A learner circles has lived in a present-perfect exercise and gets the item right. That mark tells us almost nothing about the decision. The learner may have noticed since 2021, remembered a rule about unfinished time, copied the shape of the previous item, or guessed. Ask for one sentence of explanation before revealing the answer, and the exercise starts showing how the grammar is being processed.
The claim here is narrow: a brief explanation can make grammar practice more diagnostic and can push a learner to connect a form with meaning. It cannot rescue a confusing rule, and explanation should never consume the whole lesson. For a Taiwanese learner around B1 who has completed years of multiple-choice worksheets, sixty seconds spent stating a reason may be more useful than six more unchecked items.
What an answer conceals
Selected-response practice is efficient. A teacher can scan 18 answers quickly, and a learner can complete a set on the MRT. Its weakness is equally plain. The same tick can come from several mental routes. Only some of those routes will transfer when the learner has to understand a sentence or write one without options.
This is why the prompt should ask about the evidence inside the sentence. For Mina has worked here since March, a useful explanation might be: “Since March gives a starting point, and the situation continues now.” The learner has named the time expression, the meaning, and the connection to the verb form. “Because present perfect is the rule” adds no usable information.
Grammar practice already improves when learners attend to meaning as they choose a form. Choc's article on beginning grammar practice with interpretation explains why a sentence should be understood before it is manipulated. An explanation prompt adds a second window: it lets the learner and teacher inspect the reason behind the choice.
What the evidence can support
Self-explanation has a substantial research base across school and university subjects. A 2018 meta-analysis by Bisra and colleagues combined 69 effect sizes from 64 reports and found a weighted mean effect of g = 0.55. The tasks, ages, subjects, and prompts varied. That average supports self-explanation as a learning move. it does not give a guaranteed effect for an English tense exercise. The authors also warned that scripted prompts have limits.
The language-learning evidence gives a more specific reason to attend to the learner's processing route. McManus and Marsden assigned 53 English-speaking learners of French to three treatments. All received information and interpretation practice in French. One group also received explicit information about how English and French processing routines differed. Increasing practice led to faster and more accurate performance only for that group. This was French learned by English speakers, so it cannot be copied directly onto every Mandarin-English contrast. It does show that information about the route from the first language to the second can change what practice accomplishes.
An older study supplies the mechanism behind the classroom prompt. Chi and colleagues analysed how students talked through worked examples in mechanics. Stronger learners generated explanations that connected solution steps with principles and monitored gaps in their understanding. Physics and English grammar differ in obvious ways. The useful transfer is the act of making a hidden inference available for inspection.
Keep the prompt small
A full grammatical lecture after every item would crush the pace. Use explanation where the decision carries meaning or where Mandarin offers a tempting alternative. Two prompts cover most cases:
- “Point to the words in the sentence that helped you choose.”
- “Describe the meaning that would change with the other answer.”
Consider I stopped to smoke and I stopped smoking. A learner who can explain the difference has connected the complement to two event structures. A learner who merely labels one form “infinitive” and the other “gerund” has supplied terminology, not evidence of interpretation. Ask for the scene each sentence describes. One person pauses another activity in order to smoke. the other ends the habit.
Chinese can be useful during this check. If a B1 learner needs twenty seconds of華語 to express the distinction accurately, that is better diagnostic evidence than a vague English sentence produced under pressure. The next attempt can compress the idea into English. The purpose of the explanation is to reveal a decision, so language difficulty should not hide it.
Correct the reason as well as the answer
Four outcomes are possible, and each calls for a different response.
- The answer and reason are sound. Move on.
- The answer is right and the reason is weak. Ask for the sentence cue, then test a near neighbour.
- The answer is wrong and the reason is coherent. Show where that rule stops applying.
- The answer and reason are both uncertain. Return to meaning with two contrasting examples.
The third case deserves patience. A learner might choose did you ever visit for a life-experience question because a completed trip happened in the past. The reasoning has internal logic. The teacher can contrast an open life period with a finished visit to Japan in 2023, then let the learner decide again. Simply marking the item red would lose the most informative part of the attempt.
Where explanation becomes a burden
Beginners may lack enough language and grammatical concepts to explain a choice. Time pressure also changes the task: during a timed TOEIC section, extended commentary is unrealistic. Use the prompt during learning, then remove it during fluency work and test rehearsal.
Some items are poor candidates. Articles in fixed phrases, irregular forms, and highly frequent chunks may be learned partly through repeated exposure. Forcing a rule onto every instance can produce a tidy story that the language itself does not support. “It sounds familiar in this phrase” is sometimes the honest answer, and a corpus or teacher can verify that familiarity.
A workable routine takes four minutes. Choose two items from a completed set, including one the learner found difficult. Hide the answer key. For each item, state the chosen form, point to the cue, and describe the meaning. Check the answer. If the reason failed, write one contrasting sentence and try it again 48 hours later. Two items are enough. Stop there. A learner who can point to a time expression, describe the scene it creates, explain why a competing form changes that scene, compare the decision with a nearby example, and then apply the same reasoning after the options disappear has supplied far richer evidence than a page containing ten ticks and two crosses. The revealing detail is the word the learner points to before the answer appears.