IB TOK Exhibition 选题 分析框架 知识问题
IB 知识论（TOK）的 Exhibition 展览评估，是整个 IBDP 核心课程里最容易被低估、却最容易拉开差距的一项内部评估。每位学生需要从官方给出的 35 个 IA Prompt 中选择一个，再挑选三件真实世界的展览物（objects），在 950 个英文单词以内完成分析，满分 10 分。这个分数直接决定 TOK 的等级（A 到 E），而 TOK 等级又与 Extended Essay 组合，共同决定核心课程最多 3 分的加分。换句话说，一篇写砸的 Exhibition，可能让你在 45 分总分里白白丢掉 3 分，这往往就是名校申请线上下浮动的全部差距。
The TOK exhibition is one of the most underestimated internal assessments in the IBDP core. Each student selects one of the 35 prescribed IA prompts, chooses three real-world objects, and writes an analysis of no more than 950 words, marked out of 10. That mark determines the TOK grade from A to E, which combines with the Extended Essay to decide up to 3 bonus points. A weak exhibition can quietly cost you 3 marks out of 45, which is often exactly the margin between a strong and a borderline university application.
一、Exhibition 的评分结构与硬性规则
Exhibition 占 TOK 内部评估的全部权重，评分分为两个 criterion：criterion A 考察你对 IA prompt 的理解程度以及展览物的选择是否恰当，占 5 分；criterion B 考察你能否把展览物与知识问题（knowledge question）连起来，展示出对知识如何被生产、验证、传播的洞察，也占 5 分。硬性规则有三条必须记住：第一，三件展览物必须对应同一个 IA prompt，不能各写各的；第二，展览物必须是真实世界中的具体物件，可以是一张新闻照片、一段社交媒体帖子、一件实验室器材、一幅画、一份数据图表，甚至是你每天使用的工具；第三，全文不超过 950 字，超出部分不会被评分，注释与参考文献不计入字数，但展览物说明文字也不应该用来夹带额外分析。
The exhibition carries the full weight of the TOK internal assessment and is marked against two criteria. Criterion A examines your understanding of the selected IA prompt and the appropriateness of your chosen objects, worth 5 marks. Criterion B examines whether you can connect those objects to knowledge questions and demonstrate insight into how knowledge is produced, validated, and shared, also worth 5 marks. Three rules are non-negotiable. All three objects must answer the same IA prompt. Every object must be a specific real-world item, such as a news photograph, a social media post, a piece of laboratory equipment, a painting, a data chart, or an everyday tool. The written analysis must stay within 950 words; anything beyond the limit is not assessed.
二、选题第一步：从 35 个 IA Prompt 中做正确选择
35 个 IA prompt 覆盖知识与技术、知识与语言、知识与政治、知识与宗教、知识与本土社会等核心主题。很多学生的第一个错误，是凭直觉选一个听起来最深刻的 prompt，例如关于知识是否可以独立于文化存在之类的大问题，然后发现根本找不到具体物件。正确的顺序是反过来：先回想你手边真实、具体、有出处的物件，再倒推哪个 prompt 能自然覆盖这三件物件。判断一个 prompt 是否适合自己，可以用三个测试：第一，它是否能被具体物件直接回答，而不是只能靠抽象论证；第二，它是否自带知识争议，也就是不同的人对同一件事会有不同的知识主张；第三，你是否能在 950 字里为三件物件各写出一个有深度的分析段落，而不是三句重复的话。
The 35 prescribed prompts span knowledge and technology, knowledge and language, knowledge and politics, knowledge and religion, and knowledge and indigenous societies. The most common first mistake is choosing whichever prompt sounds most profound, then discovering that no concrete object fits it. Work in the opposite direction: start from real, specific, dated objects you already have access to, then work backwards to the prompt that naturally covers all three. Three tests help you judge fit. Can the prompt be answered through concrete objects rather than abstract argument alone? Does it contain genuine knowledge controversy, where different people would hold different knowledge claims about the same thing? Can you write one substantive analytical paragraph for each of the three objects within the word limit, rather than three variations of the same sentence?
三、展览物的选择标准：具体、真实、可追溯
展览物的黄金标准是 specific real-world context，也就是每一件物件都要能说出时间、地点、来源和它为什么在你手中。比较两个例子就能看出差别。写知识如何被权威塑造时，展览物写社会课本是一个失败的选择，因为它太泛、没有具体语境；换成 2024 年某地教育部门修订后的历史教科书某一页，并附上修订前后的对照，就立刻满足了具体性要求。同样，写知识与技术时，展览物写人工智能是失败的，换成你手机上某个作文辅助应用的界面截图并注明该应用宣称的功能，就成立了。建议三件物件在媒体类型上略有差异，例如一件图像、一件文字、一件器物或数据，这样论证层次会更丰富，也更容易展示不同类型的知识生产机制。
The golden standard for an object is a specific real-world context, meaning you can state its time, place, source, and the reason it is in your hands. Two comparisons show the difference. If you are writing about how authority shapes knowledge, a social studies textbook is a weak object because it is too general; a photograph of a specific page from a textbook revised by an education authority in 2024, paired with the pre-revision version, immediately satisfies the specificity requirement. If you are writing about knowledge and technology, artificial intelligence as a topic is weak; a screenshot of one essay-assistance application on your own phone, noting exactly what it claims to do, is a valid object. Choose three objects that differ in medium, perhaps one image, one text, and one artefact or dataset, so your argument can show several mechanisms of knowledge production rather than one repeated pattern.
四、分析框架：四步法把 5 分变成稳拿的分
一个可以直接套用的四步框架是：语境、知识问题、主张与证据、论证与反思。第一步用一两句话交代物件的真实语境，说明它是谁在什么时候为什么创造的。第二步从 prompt 出发提出一个明确的知识问题，例如在什么条件下专家的证词可以被视为知识。第三步给出一个知识主张，并用物件的具体特征作为证据，一定要引用物件上真实存在的细节，例如图表中的数值、照片中的构图、文本中的措辞。第四步写论证与反思，包括这个主张的适用范围、可能出现的反例、以及不同知识领域（例如自然科学与人文）是否会得出不同结论。四步之中，评分者最看重的是第三步与第四步，因为描述任何人都能做，只有把物件变成论据才是真正的分析。
A four-step framework you can apply directly is context, knowledge question, claim and evidence, then justification and reflection. Step one sets the real-world context in one or two sentences, stating who created the object, when, and why. Step two states an explicit knowledge question derived from the prompt, for example under what conditions expert testimony counts as knowledge. Step three makes a knowledge claim and supports it with concrete features of the object, quoting real details such as a value in a chart, a compositional choice in a photograph, or specific wording in a text. Step four justifies and reflects, discussing the scope of the claim, possible counterexamples, and whether different areas of knowledge such as the natural sciences and the human sciences would reach different conclusions. Examiners weigh steps three and four most heavily, because description is easy and turning an object into evidence is what analysis actually means.
五、三件展览物之间的逻辑关系
三件展览物不必属于同一类，但必须能构成一个完整的论证弧线。常见的三种有效结构是：递进式，三件物件展示同一知识机制在不同场景下的加深，例如从个人经验到公共媒体再到制度性记录；对照式，两件物件呈现相反的知识主张，第三件解释为什么两者可以并存；层级式，三件物件分别对应知识的生产、传播与接受三个环节。需要避免的是三件物件论证同一个点，这会让评分者认为你的分析缺乏广度。写作顺序上，建议先确认三件物件的排列顺序，再给每一件写一段，最后回头检查三段之间是否有明确的连接词，例如相比之下、进一步说、因此，这些连接词能让论证结构在 950 字里依然清晰可见。
The three objects need not belong to the same category, but together they must form a complete argument arc. Three structures work well. A progressive structure shows the same knowledge mechanism deepening across contexts, from personal experience to public media to institutional record. A contrastive structure sets two objects against each other on a knowledge claim and uses the third to explain why both can coexist. A layered structure maps the objects onto the production, dissemination, and reception of knowledge. The failure mode to avoid is three objects proving the same point, which examiners read as narrow analysis. Before writing, fix the order of the three objects, then draft one paragraph each, and finally check that clear connectives link the paragraphs, such as in contrast, further, and therefore, so that the argument structure remains visible within the word limit.
六、常见失分点与高分技巧
评分报告反复提到的失分点有五个。第一，只描述不分析，整段在讲物件是什么，却没有回答知识问题。第二，物件过于笼统，例如笼统地写科学、写社交媒体，没有具体到某一篇帖子或某一项研究。第三，混淆知识与信念，把个人观点当成知识主张而不讨论其验证条件。第四，三件物件之间没有逻辑联系，读起来像三篇独立的短文。第五，字数分配失衡，第一件物件写了一整段半，第三件只剩两句话。针对这五点，对应的技巧是：每一段都用我们之所以认为这个主张成立，是因为这样的句式把物件与知识主张绑在一起；准备一个细节清单，写下每件物件上真正看得到的元素；提前规划每件物件约 250 字的配额；写完通读一遍，把所有可以删掉的形容词换成更精确的名词。
Examiner reports repeatedly flag five weaknesses. First, description without analysis, where the paragraph explains what the object is but never answers the knowledge question. Second, objects that are too general, such as science or social media in the abstract rather than one specific post or study. Third, confusing knowledge with belief, treating a personal opinion as a knowledge claim without discussing the conditions under which it could be validated. Fourth, no logical link between the three objects, so the piece reads as three separate mini-essays. Fifth, imbalanced word allocation, where the first object takes a paragraph and a half and the third gets two sentences. The countermeasures follow directly. Bind every paragraph to a knowledge claim with a sentence such as we treat this claim as knowledge because. Keep a detail list of what is actually visible in each object. Budget roughly 250 words per object. Then reread and replace vague adjectives with precise nouns.
七、一个可以模仿的高分段落示范
以下段落示范四步框架如何落地。我选择的展览物是一份 2023 年公开发布的空气质量监测报告首页，它来自一个城市环境部门，用于每周向公众通报空气质量。由此我提出知识问题：当数据由发布者自身机构采集时，公众如何在信任与怀疑之间作出判断。本物件同时提供支持与限制：报告公开了监测站位置与采样频次，这些信息让数据可以被独立复核，因此具备知识的可检验特征；但报告并未公开原始逐时数据，公众只能接受汇总后的结论，这说明知识传播的过程本身就包含了取舍。进一步看，若同一时段的独立监测机构给出不同数值，公众面临的不是无知，而是相互竞争的知识主张，此时判断标准应从数据是否存在转向数据由谁采集、依据何种方法。由此可见，在环境科学这类高度依赖机构数据的领域中，知识的社会信任结构与实证证据同样重要。
This paragraph shows the framework in action. The object is the front page of an air quality monitoring report published in 2023 by a city environmental department and released weekly to the public. The knowledge question that follows is how the public judges between trust and doubt when data is gathered by the very body that publishes it. The object supplies both support and limitation. The report discloses station locations and sampling frequency, which allows independent verification and therefore gives the data a testable character. Yet it does not release raw hourly data, so the public can only accept the summarised conclusion, showing that dissemination itself involves selection. If an independent monitoring body reports different values for the same period, the public faces not ignorance but competing knowledge claims, and the criterion shifts from whether data exists to who gathered it and by what method. In a field as dependent on institutional data as environmental science, the social structure of trust matters as much as empirical evidence.
八、四周备考时间表
第一周完成两个动作：通读 35 个 IA prompt，为其中三到四个写下你能想到的具体物件，然后从中锁定一个 prompt。第二周确定三件展览物，收集每件物件的来源信息与细节清单，确认它们真的构成一条论证弧线。第三周写第一稿，不要一开始就追求 950 字，先把三段核心论证写清楚，再补语境与反思。第四周做两件事：把草稿压缩到字数上限以内，并请老师或同学只看一句话，也就是让他们读完每一段后说出这一段的知识主张是什么，如果他们说不出来，那一段就需要重写。最后检查注释格式与展览物图片的清晰度，因为展览物呈现方式本身也属于评估的一部分。
Week one has two tasks. Read all 35 IA prompts, jot down concrete objects you can actually access for three or four of them, then commit to one. Week two confirms the three objects, gathers source information and a detail list for each, and verifies that together they form one argument arc. Week three produces a first draft. Do not chase the word limit immediately; write the three core analytical paragraphs clearly first, then add context and reflection. Week four has two tasks. Compress the draft within the limit, and run a single-sentence test with a teacher or classmate, asking them to state the knowledge claim of each paragraph after reading it. If they cannot, that paragraph needs rewriting. Finally, check referencing and the clarity of your object images, because presentation of the objects is itself part of what is assessed.
把 Exhibition 看成一次受限条件下的论证训练，而不是一次知识展示，你的写作重心就会自然从这件东西是什么，转向我们凭什么把它当作知识。前者只能拿到描述分，后者才能真正触及 criterion B 想要看到的洞察。
Treat the exhibition as training in arguing under constraints rather than a showcase of what you know, and your focus will shift from what this thing is to why we count it as knowledge. The first earns description marks; the second reaches the insight that criterion B rewards.
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，TOK 与 EE 专项指导
或截图后用「扫一扫 → 相册」选择本图