Learning Science · Learning with digital environments

Interaction, feedback and meaningful practice

Design interactions that require meaningful decisions and feedback that learners can use.

Use the examples to connect this lesson with the earlier topics. No specialist equipment is needed.

After reading, you can

Distinguish meaningful interaction from repeated clicking.
Connect feedback to a revised attempt.

A click is not necessarily a thought

A quiz awards points whenever a learner clicks the next button. Another asks the learner to choose an explanation, justify it and revise after feedback. Both are interactive in a technical sense. Only the second clearly requires the target reasoning. Interaction quality depends on the decision and its consequences for learning.

Make the response diagnostic

Ask for something that reveals the relevant idea: a prediction, a method choice, a placement on a diagram or a short explanation. A multiple-choice response can be informative when alternatives reflect meaningful distinctions, but a guessed correct answer may still conceal uncertainty. Add a reason or a fresh case when needed.

Feedback should identify a useful relation between the response and the target. “Wrong” gives little direction. An explanation that shows why a denominator is inappropriate can guide a new attempt. Showing the full solution immediately may be useful at one point and may remove needed practice at another.

Close the loop

Give learners time and a task in which to use feedback. A correction that appears only after the final submission cannot guide the completed attempt, though it may guide future work. Provide another opportunity while the relationship remains clear.

Watch for behaviours that bypass reasoning, such as clicking every option until the system accepts one. Restricting attempts is not the only response; require a prediction or explanation and use a new case after feedback. Keep navigation and response accessible so interface skill does not become an accidental requirement.

Worked example

An informative ratio interaction

Why are 3:5 and 6:8 not equivalent, and what should a feedback interaction ask next?

The learner chooses whether 3 to 5 and 6 to 8 are equivalent and explains the choice. If the learner adds three to both quantities, feedback asks what happens to each quantity under a common multiplication. A diagram shows the first pair scaled by two, producing 6 to 10. The learner then judges a different pair and explains it. The interaction connects a visible error to the target relation and a subsequent check.
A delayed version of the ratio question should change the numbers and remove the feedback diagram initially. If the learner now explains a common scale factor, the repair has some support. If the learner simply remembers which option was accepted before, the record of successful clicks overstated the result. The new question distinguishes these possibilities without adding unrelated content.

Scaling both terms of 3:5 by two gives 6:10. After explaining the additive error, use a fresh pair without the accepted-answer cue to inspect common-factor reasoning.

A common factor differs from a common addition

The first comparison scales both 3 and 5 by 2, producing 6:10 and preserving the ratio. The second adds 3 to each, producing 6:8, which does not preserve 3:5. Arrows describe these arithmetic operations, not observed learning gains.

Try it yourself

Think through each question before opening the answer.

Why can unlimited attempts produce a misleading success record?
Show the answer and reasoning
A learner may reach acceptance by trying every option without learning the relationship. A fresh explanation or task can check what changed.
Acceptance after trying every option records the system’s stopping condition, not necessarily a learned relation. A new reasoned choice distinguishes reasoning from option exhaustion.
What makes feedback actionable?
Show the answer and reasoning
It identifies a relevant gap or successful feature and helps the learner choose and carry out a next step.
Actionable feedback identifies a criterion-relevant feature and permits a next attempt. A bare verdict does not show how to revise the method.

What to carry forward

Require the decision that matters for the target.
Give feedback a route into a revised attempt.

References

Sources supporting the explanations and learning checks in this lesson.