TARA批判性思维 22题40分钟 七种题型
TARA 的批判性思维模块只有 22 道选择题，考试时间 40 分钟，平均每道题不到两分钟。很多学生以为这一模块靠语感和常识，题海刷一遍就能应付。它实际考的是论证结构，也就是理由、结论与假设三者之间的关系。UAT-UK 官网的 TARA Content Specification 适用于 2026 年 10 月与 2027 年 1 月场次，把这一模块写成 reasoning using everyday written language，即用日常书面语言做推理。牛津与 UCL 用这门笔试筛选申请人，考的是你能不能在有限时间里把一段话里的论证拆开。
Critical Thinking is the first of the three TARA modules and the one candidates find hardest to pace. It contains 22 five-option multiple-choice questions, all built from everyday written language rather than any school subject. UAT-UK states that you should treat the reasons given in a passage as true, so the task is never to fact-check the topic; it is to judge whether those reasons carry the conclusion. Each question is worth one mark and no marks are deducted for a wrong answer, so every question is worth an attempt.
批判性思维考的是论证而不是学科常识
UAT-UK 官网在 Content Specification 里把批判性思维定义为用日常书面语言做推理。历史学家用论证解释与解读过去的事件。科学家用论证评估实验得到的数据。这一模块的题目只借这两种场景，不涉及任何课程内容。很多学生以为读懂英文段落就能做对题，其实题干里的每一句都可能只是在给理由。真正被问的是这段话想让你接受什么。判断标准只有一条，就是结论能不能从这些理由里推出来。
The specification describes Critical Thinking as reasoning in everyday written language, and the question guide repeats that arguments are the unit of assessment. Historians argue to interpret past events; scientists argue to evaluate the evidence from an experiment, and the TARA borrows both habits without borrowing any curriculum content. Candidates often assume that fluent reading is enough, yet a well-written passage can still carry a weak inference. The only question that matters is whether the stated reasons force the conclusion to be accepted.
一段论证由理由 结论与假设三个部件组成
任何一段论证都由三样东西组成：理由、结论与假设。理由是被摆出来支持结论的陈述，结论是被支持的判断，假设是没有写出来但必须成立的那一步。UAT-UK 的试题指南举过一个很短的例子：她没什么机会了，北极熊就在她身后。这句话里没有人写出北极熊危险，可结论依赖这个默认前提。找假设的题因此要先找到结论，再问这段话缺了哪一块。常见的错误是直接把选项里最像常识的一句选上去，而它可能只是背景信息。
Every argument in this module can be broken into reasons, a conclusion and at least one assumption. The reasons are the statements offered as grounds; the conclusion is what those grounds are meant to establish; the assumption is the unstated step that has to hold for the move to work. UAT-UK uses a two-line example about a polar bear to show how much weight an unwritten belief can carry. When a question asks for an assumption, the fastest route is to fix the conclusion first and then ask what the passage has silently taken for granted.
结论的位置不固定 中间结论容易冒充主结论
结论不一定出现在段落末尾，它可能在开头，也可能夹在中间。试题指南的第一个官方例子是一段关于英国皇家学会投资化石燃料的材料。那段话先给出三个理由，再得出一个中间结论，说这种做法与学会自己的立场相矛盾，最后才落到主结论：学会应当撤出这些投资。中间结论只是支撑主结论的一环，不是这段话真正想让你接受的东西。很多学生以为语气最强的那一句就是主结论，其实要看它是否被后文继续支持。错得最多的是把中间结论当成答案。
The guide's first worked example shows that a conclusion can sit at the end of a passage while an earlier judgement plays only a supporting role. In that example the argument lists three reasons about an organisation's investments, then reaches an intermediate conclusion about inconsistency, and only afterwards states the recommendation that forms the main conclusion. Options in this question type are usually built from exactly those intermediate steps. Candidates who pick the most emphatic sentence rather than the one the rest of the passage supports tend to lose the mark here.
官方把批判性思维分成七种题型
TARA Question Guide 把批判性思维拆成七种题型。它们分别是找出主结论、从材料推出结论、找出假设、判断新增证据的影响、指出推理错误、匹配论证结构与套用一条原则。每一道题只有五个选项，也只有一个正确答案。七种题型对应七种不同的动作，用一种方法套所有题会很吃亏。找出假设要你补上缺失的那一环，指出推理错误却要你说出这一环为什么断掉了。两个动作的方向正好相反。
The question guide names seven Critical Thinking types, and each one asks for a different operation on the same material. Identifying the main conclusion and drawing a conclusion both concern conclusions, but only the first expects you to find one already in the passage; identifying an assumption asks you to supply a missing bridge, while detecting reasoning errors asks you to explain how that bridge fails. Matching arguments ignores topic and compares structure, and applying principles asks you to state a general rule and then find the case it governs. Treating all seven as one generic skill is the most common waste of preparation time.
第一种题型 找出主结论
找出主结论的题给你一段话，问哪一句最准确地表达了主结论。做法是先问这段话想让你接受什么，再回头检查其余句子是不是都在为它服务。被 so 或 therefore 引出的句子往往是结论，但没有这些词也可能是结论。结论同样可能排在开头。常见的错误是把一个理由当成结论，因为理由常常读起来更像事实。真正的结论通常带有一个动作或者一个判断，而理由多数只是对现状的描述。
In this type, five candidate statements are offered and exactly one of them is the main conclusion. The reliable route is to decide first what the passage is trying to persuade you to accept, then check whether the remaining sentences do any work other than supporting it. Words such as so, therefore and thus are useful signals, yet the guide is explicit that a conclusion may appear with no signal at all, at the beginning or in the middle of a passage. Options that read like plain facts are usually reasons dressed up as conclusions.
第二种题型 从材料里推出结论
这一类的特别之处在于，结论根本不在材料里。题目给你一组事实，让你判断哪一句是可以从中推出的结论。答题时要克制，不要推出材料没有支持的东西。官方那道关于狼与羊群的材料给了五个事实，包括补偿金额高于牲畜本身的价值，以及狼与狗的攻击痕迹很难区分。正确的选项只说这类索赔应当谨慎对待。说得更绝对、更像道德判断的选项都超出了材料的支撑。
Drawing a conclusion is the mirror image of the previous type: the passage contains facts, and the conclusion has to be chosen from the options rather than located in the text. The discipline here is to claim no more than the passage supports, which means rejecting any option that turns a tendency into a certainty or a factual pattern into a moral judgement. In the guide's worked example about wolves, compensation and the difficulty of telling wolf attacks from dog attacks, the credited answer is deliberately hedged. Every distractor adds a judgement that the facts cannot carry.
第三种题型 找出没有写出来的假设
假设题的常见问法是哪一个选项属于这段论证的隐含前提。假设一定没有被写出来，写出来的就是理由。做法是先确定结论，再问这个结论要站得住，必须默认什么。官方例子涉及英格兰的宗教学校，材料给出家长第一志愿填非宗教学校却被分配宗教学校的比例，结论说很多孩子被强加了宗教。这一步要成立，就要默认那些家长填非宗教学校，正是因为学校不带宗教色彩。五个选项里通常有三四个本身可能为真，却对结论并不必要。
Assumption questions ask for the necessary unstated bridge, and the word necessary is where most marks are lost. A statement can be true, relevant and still not required by the argument, so the test is whether removing it would break the inference. In the guide's example about faith school admissions, the conclusion only follows if the parents who ranked a non-faith school first did so because it was non-faith. The guide marks the other four options as possible but unnecessary, which is exactly the trap this type is built around.
第四种题型 判断新增证据的影响
这一类题目问你哪一条新信息会加强或者削弱原来的论证。方法是先把结论与核心依据分开，再看新信息作用在哪一块上。作用在结论上却绕开依据的选项通常没有用。官方例子讨论感冒时吃止痛药，论证说止痛药会降低体温，而体温是对抗病毒的手段，所以吃药可能加重传播。能让这条推理更紧的选项，是那个说明服药者更早带病返工的信息。只谈止痛药别的副作用的选项，都打不到这条推理上。
Here a new fact is offered and you have to judge what it does to the argument: strengthen it, weaken it, or leave it untouched. The efficient method is to separate the conclusion from the reason it rests on and then locate the new fact on that line of inference. A distractor often raises a real concern about the topic, such as a different side effect of a medicine, without touching the link the argument actually depends on. In the guide's example about painkillers and flu transmission, the credited answer strengthens the original reason instead of introducing an independent risk.
第五种题型 指出论证里的推理错误
找错误的题要求你说出结论为什么不能从理由里推出来。题干里的论证往往读起来很顺，问题就出在中间的跳跃上。官方例子把两百年来出生率下降与女性预期寿命上升放在一起，直接得出多生育会让女性更快衰老的结论。这一段的毛病在于把两件并行发生的事当成了因果关系，忽略了收入、饮食与传染病风险这些别的解释。答题时要能说出这条跳跃在哪，而不是简单否定材料谈论的话题。五个选项里最像常识的那一句，常常只是论证从来没有主张过的东西。
A reasoning error question asks you to name the defect in the move from reasons to conclusion. The guide's example uses historical birth rates and life expectancy to show a conclusion built on a bare causal link: two trends moved together, so one was said to cause the other, with income, diet and infectious disease left out of the account. The right answer describes that gap precisely, while the wrong answers usually describe something the argument never claimed or never needed. Rejecting the topic or the writer's values is not the same as finding the flaw.
第六种题型 匹配论证的结构
匹配题要找的是结构相同的论证，不是主题相同的论证。题干与选项可能谈完全不同的领域，只要理由与结论的组织方式一致就算匹配。选项里的句子顺序往往被打乱，用来掩盖这种相似。题干里的论证本身可能是无效的，这并不影响匹配。官方例子里的医院削减赤字那段推理并不成立，正确的选项照样要与它同构。看关系，不看对错，也不看领域。
Matching arguments looks at pattern rather than subject matter, and the correct option may sit in a completely different field from the passage. Statements in the options are often reordered, so the parallel is structural: which claim supports which, and in what role. The guide notes that the argument in the passage can itself be invalid, and that this does not matter, because the question concerns form alone. Candidates who compare topics instead of relationships usually find two plausible options and then choose the wrong one.
第七种题型 套用一条一般原则
套用原则题先给你一条一般规则，再问你哪一个情境受这条规则管辖。做法是把规则读成两句：前提条件是什么，结论要求是什么。然后逐个检查选项是否都满足这两句。容易出问题的地方是规则里带有程度词，比如只要可能就应当，或者除非必要不得。这些词决定了规则适用的边界。把规则读宽或者读窄，都会选到看似相关却不合格的选项。
Applying principles starts from a general rule and asks which situation falls under it. Break the rule into its trigger and its requirement, then test each option against both parts rather than against the general topic. Degree words carry the whole question: a rule phrased as only if, unless or whenever applies far more narrowly or widely than a loose paraphrase would suggest. Reading the rule slightly too broadly is a common way to land on a related but ineligible option.
官方材料怎么用 试题指南与BMAT TSA旧卷
UAT-UK 官网建议先读 Content Specification，再读 TARA Question Guide，然后开始做题。Pearson VUE 页面上有两套官方的练习，界面与真实考试一致，用来熟悉机考流程。官网还写明批判性思维与问题解决这两部分，与 BMAT Section 1 和 TSA 的思维技能考试一脉相承。官网因此直接给出了 BMAT Section 1 的历年试卷。这些旧卷从 2003 年排到 2023 年，每份都配答案。TSA 的旧卷同样列出。用它们练手要注意，它们只是同源材料，不是 TARA 的真题。
UAT-UK's own guidance is to read the specification first, then the question guide, and only then start practising. Two practice tests are hosted on the Pearson VUE landing page and are built on the same software as the live test, which makes them the cheapest way to remove interface surprises. The preparation page also states that the Critical Thinking and Problem Solving components share a common history with BMAT Section 1 and the Thinking Skills Assessment, and it links the archived papers directly. Those archives run from 2003 to 2023 for BMAT with answer keys, and TSA papers are listed alongside them; they train the same skills, but they are not released TARA questions.
考场细节与时间分配的三个硬规则
考试当天有三条硬规则。模块之间单独计时，提前做完的时间不会结转。任何模块都不允许使用计算器与词典。考场会发一本可擦写的草稿册，用来划出理由与结论。批判性思维与问题解决各自单独报告成绩，分数从 1 到 9，保留一位小数，官方明确没有及格线。每道题答错不扣分，所以官方写明所有题都值得作答。22 道题 40 分钟意味着每题约 1 分 49 秒，长材料题往往要两三分钟，短题必须补回来。留出最后两分钟检查空白选项，比反复琢磨一道难题更划算。
Three rules govern the day itself. Each module is timed separately and unused time is not carried forward, so banking minutes from an early finish gains nothing; calculators and dictionaries are banned in every module; and an erasable booklet is provided for your working, which is the natural place to write the conclusion and the reasons down. Critical Thinking and Problem Solving are reported separately on a scale from 1.0 to 9.0 to one decimal place, with no pass or fail. Since no marks are deducted for a wrong answer, the guide tells candidates to attempt every question, which turns pacing into the real constraint: 22 questions in 40 minutes is roughly one minute forty-nine seconds each, and a long passage can easily consume three.
牛津高阶英汉双解词典 第10版 商务印书馆¥168📱 去京东购买京东精选需要一对一辅导？Need One-on-One Tutoring?
A-Level / GCSE / IB / KS2-KS3 全科辅导
16621398022 同微信
Visit aleveler.com
或截图后用「扫一扫 → 相册」选择本图