The Review Question That Tells You Nothing

"What did we learn today?"

Watch what happens when a teacher asks that at the end of a lesson. The three kids who always talk say something close to what the lesson was about. Everyone else waits it out. And the teacher walks away thinking the material landed, because a few confident voices repeated the main point back.

It didn't land. Not for most of the room. That closing question measures who was paying attention in the last ninety seconds, not who will remember any of this next week. And retention next week is the only thing that actually matters, because Sunday school is a serialized format. You're not teaching one lesson. You're teaching lesson 34 to kids who half-remember lessons 1 through 33.

Here's what the recall research has been saying for a long time, and what most children's ministry curriculum still ignores: the act of retrieving information is what strengthens memory, not the act of receiving it. Reviewing your notes feels productive and does almost nothing. Trying to pull the answer out of your own head, struggling with it a little, that's what makes it stick. This is the testing effect, and it's one of the most replicated findings in learning science. Roediger and Karpicke's work on retrieval practice showed students who were quizzed on material retained far more than students who reread it the same number of times.

So the problem with "what did we learn today" isn't that it's a review question. It's that it's a group review question with an escape hatch. Any kid can stay silent and let the extroverts carry it.

The difference between recognition and recall

Most lesson-ending review is recognition, not recall. Recognition is "Was Zacchaeus short or tall?" The answer is sitting right there in the question. Recall is "Why did Zacchaeus climb the tree?" Now the kid has to reconstruct the whole scene: the crowd, his height, wanting to see Jesus. That reconstruction is the actual work of learning.

In my experience watching lessons across different rooms, teachers default to recognition because recognition produces faster, more confident answers, and a room full of confident answers feels like a room that learned something. It's a false signal. I've watched a class nail every recognition question and then, seven days later, not remember which story we were even talking about.

The fix is not complicated, but it costs you something: silence. Real recall questions produce pauses. Uncomfortable ones. You ask, "How would you explain what grace means to someone who's never heard the word?" and then you have to sit in three seconds of nothing while eight-year-olds actually think. Most teachers can't tolerate that pause and rescue the room by answering their own question. Every time.

What to do instead this Sunday

Stop asking the whole group at once. Distribute the retrieval so no one can hide, and space it out instead of dumping it all at the end.

• Ask individuals.
• Open the next week's lesson by asking them to recall last week's story before you give them any reminder at all, which forces retrieval across a gap of seven days, and spacing retrieval across time is where the memory research gets genuinely strong. That one move, a two-minute cold recall at the start, probably does more for retention than the entire closing review you're currently running.
• Have them explain it to a partner, because generating an explanation is a harder retrieval task than answering a question, and harder retrieval (up to a point) builds stronger memory.

Here's my honest caveat. I don't know the exact difficulty sweet spot for a given age group, and I'm skeptical of anyone who claims they do. Push too hard and a struggling reader or a kid with a learning difference just shuts down, and now you've taught them that Sunday morning is another place they fail. The research on "desirable difficulty" is real, but the word desirable is doing a lot of load-bearing work, and where that line sits depends on the specific kid in front of you. I've seen the same recall question energize one class and flatten another.

What I'm confident about is narrower: the closing group question you're relying on is measuring the wrong thing. It tells you who's awake. It doesn't tell you who learned. If you want to know whether the lesson worked, ask one quiet kid, next week, before you've reminded them of anything, and see what comes out.

Usually it's less than you hoped. That gap is the actual information. Everything before it was just noise.