PTE Highlight Incorrect Words
Learn PTE Highlight Incorrect Words with current Pearson scoring, a cursor-tracking method, recovery rules, original examples, and false-alarm practice.
In PTE Highlight Incorrect Words, click only when the spoken word at the current transcript position is genuinely different from the displayed word. Keep your eyes, ears, and cursor aligned, because the recording plays once and an incorrect selection costs a point.
The quickest way to lose this task is to win an argument with a word that has already passed. You hear something doubtful, replay it mentally, and by the time you decide, the speaker is several words ahead. Use a forward-only rhythm: hear → compare → click if clear → release → rejoin.
Current PTE Highlight Incorrect Words rules and scoring
Official-source check: 26 August 2026. This guide uses the current PTE Academic and PTE Academic UKVI format. Pearson's 2025 update applies to tests after 7 August 2025. Pearson also stated that the existing question types kept their task formats, timings, and expected response types when the two new speaking question types were added, so Highlight Incorrect Words remains an existing listening task rather than a newly redesigned one.
| Current Pearson detail | What it means for you |
|---|---|
| Task | You see a transcript containing words that differ from the recording and select the words that are different. |
| Recording length | 15–50 seconds. |
| Preview | Pearson gives 10 seconds before the recording and recommends skimming for the topic and information-bearing words. |
| Playback | The audio plays automatically once. There is no second listen during the item. |
| Current item range | The current Pearson Test Taker Score Guide lists 2–3 Highlight Incorrect Words questions. |
| Skills assessed | Listening and Reading. |
| Scoring | Each correctly selected word earns +1. Each incorrectly selected word receives −1. The item has a minimum score of 0, so the item itself does not finish below zero. Partial credit applies. |
That scoring rule is why “click anything suspicious” is a bad strategy. A word sounding unfamiliar is not enough. You need evidence that the speaker actually said a different word at that position.
Also notice what the current official sources do not give you: they do not establish a universal number of mismatches inside every item, and the task pages checked for this guide do not publish a fixed percentage weight such as “this task is exactly X% of your Listening score.” Do not build your strategy around those prep-site shortcuts.
If you need the broader section structure beyond this one task, use the PTE Listening format and study plan. Here, stay narrow: transcript versus recording, word by word.
Use the 10-second preview to build a map, not predict answers
You are not trying to memorize the paragraph in 10 seconds. That is a fine way to turn a short preview into an unnecessary stress experiment. You are building landmarks so your moving window has somewhere to go.
There is an important distinction here: content words deserve preview attention, but every word still deserves comparison during the recording. A small preposition can be the genuine mismatch. “Important word” is a preview concept, not a scoring rule.
Keep a moving window while the audio plays
Once the recording starts, stop reading like someone studying a paragraph and start tracking like someone following a moving subtitle. Your useful attention is narrow: the word or short phrase you are hearing now, plus just enough text ahead to stay oriented.
- Hear: receive the current word or short phrase before deciding anything.
- Compare: match it to the word at the same position on screen.
- Click if clear: if the speaker clearly used a different word, select the displayed word.
- Release: stop analysing that decision as soon as it is made.
- Rejoin: move your eyes and cursor with the next spoken phrase.
Pearson specifically recommends following the text with the cursor and not taking notes for this task. That makes sense: your scarce resource is alignment. If your hand is writing while your eyes are hunting backward, your moving window has turned into three separate windows, which is not the productivity breakthrough it sounds like.
If you lose your place, recover forward
Suppose you hesitate near regional rainfall patterns. By the time you decide that you did not hear the phrase clearly, the speaker has moved on. Do not drag the cursor backward and mentally replay the old phrase. Scan a few words ahead for the next strong landmark you can hear — perhaps the second phase or annual measurements — reconnect there, and continue.
The missed position may be gone. Protect the positions that are still available.
Different word, or just a different sound?
This is the decision that separates controlled clicking from suspicion-clicking. Your question is not “Did that sound strange?” It is:
Did the speaker say a different word at this exact transcript position?
| What happened | Action | Why |
|---|---|---|
| The transcript says gradually; the speaker clearly says rapidly. | Click. | A different word was spoken at the same position. |
| The transcript says for; the speaker says the same word very weakly and quickly. | Do not click. | The pronunciation changed, not the word. |
| The transcript says data; the speaker uses another common pronunciation of data. | Do not click. | Accent or pronunciation variation does not create a new word. |
| The transcript says in the final stage; the speaker clearly says at the final stage. | Click in. | The small function word changed. Small does not mean unscored. |
| A word sounds unclear and the audio has already moved on. | Usually leave it and rejoin. | Uncertainty is not evidence of a different word, and chasing the old sound can cost later positions. |
A four-question evidence filter
- Am I comparing the spoken word with the displayed word at the same position?
- Did I actually hear a different word, rather than the same word with a different pronunciation?
- Is my evidence coming from the recording, not from what I think the grammar or topic should say?
- If the evidence is already gone, can I release it and rejoin the next clear landmark?
Meaning and grammar can help you confirm what you heard, but they do not replace the recording. Both alternatives can be perfectly grammatical. The audio is the evidence.
When the audio stops: audit evidence, do not invent it
The live rule gets you through the recording. After the recording has finished, your job changes slightly: review the evidence you still remember clearly rather than starting a new round of grammar-based guessing.
- Keep a selection when you clearly remember hearing a different word at that exact position.
- Deselect a word if you realize the speaker actually said the same word and it only sounded unfamiliar because of pronunciation, reduction, linking, or hesitation. Pearson allows you to click a selected word again to deselect it.
- Do not add a fresh selection merely because the written sentence now looks odd, unexpected, or less logical than another possible wording.
- Leave unresolved uncertainty alone when the remembered audio evidence is genuinely unclear. A new guess can create a −1 incorrect selection; the current item score has a minimum of 0.
Do not invent a fixed post-audio review routine or countdown. The useful rule is simpler: keep clear evidence, remove a clear false alarm, and do not manufacture a new mismatch after the sound is gone.
Original worked examples: mismatches and false alarms
The examples below are written specifically for this guide. They are not copied from a real PTE question, test bank, recording, transcript, or answer set. Because there is no audio here, each example gives you both the displayed transcript and an audio-independent spoken version. Use the pair to practise the decision logic.
Example: a content-word change plus a small preposition
Displayed transcript: The city’s new tree-planting program is expected to reduce summer temperatures gradually. Researchers will monitor shaded streets for three years and compare them with nearby areas that receive no additional planting.
Spoken version: The city’s new tree-planting program is expected to reduce summer temperatures noticeably. Researchers will monitor shaded streets for three years and compare them to nearby areas that receive no additional planting.
Show the words to select and the reasoning
Correct selections: gradually and with.
- Gradually must be selected because the spoken word is noticeably.
- With must be selected because the speaker says to. Both versions can form acceptable English in this example, so grammar does not decide the answer; the spoken word does.
- For should not be selected even if it sounds weak in natural speech. In this example the speaker still says for. That is a pronunciation false alarm, not a word mismatch.
Example: two plausible words, one recording
Displayed transcript: The survey found that students who reviewed the material briefly before class were more confident during group discussions, although the effect was smaller for students who already knew the topic well.
Spoken version: The survey found that students who reviewed the material regularly before class were more confident during group discussions, although the effect was smaller for students who already knew the topic well.
Show the word to select and the false-alarm check
Correct selection: briefly.
Both sentences are grammatical and believable. That is exactly why guessing from meaning is unreliable. The displayed word is briefly; the speaker says regularly; select the displayed word. Nothing else changes.
Example: a tiny substitution you cannot dismiss
Displayed transcript: The samples were stored in the final stage of the experiment and examined again after forty-eight hours.
Spoken version: The samples were stored at the final stage of the experiment and examined again after forty-eight hours.
Show the answer
Correct selection: in.
A preposition may look too boring to be important, but this task scores word identity. The speaker says at, so in is a genuine mismatch.
Decision lab: would you click it?
Make each decision before opening the explanation. The goal is not to collect six easy “right answers.” It is to notice which rule you reach for when the sound or sentence feels uncertain.
Five failure modes that cost attention or points
| Failure mode | What goes wrong | Exact repair |
|---|---|---|
| Clicking every uncertain word | “That sounded odd” becomes “that must be wrong.” With negative marking, suspicion has a price. | Demand a different spoken word at the same position. If you cannot identify the difference clearly, do not manufacture certainty. |
| Losing position after one mismatch | You open a mental courtroom for the old word while the speaker keeps talking. | Release the decision and recover at the next strong phrase you can both hear and see. |
| Treating pronunciation variation as a text error | An accent, weak vowel, linked phrase, or fast delivery makes the same word sound unfamiliar. | Ask whether the word itself changed. Same word, different pronunciation: leave it. |
| Overlooking function-word substitutions | You focus so heavily on nouns and verbs that in → at or another small change slips past. | Use content words as preview anchors, but compare every position during audio. |
| Guessing from grammar or topic | You click the word you dislike rather than the word that actually differs from the recording. Grammar becomes an unreliable alibi. | Use meaning only to check what you heard. The recording decides the word identity. |
The 10-second map challenge
Set a 10-second timer. Read the original transcript below once. Do not try to memorize it. When the timer ends, name the topic in a few words, identify two structural landmarks, choose three strong anchor words, notice one or two function words that may sound reduced, and know where your cursor would start.
Urban wetlands can reduce flood risk by temporarily storing storm water after heavy rainfall. Researchers are now comparing restored wetlands with conventional drainage systems to measure how quickly each approach removes water from residential areas. The project will also examine maintenance costs and changes in local biodiversity over a five-year period.
Show a model 10-second map
One reasonable map would be:
- Topic: wetlands and urban flood control.
- Landmark: the first sentence ends after heavy rainfall.
- Landmark: the final sentence turns to maintenance costs and biodiversity.
- Strong anchors: wetlands, drainage systems, biodiversity.
- Likely weak words: to, from, or and may be reduced in fluent speech.
- Cursor: ready at Urban, not floating halfway down the paragraph.
Your map can differ and still be useful. The purpose is orientation, not predicting which words will be mismatches.
A compact practice plan that transfers to unfamiliar speech
Do not practise until you memorize where the mistakes are. That trains memory, not Highlight Incorrect Words. Rotate speakers, accents, and academic topics, and track the first place your process breaks: preview, alignment, decision, or recovery.
| Session | Main drill | What to measure |
|---|---|---|
| Tracking | Use a short lawful audio clip with an accurate transcript. Follow once at normal speed when possible. Do not create mismatches yet; train eyes, ears, and cursor to stay together. | Where did your moving window first drift away from the speaker? |
| False alarms | Use a new speaker or accent with an accurate transcript. Notice words that sound unfamiliar but are still the same written words. | How many times did “different pronunciation” tempt you to invent “different word”? |
| Mismatch decisions | Ask a study partner to replace a few words in a copy of an accurate transcript with plausible alternatives. Listen once and mark only clear differences, then compare with the untouched transcript. | Separate missed mismatches from false clicks. They are different problems. |
| Mixed transfer | Choose a new academic topic and unfamiliar speaker. Use the 10-second map, one-pass tracking, evidence rule, and forward recovery as one routine. | Which part fails first under novelty: map, move, decide, release, or rejoin? |
Keep normal speed as your destination. If you slow difficult practice audio during review, use the slower pass to hear the boundary more clearly, then return toward normal speed rather than making slow audio your permanent home.
Keep the window moving
You do not need perfect confidence about every sound. You need a disciplined standard for what earns a click and a recovery habit for what does not.
Map the paragraph before audio. Keep the comparison window narrow. Hear, compare, click if the difference is clear, release the decision, and rejoin. When pronunciation is merely unfamiliar, leave it. When a genuine word substitution is clear — even a tiny one — select it. When the evidence is gone, do not hold a trial for a word the recording left behind.
That is the useful kind of confidence in PTE Highlight Incorrect Words: not “I heard everything perfectly,” but “I stayed with the speaker and clicked only what I could defend.”
Sources
- Pearson PTE — PTE Academic & UKVI test format: Listening — current task instructions, recording length, one-playback rule, 10-second preview, cursor guidance, assessed skills, deselection behavior, and per-selection scoring.
- Pearson PTE — PTE Academic Test Taker Score Guide — current question range and scoring row for Highlight Incorrect Words, including partial credit, +1, −1, and the 0 minimum item score.
- Pearson PTE — Updates to PTE Academic — current post-7-August-2025 PTE Academic and PTE Academic UKVI version context.
- Pearson PTE — PTE changes 2025: everything you need to know — dated 10 July 2025; explains that the existing question types kept their task formats, timings, and expected response types when two speaking question types were added.
- Pearson PTE — PTE Academic scoring — current score-report terminology and access to Pearson's current Test Taker Score Guide.