In Pearson’s live test a person reviews the Content score on seven long tasks. That is Pearson’s process only: EdKnot practice scoring is entirely automated, and no person ever sees or scores a practice response.
Glossary term
Human Review
Human review is Pearson’s expert check of the Content score on seven PTE tasks before the final score is released: Describe Image, Retell Lecture, Respond to a Situation, Summarize Group Discussion, Summarize Written Text, Write Essay and Summarize Spoken Text. Write Essay also gets human review of Development, Structure and Coherence and General Linguistic Range. If the human and the AI disagree, a second human decides.
Not visible on the report. Pearson’s score guide states which tasks include a human expert review and that all other tasks are scored automatically. On EdKnot, every practice and mock response is scored automatically with no human involvement at any stage.
What the reviewer checks
The reviewer scores Content against the same descriptors the AI uses, and those descriptors are written for a human reader: “a listener could build a complete mental picture”, “paraphrasing the main ideas seamlessly in own words”, “repetition of language from the prompt without reformulation”. The review is what makes those judgements reliable.
It also explains why certain tactics fail. A Describe Image template with empty slots reads to a person as “a list of points without description”, which is the Content 1 descriptor. A memorised essay introduction reads as “generic statements” and “over-reliance on repeating language from the prompt”, the Content 2 descriptor. The Write Email Content 0 descriptor names “significant amounts of pre-prepared or memorised material” outright.
How this shapes practice
Write and speak for a reader. On the long tasks, say what is there in your own words with one specific detail per point, and use templates only as the order of moves, filling every slot with a real feature before you begin. On the short tasks, which are scored automatically, mechanical accuracy is what counts: every word in Read Aloud, every word in sequence in Repeat Sentence, every word spelled correctly in dictation.
Human review is not a rescore. It happens on every test before the result is issued. A rescore is the separate paid option to have the whole test re-marked afterwards.
The numbers that matter
| Tasks with human review of Content | Describe Image, Retell Lecture, Respond to a Situation, Summarize Group Discussion, Summarize Written Text, Write Essay, Summarize Spoken Text |
|---|---|
| Write Essay additional review | Development, Structure and Coherence; General Linguistic Range |
| Disagreement | A second human makes the final judgement |
| All other items | Scored by AI only |
| Separate from a rescore | A rescore is a paid, whole-test re-mark you request after results; human review happens before results |
Worked example
“Clouds. Water vapor. Cooling. Condensation nuclei. Dust, salt, pollution. Clouds form.” Every key term from the lecture is present and a keyword count would look healthy. A reviewer scores it against “isolated words and phrases from the lecture with no adequate context or meaning”: Content 1.
What it shows: The review is why the descriptors, not the keywords, are what to write for.
Frequently asked questions
Is PTE scored by humans or by computer?
Every item is scored by AI. On seven long tasks a human expert also reviews the Content score before the result is final, and on Write Essay two further traits. Human raters also train the scoring engine by rating responses to every speaking and writing item.
Does human review slow down results?
No. Results typically arrive within 48 hours and never later than five business days, with the review included.
Can I request a human review of my score?
That is a rescore: a paid re-mark of the whole test requested within 14 days of the result, with the fee refunded if any score changes. Human review of Content on the seven tasks happens automatically before every result.
Related terms
More PTE terms
AppropriacyCEFRCLB (Canadian Language Benchmarks)Communicative SkillsCompetent, Proficient and Superior EnglishConcordance (PTE to IELTS)Development, Structure and CoherenceEmail ConventionsEnhanced PTE Academic (August 2025)Erasable NoteboardExpress EntryFormGeneral Linguistic RangeGrammarGSE (Global Scale of English)Integrated TaskMock TestmyPTENegative MarkingOne Skill RetakeOral FluencyOverall ScorePartial CreditPersonal IntroductionPronunciationPTE CorePTE HomeReschedule and Cancellation PolicyScore BandScore Report CodeScore ValiditySELT (Secure English Language Test)Sending ScoresShared Clock (Reading and Listening)Skills ProfileSpellingThree-Second Silence RuleUKVIURN (Unique Reference Number)Vocabulary