insult: cheap talk.
It means communication that costs nothing and commits no one. A message sent before a game begins. A hint. A conversation that doesn't change the score. For a long time, economists assumed cheap talk didn't matter much. If words are free and nothing rides on them, why would anyone trust them or act on them?
It turns out they do. Under the right conditions, free, low-pressure communication changes behavior a great deal. It builds trust, shares information and helps people coordinate.
My doctoral research applied that idea to the classroom, and it became the core of my book Cheap Talk, Big Gains in Education. The argument is simple: students learn more from frequent, low-stakes conversation with their teachers than from rare, high-pressure tests.
What high stakes do to learning
Think about what a big exam asks a student to do. Learn the material, hold it in memory until a set date, reproduce it accurately under time pressure, and then, in practice, forget most of it.
The student isn't being foolish. They are responding rationally to the game they have been placed in. When the reward is attached to performance on one day, the sensible strategy is to perform on that day. Understanding is optional. Appearing to know is what the system pays for.
This is the core problem: incentive structures in traditional schooling reward the appearance of knowledge rather than the real thing. If we want different behavior, we have to change the game.
What low stakes make possible
Now picture a different classroom. The teacher asks questions all through the lesson. Students answer, guess, get things wrong and try again. Nothing they say goes into a grade. The talk is cheap.
Three things happen.
Students say what they actually think. When a wrong answer costs nothing, students stop hiding their confusion. The teacher can finally see what is going on inside their heads.
Mistakes become information. A wrong answer in conversation is a clue about where understanding broke down. A wrong answer on a final exam is just a lost point.
Memory gets stronger. Research on retrieval practice, including well-known studies by Henry Roediger and Jeffrey Karpicke, has found that recalling information through low-stakes quizzing helps people remember it longer than simply re-reading it. Being asked, often and without pressure, is one of the best ways to learn.
This is an old idea
None of this is new. Socrates taught almost entirely through questions. He believed real understanding could only come from live dialogue, where ideas are tested, challenged and revised between minds that are actually engaged.
I trace that tradition in my book Dialogic Education: A Historical and Contemporary Perspective, from ancient Athens to the age of artificial intelligence. What game theory adds is an explanation of why it works. Dialogue changes the incentives. When talking is safe, honesty becomes the best strategy.
What this looks like in practice
For teachers and parents, the shift doesn't require a new curriculum. It requires a change in how often we ask and how much rides on the answer.
Ask more, grade less. Short, frequent questions do more than one big test.
Reward reasoning, not just results. Ask “how did you get there?” as often as “what’s the answer?”
Make wrong answers useful. Treat them as the start of a conversation, not the end of one.
Let students ask too. A question from a student is the clearest signal of where their thinking is.
The bigger lesson
We often assume that pressure produces effort and effort produces learning. Sometimes it does. But pressure also produces performance, and performance is not understanding.
The classrooms that produce durable learning tend to be the ones where talk is cheap, questions are constant and nobody is afraid to be wrong out loud. That is not a soft approach. It is a better-designed game.
Dr. Matthew Weinberg, PhD, MEd is an educator and the author of nine books, including Cheap Talk, Big Gains in Education and Dialogic Education: A Historical and Contemporary Perspective.
Comments