GRE 全真模擬考 #2：科學與科技
難度：中等 | 主題：科學、科技 | 出題老師：威威老師
考試總覽
| 部分 | 題型 | 題數 | 建議時間 |
|---|---|---|---|
| Verbal Reasoning 1 | RC + TC + SE | 20 題 | 30 分鐘 |
| Verbal Reasoning 2 | RC + TC + SE | 20 題 | 30 分鐘 |
| Quantitative Reasoning | 數學英文題 | 10 題 | 15 分鐘 |
| Analytical Writing | Issue + Argument | 2 篇 | 各 30 分鐘 |
SECTION 1：VERBAL REASONING（20 題）
PART A：READING COMPREHENSION（10 題）
Long Passage（Questions 1-4）
Read the following passage and answer questions 1-4.
The discovery of CRISPR-Cas9 as a gene-editing tool in 2012 was heralded as a revolution in molecular biology, and for good reason. Unlike previous gene-editing technologies, which were cumbersome, expensive, and imprecise, CRISPR allows researchers to target specific DNA sequences with unprecedented accuracy and at a fraction of the cost. Within a decade, the technique has been deployed in agriculture to engineer drought-resistant crops, in medicine to develop potential cures for genetic disorders such as sickle cell anemia, and in basic research to unravel the functions of previously mysterious genes. The democratization of gene editing—CRISPR kits can now be ordered online for a few hundred dollars—has transformed what was once the exclusive domain of well-funded laboratories into a technology accessible to community college biology classes and even amateur “biohackers.”
Yet this democratization is precisely what makes many scientists and ethicists uneasy. The ease and low cost of CRISPR mean that gene editing could be performed without the institutional oversight that traditionally accompanies high-stakes biological research. In 2018, a Chinese researcher shocked the international community by announcing that he had used CRISPR to edit the genomes of twin embryos, ostensibly to confer resistance to HIV. The experiment, widely condemned as premature and unethical, demonstrated that the gap between what CRISPR makes technically possible and what society has agreed is ethically permissible is dangerously wide. The episode also exposed the limitations of existing regulatory frameworks, which vary dramatically across national boundaries.
More subtle but equally troubling are the ecological implications of CRISPR-modified organisms released into the wild. Gene drives—CRISPR-based systems designed to spread a particular genetic trait through an entire population—hold promise for eradicating mosquito-borne diseases like malaria. But the same technology, if deployed carelessly or maliciously, could irreversibly alter ecosystems in ways that are impossible to predict or reverse. Unlike chemical pollutants that degrade over time, a gene drive is designed to perpetuate itself. The very feature that makes it powerful—self-propagation—makes it uniquely dangerous.
1. The primary purpose of the passage is to:
(A) Celebrate CRISPR’s transformative impact on molecular biology and medicine (B) Argue that CRISPR technology should be banned entirely due to its risks (C) Present both the remarkable potential and the serious ethical and ecological concerns raised by CRISPR (D) Defend the Chinese researcher’s embryo-editing experiment as necessary scientific progress (E) Explain in technical detail how CRISPR-Cas9 functions at the molecular level
2. According to the passage, which of the following is a feature of gene drives that makes them “uniquely dangerous”?
(A) They cannot be used to target disease-carrying organisms (B) They are designed to spread and perpetuate themselves through populations (C) They are more expensive than traditional forms of pest control (D) They require international cooperation to implement effectively (E) They degrade over time and become less effective
3. The author mentions the Chinese researcher’s embryo-editing experiment primarily to:
(A) Suggest that Chinese scientists are more innovative than their Western counterparts (B) Illustrate the gap between technical capability and ethical consensus (C) Provide evidence that CRISPR is safe for use in human embryos (D) Argue that national regulatory frameworks are sufficient to govern gene editing (E) Demonstrate that HIV resistance is the most promising application of CRISPR
4. Select the sentence in the passage that best describes what the author considers the paradoxical nature of gene drives.
Short Passage A（Questions 5-6）
Read the following passage and answer questions 5-6.
The Turing test, proposed by Alan Turing in 1950, has been both celebrated as a foundational concept in artificial intelligence and criticized as a deeply flawed metric for machine intelligence. The test’s premise—that a machine can be considered intelligent if it can converse in such a way that a human interlocutor cannot reliably distinguish it from another human—privileges linguistic mimicry over genuine understanding. A system might pass the Turing test by deploying sophisticated pattern-matching algorithms to generate plausible conversational responses without possessing anything resembling comprehension, intentionality, or consciousness. The philosopher John Searle’s famous “Chinese Room” thought experiment was designed to expose precisely this limitation: a person who follows rules to manipulate Chinese symbols without understanding Chinese can, in principle, produce outputs indistinguishable from those of a native speaker. The Turing test, Searle argued, confuses simulation with the real thing.
5. The author mentions Searle’s “Chinese Room” primarily to:
(A) Defend the Turing test against philosophical criticism (B) Demonstrate that machines will never be able to converse in Chinese (C) Support the argument that passing the Turing test does not imply genuine understanding (D) Prove that linguistic mimicry requires consciousness (E) Illustrate how human translators differ from machine translation systems
6. It can be inferred that the author would most likely agree with which of the following?
(A) The Turing test is the definitive measure of machine intelligence. (B) Simulation of intelligent behavior is equivalent to genuine intelligence. (C) The Turing test’s focus on linguistic output may not adequately capture what intelligence truly is. (D) John Searle’s argument has been definitively refuted by modern AI research. (E) Machines that pass the Turing test necessarily possess consciousness.
Short Passage B（Questions 7-8）
Read the following passage and answer questions 7-8.
The concept of the “technological fix”—the idea that complex social, political, or environmental problems can be solved primarily through technological innovation—has been a recurring theme in American culture since the nineteenth century. From the belief that nuclear power would provide “electricity too cheap to meter” to contemporary hopes that carbon capture will solve climate change without requiring changes in consumption patterns, the technological fix promises progress without sacrifice. Historians of technology have noted a consistent pattern: technological fixes that ignore the social and political dimensions of the problems they address tend to generate new, often more intractable problems. The automobile solved the problem of urban horse manure but created smog, suburban sprawl, and carbon emissions. The lesson is not that technology is useless but that technological solutions are most effective when embedded within broader social, political, and behavioral transformations.
7. The “pattern” identified by historians of technology is best described as:
(A) Technological innovations regularly fail to deliver any benefits whatsoever (B) Solutions that ignore social and political contexts tend to create new problems (C) Nuclear power and carbon capture are fundamentally the same technology (D) American culture has consistently rejected technological innovation (E) Environmental problems can only be solved through behavioral change alone
8. The passage mentions “urban horse manure” primarily to:
(A) Suggest that nineteenth-century cities were cleaner than modern ones (B) Illustrate that automobiles were a wholly negative development (C) Provide a historical example of how technological solutions can create new problems while solving old ones (D) Argue that horses should be reintroduced as urban transportation (E) Demonstrate that manure was a more serious problem than climate change
Short Passage C（Questions 9-10）
Read the following passage. For question 9, consider each answer choice separately and select ALL that apply. For question 10, select one answer choice.
Climate modeling has advanced dramatically since the first general circulation models of the 1960s. Modern Earth system models incorporate atmospheric chemistry, ocean circulation, ice sheet dynamics, and terrestrial ecosystems into integrated frameworks that run on supercomputers. Yet for all this sophistication, a fundamental tension persists: the phenomena that matter most to policymakers—regional precipitation changes, hurricane frequency, local temperature extremes—are precisely those that climate models are least reliable at predicting. Global mean temperature projections have proven remarkably accurate over multi-decadal timescales, but downscaling these global trends to actionable local predictions remains an enormous scientific challenge. This “scale gap” between what models can do well and what decision-makers need has become one of the central preoccupations of climate science, driving investments in high-resolution regional modeling and in statistical techniques for extracting local information from coarse-grained global simulations.
9. Which of the following can be inferred from the passage about modern climate models? Select ALL that apply.
[A] They are less accurate than the first models developed in the 1960s. [B] They incorporate multiple interacting Earth systems into integrated simulations. [C] They are more reliable for predicting global mean temperatures than for predicting local precipitation patterns.
10. The phrase “scale gap” refers to the:
(A) Difference in funding between global and regional climate research (B) Discrepancy between model accuracy for global trends and the local predictions policymakers need (C) Growing divide between climate scientists and the general public (D) Time lag between model development and practical implementation (E) Inequality between wealthy and developing nations in access to climate modeling technology
PART B：TEXT COMPLETION（6 題）
11. (One Blank) The physicist’s later work was ______ by her earlier discoveries; having set such a high standard with her groundbreaking research in quantum mechanics, everything she published afterward seemed, by comparison, merely competent.
(A) enhanced (B) overshadowed (C) contradicted (D) corroborated (E) anticipated
12. (One Blank) The software company’s claim that its new algorithm was “revolutionary” proved to be ______; independent testing revealed that its performance was marginally better than existing solutions at best.
(A) prescient (B) hyperbolic (C) modest (D) vindicated (E) unassuming
13. (Two Blanks) The researchers’ hypothesis was (i) ______ by their experimental data, but they cautioned that the sample size was too small to draw (ii) ______ conclusions.
| Blank (i) | Blank (ii) |
|---|---|
| (A) refuted | (D) tentative |
| (B) bolstered | (E) definitive |
| (C) complicated | (F) intriguing |
14. (Two Blanks) The renowned biologist was known for her (i) ______ approach to peer review, submitting colleagues’ manuscripts to a level of (ii) ______ that, while scientifically valuable, could strain professional relationships.
| Blank (i) | Blank (ii) |
|---|---|
| (A) perfunctory | (D) scrutiny |
| (B) exacting | (E) indifference |
| (C) indulgent | (F) encouragement |
15. (Three Blanks) Advances in machine learning have been so (i) ______ that even experts struggle to keep pace; algorithms that were considered (ii) ______ just five years ago are now regarded as (iii) ______, useful primarily as teaching examples for introductory courses.
| Blank (i) | Blank (ii) | Blank (iii) |
|---|---|---|
| (A) incremental | (D) heretical | (G) indispensable |
| (B) rapid | (E) state-of-the-art | (H) obsolete |
| (C) predictable | (F) fallacious | (I) controversial |
16. (Three Blanks) The historian of science argued that the (i) ______ of the telescope was not, as commonly assumed, a sudden (ii) ______ but rather a gradual process of refinement that depended as much on advances in glassmaking and lens-grinding as on (iii) ______ insight.
| Blank (i) | Blank (ii) | Blank (iii) |
|---|---|---|
| (A) obsolescence | (D) catastrophe | (G) theoretical |
| (B) invention | (E) breakthrough | (H) commercial |
| (C) calibration | (F) impediment | (I) accidental |
PART C：SENTENCE EQUIVALENCE（4 題）
17. The study’s conclusions, though ______ by a majority of researchers in the field, were sharply criticized by a vocal minority who questioned the statistical methods employed.
(A) repudiated (B) endorsed (C) conceived (D) embraced (E) fabricated (F) disputed
18. The engineer’s design was praised for its ______: it achieved the same functionality as the previous model with half the components and a third of the cost.
(A) complexity (B) elegance (C) redundancy (D) simplicity (E) opacity (F) extravagance
19. The science writer’s latest book attempts to make quantum physics ______ to general readers, using vivid analogies and minimal technical jargon to convey concepts that are notoriously difficult to grasp.
(A) perplexing (B) accessible (C) inscrutable (D) unintelligible (E) comprehensible (F) esoteric
20. The panel of experts was ______ in its assessment of the new drug, emphasizing that while early results are promising, much larger clinical trials are needed before any firm conclusions can be drawn.
(A) sanguine (B) categorical (C) unequivocal (D) circumspect (E) dogmatic (F) cautious
SECTION 2：VERBAL REASONING（20 題）
PART A：READING COMPREHENSION（8 題）
Long Passage（Questions 1-3）
Read the following passage and answer questions 1-3.
The relationship between scientific progress and government funding is more complex than the simple narrative of “fund science and prosperity follows” would suggest. The Manhattan Project and the Apollo program loom large in the public imagination as demonstrations that massive state investment can achieve seemingly impossible technological goals. These successes gave rise to what might be called the “Apollo model” of science policy: identify a clear objective, allocate substantial resources, and assemble the best minds to pursue it. This model has been invoked, with varying degrees of appropriateness, to justify everything from the War on Cancer to contemporary proposals for a “moonshot” approach to climate change.
But the Apollo model has significant limitations, particularly when applied to problems that lack the clear definition and linear structure of putting a human on the moon. Cancer, it turns out, is not one disease but hundreds, each with its own molecular logic, and the “War on Cancer” did not produce a single decisive victory so much as a long series of incremental advances in understanding, detection, and treatment. The history of postwar science policy is, in many respects, a history of learning this lesson repeatedly: state-directed, mission-oriented research excels at solving well-defined technical problems but is far less effective at generating the kind of fundamental, curiosity-driven discoveries that ultimately enable those technical solutions.
This tension is currently playing out in debates over artificial intelligence research funding. Some policymakers advocate for a “Manhattan Project for AI” that would concentrate resources on achieving specific benchmarks in machine capability. Critics of this approach argue that the most important advances in AI—neural networks, backpropagation, reinforcement learning—emerged not from mission-oriented programs but from decades of exploratory research whose practical applications were not immediately apparent. The risk of an Apollo-style approach to AI is that it would channel resources toward near-term performance metrics at the expense of the foundational research that enables genuine breakthroughs. The challenge for science policy is not to choose between mission-oriented and curiosity-driven research but to design funding ecosystems that sustain a productive interplay between the two.
1. Which of the following best describes the organization of the passage?
(A) A historical narrative followed by a prediction about future scientific developments (B) A critique of a widely accepted model of science funding, supported by historical examples, followed by an application to a contemporary debate (C) A comparison of two competing scientific theories followed by a synthesis that resolves the conflict (D) A defense of government-funded science against recent criticisms, supported by statistical evidence (E) An explanation of how artificial intelligence research has evolved over the past several decades
2. According to the passage, the “War on Cancer” is mentioned as an example of:
(A) A successful application of the Apollo model to medical research (B) A case in which the Apollo model proved less effective than expected due to the nature of the problem (C) The superiority of curiosity-driven research over all forms of mission-oriented funding (D) A scientific initiative that failed entirely to produce any useful results (E) The dangers of excessive government involvement in scientific research
3. Select the sentence that best expresses the author’s proposed solution to the tension between mission-oriented and curiosity-driven research.
Short Passage A（Questions 4-5）
Read the following passage and answer questions 4-5.
The placebo effect has long been treated as a nuisance variable in clinical research—something to be controlled for, subtracted out, eliminated from the “real” signal of a drug’s efficacy. But this framing has come under increasing scrutiny from researchers who argue that the placebo response is itself a legitimate biological phenomenon worthy of study in its own right. Neuroimaging studies have demonstrated that placebo interventions can produce measurable changes in brain activity, particularly in regions associated with pain perception and emotional regulation. Far from being a mere psychological trick, the placebo effect appears to involve genuine neurobiological mechanisms, including the release of endogenous opioids and dopamine. Some researchers have gone so far as to suggest that harnessing these mechanisms—rather than simply controlling for them—could open new therapeutic avenues, particularly for conditions like chronic pain and depression where existing pharmaceutical treatments have significant limitations.
4. The passage suggests that the traditional view of the placebo effect as a “nuisance variable” is:
(A) Entirely justified and should guide future research design (B) An oversimplification that ignores the genuine biological mechanisms involved (C) Supported by recent neuroimaging studies of placebo interventions (D) The only scientifically rigorous way to approach clinical trials (E) Primarily a concern of pharmaceutical companies rather than academic researchers
5. The author mentions “endogenous opioids and dopamine” primarily to:
(A) Argue that placebos are more effective than pharmaceutical painkillers (B) Provide evidence that the placebo response has a genuine neurobiological basis (C) Suggest that chronic pain can only be treated through placebo-based therapies (D) Demonstrate that neuroimaging is an unreliable research technique (E) Criticize researchers who have studied the placebo effect
Short Passage B（Questions 6-8）
Read the following passage. For question 6, consider each answer choice separately and select ALL that apply. For questions 7-8, select one answer choice.
The popular narrative of scientific discovery tends to emphasize the “Eureka moment”—the sudden flash of insight that resolves a long-standing puzzle. Archimedes leaping from his bath, Newton under his apple tree, Kekule dreaming of a snake biting its tail and thereby intuiting the structure of benzene: these stories have become part of our cultural mythology. However, historians and sociologists of science have demonstrated that such narratives are almost always retrospective constructions that compress complex, collaborative, and incremental processes into dramatic individual revelations. The reality of scientific discovery is messier: it involves dead ends, ambiguous data, contested interpretations, and contributions from multiple researchers that are later simplified into a single origin story.
The persistence of the Eureka narrative is not merely a matter of historical inaccuracy; it has practical consequences. By presenting scientific progress as the product of isolated genius rather than sustained collective effort, the narrative discourages the kinds of collaborative, interdisciplinary work that contemporary science increasingly demands. It also contributes to the “impostor phenomenon” experienced by many scientists who measure their own work against these mythologized standards and find themselves wanting. The Eureka story is a compelling drama, but it is a poor model for how science actually advances.
6. The author suggests that Eureka narratives have which of the following consequences? Select ALL that apply.
[A] They undermine collaborative and interdisciplinary approaches to science. [B] They contribute to feelings of inadequacy among practicing scientists. [C] They have been definitively proven to be historically accurate.
7. The author characterizes stories like Archimedes in his bath as:
(A) Completely fabricated with no basis in historical events (B) Retrospective simplifications of more complex processes (C) Essential motivational tools for aspiring scientists (D) The most accurate accounts we have of scientific discovery (E) Evidence that ancient science was more intuitive than modern science
8. According to the passage, the “reality of scientific discovery” includes all of the following EXCEPT:
(A) Dead ends and failed experiments (B) Ambiguous and contested data (C) Contributions from multiple researchers (D) Sudden, isolated flashes of individual genius (E) Processes that are later simplified into origin stories
PART B：TEXT COMPLETION（7 題）
9. (One Blank) The artificial intelligence researcher warned that the technology’s rapid advancement has ______ the development of ethical guidelines; we are deploying systems whose societal implications we have not yet fully understood.
(A) outpaced (B) necessitated (C) forestalled (D) paralleled (E) facilitated
10. (One Blank) The biologist’s theory, initially dismissed as ______, gained credibility as evidence from multiple independent laboratories accumulated in its favor.
(A) prescient (B) heretical (C) conventional (D) orthodox (E) incontrovertible
11. (Two Blanks) The tech entrepreneur’s memoir is both (i) ______ and (ii) ______: it offers genuinely useful advice for aspiring founders while also indulging in self-aggrandizement that undermines its credibility.
| Blank (i) | Blank (ii) |
|---|---|
| (A) vacuous | (D) modest |
| (B) insightful | (E) self-serving |
| (C) tedious | (F) altruistic |
12. (Two Blanks) Although gene therapy was once considered a (i) ______ approach limited to a handful of rare disorders, recent clinical successes have (ii) ______ its potential to treat a much wider range of diseases.
| Blank (i) | Blank (ii) |
|---|---|
| (A) versatile | (D) confirmed |
| (B) niche | (E) restricted |
| (C) universal | (F) broadened |
13. (Two Blanks) The scientist’s public statements were remarkably (i) ______ given the (ii) ______ nature of the evidence; she spoke with an assurance that the data simply did not warrant.
| Blank (i) | Blank (ii) |
|---|---|
| (A) tentative | (D) conclusive |
| (B) emphatic | (E) equivocal |
| (C) equivocal | (F) exhaustive |
14. (Three Blanks) The historian of technology argued that the printing press was not so much an (i) ______ agent of change as a (ii) ______ of existing social forces; its revolutionary effects were (iii) ______ upon the prior emergence of a literate public eager for the texts it produced.
| Blank (i) | Blank (ii) | Blank (iii) |
|---|---|---|
| (A) autonomous | (D) catalyst | (G) contingent |
| (B) insignificant | (E) inhibitor | (H) independent |
| (C) belated | (F) beneficiary | (I) superfluous |
15. (Three Blanks) The neuroscientist’s research was (i) ______ by its reliance on a technique whose (ii) ______ had been questioned by several prominent researchers; by the time she acknowledged these concerns, her findings had already been (iii) ______ into the scientific literature.
| Blank (i) | Blank (ii) | Blank (iii) |
|---|---|---|
| (A) vindicated | (D) precision | (G) incorporated |
| (B) compromised | (E) validity | (H) expunged |
| (C) distinguished | (F) popularity | (I) overlooked |
PART C：SENTENCE EQUIVALENCE（5 題）
16. The scientific paper was written in prose so dense and ______ that even specialists in the field struggled to extract its central argument from the thicket of jargon and subordinate clauses.
(A) lucid (B) convoluted (C) pellucid (D) tortuous (E) elegant (F) succinct
17. The technology company’s promises about its new product were so consistently ______ that industry analysts had learned to discount them by at least fifty percent.
(A) inflated (B) modest (C) exaggerated (D) sober (E) understated (F) conservative
18. Far from being a ______ of computing power, the smartphone represents a remarkable concentration of it; the device in your pocket has more processing capability than the supercomputers of the 1980s.
(A) pinnacle (B) diminution (C) triumph (D) culmination (E) reduction (F) milestone
19. The two research teams, working independently on opposite sides of the world, arrived at ______ conclusions, lending considerable weight to the hypothesis that they had both been testing.
(A) contradictory (B) convergent (C) incompatible (D) congruent (E) preliminary (F) divergent
20. The astronomer’s hypothesis was ______ by the discovery of a new class of celestial objects whose properties matched her predictions with startling precision.
(A) repudiated (B) undermined (C) invalidated (D) corroborated (E) authenticated (F) refuted
SECTION 3：QUANTITATIVE REASONING（10 題）
說明： 以下數學題以英文呈現，測試閱讀英文數學題目的能力。
1. A research grant of $150,000 is divided among three laboratories in the ratio 2:3:5. How much does the laboratory receiving the largest share get?
(A) 45,000 (C) 60,000 (E) $75,000
2. If f(x) = x² - 3x + 2, what is f(-2)?
(A) -8 (B) -4 (C) 0 (D) 8 (E) 12
3. A data center uses 840 kilowatt-hours of electricity per week. If electricity costs $0.15 per kilowatt-hour, what is the annual electricity cost? (1 year = 52 weeks)
(A) 6,720 (C) 7,488 (E) $7,800
4. The standard deviation of a set of 20 measurements is 0. If the sum of all 20 measurements is 200, what is each individual measurement?
(A) 0 (B) 5 (C) 10 (D) 20 (E) Cannot be determined
5. A pharmaceutical company finds that 8% of patients experience a side effect from Drug A, and 12% experience a side effect from Drug B. If the events are independent, what is the probability that a patient taking both drugs experiences at least one side effect?
(A) 0.0096 (B) 0.04 (C) 0.1904 (D) 0.20 (E) 0.96
6. A cylindrical water tank has a radius of 3 meters and a height of 5 meters. What is the volume of the tank in cubic meters? (Use π = 3.14)
(A) 47.1 (B) 94.2 (C) 141.3 (D) 282.6 (E) 423.9
7. A computer processes 2.4 × 10⁶ instructions per second. How many instructions does it process in 2.5 × 10³ seconds?
(A) 4.8 × 10⁹ (B) 6.0 × 10⁹ (C) 6.0 × 10³ (D) 9.6 × 10⁹ (E) 1.2 × 10¹⁰
8. A set of data has a mean of 50 and a median of 55. Which of the following must be true?
(A) The data set is symmetric. (B) The data set is skewed to the left. (C) The data set is skewed to the right. (D) The data set has no mode. (E) The range is greater than 5.
9. In a probability experiment, a fair coin is flipped 4 times. What is the probability of getting exactly 3 heads?
(A) 1/16 (B) 1/8 (C) 1/4 (D) 3/8 (E) 1/2
10. The concentration of a chemical solution decreases by 10% each hour. If the initial concentration is 100 units, which expression gives the concentration after t hours?
(A) 100 - 10t (B) 100(0.9)ᵗ (C) 100(1.1)⁻ᵗ (D) Both B and C (E) 100/(1 + 0.1t)
SECTION 4：ANALYTICAL WRITING
ISSUE ESSAY
Prompt
“The rapid pace of technological innovation has made it more important than ever for educational institutions to focus on teaching students how to think critically and adapt, rather than on transmitting any specific body of knowledge, which may quickly become obsolete.”
Task: Write a response in which you discuss the extent to which you agree or disagree with the statement. In developing your position, address the most compelling reasons and/or examples that could be used to challenge your viewpoint.
腦力激盪指南（威威老師建議）
支持論點：
- 科技更新太快——程式語言、軟體平台壽命越來越短，任何 specific knowledge 都有被淘汰的風險
- 批判思考與適應力是 meta-skills：有了它們，學生可以自主學習任何新知識
- Google/搜尋引擎時代的事實性知識價值下降，重點是判斷資訊可靠性與整合不同來源的能力
反對論點：
- 批判思考不能憑空運作——必須建立在扎實的領域知識之上。你無法 critical think about something you know nothing about
- 基礎學科知識（數學、科學原理、歷史架構）相對穩定，不會那麼快「過時」
- 過度聚焦 skills 而忽略 knowledge 已在部分教育改革實驗中證實會降低學術水準
我的立場： 傾向同意，但強調 knowledge 是 skills 的載體——兩者不可分割。教育應以 critical thinking 為目標，但透過具體知識領域來鍛鍊。
模範作文（478 words）
The claim that education should prioritize critical thinking and adaptability over transmitting specific knowledge is a response to a genuine challenge: the accelerating obsolescence of many forms of specialized knowledge. In a world where programming languages rise and fall within a decade and where entire industries are reshaped by technological disruption, an education focused solely on specific facts and techniques risks preparing students for a world that will no longer exist by the time they graduate. There is considerable wisdom in this argument, but it draws a distinction between thinking skills and knowledge that is ultimately unsustainable. Critical thinking cannot be taught in a vacuum; it requires rich content through which to develop and express itself.
The argument’s strongest element is its recognition that the half-life of specific technical knowledge is shrinking. A computer science student who learns only a particular programming language or framework is indeed likely to find that knowledge partially obsolete within a few years. What endures is not the syntax of Java or Python but the capacity to reason algorithmically, to decompose complex problems, to debug systematically, and to learn new languages as needed. These meta-cognitive abilities—what the prompt calls “thinking critically and adapting”—are genuinely more durable than any specific technical skill. Educators who ignore this reality risk giving students the educational equivalent of perishable goods.
However, the claim’s weakness lies in its implicit suggestion that these durable skills can be developed independently of specific content knowledge. The cognitive science research on this question is clear: critical thinking is domain-specific in important ways. The ability to evaluate evidence in historical analysis does not automatically transfer to evaluating evidence in a physics experiment. Expertise in any field requires a substantial base of organized knowledge—facts, concepts, frameworks—that the expert mind draws upon when reasoning. A student who has been taught “critical thinking” in the abstract, without the rich disciplinary content that gives it traction, is like a carpenter taught to swing a hammer but never given wood and nails.
The most productive resolution of this tension is to recognize that knowledge and thinking are not competitors for curricular space but partners in intellectual development. The goal is not to choose between teaching facts and teaching thinking but to teach facts in ways that require thinking—and to teach thinking through engagement with substantive factual content. A history class that asks students to analyze primary sources is simultaneously building knowledge of the period and developing the skill of source evaluation. A biology lab that requires students to design their own experiments teaches both scientific concepts and the methodology of scientific inquiry.
Education should indeed aim to produce adaptable, critical thinkers, but it should pursue this aim by immersing students in rich, challenging content—not by abandoning content in favor of decontextualized skills instruction. The question is not whether to choose knowledge or thinking but how to teach both in a mutually reinforcing relationship.
ARGUMENT ESSAY
Prompt
The following appeared in a technology magazine’s editorial:
“Last year, the city of Greenfield installed smart traffic lights at 20 of its busiest intersections, at a cost of $2 million. According to the city’s transportation department, average commute times decreased by 15% over the following six months. Furthermore, a survey of Greenfield residents found that 68% reported improved satisfaction with traffic conditions. These results clearly demonstrate that smart traffic light systems are a cost-effective solution to urban congestion, and cities across the country should adopt similar systems without delay.”
Task: Write a response in which you discuss what questions would need to be answered in order to decide whether the recommendation is likely to have the predicted result. Be sure to explain how the answers to these questions would help to evaluate the recommendation.
邏輯謬誤分析
Flaw 1 — 因果關係不明（Unaddressed Confounding Variables）： 通勤時間縮短 15% 可能不是因為 smart traffic lights。同期可能有其他變化——新的道路、經濟衰退導致車流量減少、更多人在家工作。Without controlling for these, the claim fails.
Flaw 2 — 居民問卷的主觀性與涵蓋範圍（Subjective Survey and Selection Bias）： 「68% 居民說滿意度提升」——未說明問卷對象（只有開車族？坐公車的要如何感受交集號誌改善？）。樣本大小？回應率？問卷措辭可能誘導回答。
Flaw 3 — 成本效益分析不足（Cost-Effectiveness Not Proven）： $2 million 的花費是否值得？15% 的 commute time 減少換算成經濟價值是多少？有無更便宜的做法（optimize signal timing without new hardware）可達到類似效果？
Flaw 4 — 適用性問題（External Validity / Generalization）： Greenfield 的結果不一定能推論到「所有城市」。城市規模、氣候（snow affects sensors?）、既有交通基礎建設都會影響效果。
模範作文（463 words）
The editorial confidently recommends that cities nationwide adopt smart traffic light systems based on evidence from a single city, Greenfield. While the reported results are superficially encouraging, the recommendation rests on a series of unanswered questions that must be addressed before any rational policy-maker could endorse the proposed course of action.
The most pressing question is whether the 15% reduction in commute times can actually be attributed to the smart traffic lights. The six-month period following installation may have coincided with other changes that independently reduced traffic congestion. Did Greenfield experience a decline in employment during this period, reducing the number of commuters? Were alternative transportation options—new bus routes, bike lanes, or remote work policies—introduced simultaneously? Did road construction projects elsewhere in the city temporarily redirect traffic patterns? Without data controlling for these potential confounding variables, the observed reduction in commute times could be entirely coincidental. If a rigorous analysis found that most of the improvement was attributable to factors other than the traffic lights, the case for adoption would collapse.
A second set of questions concerns the resident satisfaction survey. The editorial reports that 68% of residents reported improved satisfaction, but critical methodological details are absent. What was the sample size, and how were respondents selected? If the survey primarily reached residents who live near the upgraded intersections, the sample may not represent citywide opinion. How were survey questions worded? A leading question such as “Do you agree that the new smart traffic lights have improved your commute?” would generate very different results from an open-ended question about traffic experiences. Furthermore, did the survey distinguish between different types of road users—drivers, pedestrians, cyclists, and public transit riders—who may experience intersection changes very differently? If the survey methodology was flawed or the sample was unrepresentative, the satisfaction finding would carry little weight.
The question of cost-effectiveness also demands scrutiny. The editorial asserts that the system is “cost-effective” without providing any meaningful cost-benefit analysis. What is the economic value of the 15% time savings, expressed in terms of productivity gains or fuel cost reductions? How does this value compare with the $2 million investment and ongoing maintenance costs? Have alternative, less expensive interventions—such as retiming existing traffic signals or adjusting intersection geometry—been evaluated? A 15% improvement is meaningless in economic terms without knowing what it cost to achieve relative to other possible uses of municipal funds.
Finally, the most important unanswered question is whether Greenfield’s results can be generalized to other cities. Greenfield may have particular characteristics—population density, street grid layout, climate conditions, or pre-existing traffic signal infrastructure—that made smart lights unusually effective there. Cities with different characteristics might see dramatically different results. Without evidence from diverse urban contexts, the editorial’s nationwide recommendation is premature at best and recklessly speculative at worst.
ANSWER KEY：答案與詳解
🔐 此內容需要解鎖碼才能查看。輸入解鎖碼 →
| 題號 | 答案 | 詳解（中文） |
|---|---|---|
| 1 | C | 主旨題。文章先介紹 CRISPR 的革命性影響，再探討倫理與生態風險，C 選項（balance of potential and concerns）最準確。 |
| 2 | B | 細節題。最後一段強調 gene drives 的危險來自自我傳播的特性：“designed to perpetuate itself”。 |
| 3 | B | 功能題。中國研究者的實驗展示了技術能力超越倫理共識的危險，這是 “gap between what CRISPR makes technically possible and what society has agreed is ethically permissible”。 |
| 4 | ”The very feature that makes it powerful—self-propagation—makes it uniquely dangerous.” | 選句題。這句話最直接表達 gene drive 的矛盾性：讓它強大的特質同時是它最大的危險。 |
| 5 | C | 功能題。Searle 的 Chinese Room 是被用來論證「通過 Turing test 不代表真正理解」。 |
| 6 | C | 推論題。作者認為 Turing test 太注重語言輸出，可能未能充分捕捉 intelligence 的本質。 |
| 7 | B | 細節題。pattern 就是 “technological fixes that ignore social and political dimensions tend to generate new, more intractable problems”。 |
| 8 | C | 功能題。汽車解決了馬糞問題但創造了霧霾等新問題——這是科技方案換湯不換藥的歷史案例。 |
| 9 | B, C | 多選題。B 正確（incorporate multiple systems）；C 正確（global mean temperature 預測比 local 預測更可靠）；A 錯誤（modern models 明顯比 1960s 的更先進）。 |
| 10 | B | 推論題。“scale gap” 特指模型在全球趨勢預測與地方級預測之間的精準度落差。 |
| 11 | B | overshadowed（使黯然失色）。後續作品被早期重大發現的陰影所掩蓋。 |
| 12 | B | hyperbolic（誇張的）。“marginally better” 與 “revolutionary” 形成對比，證明宣告是誇大的。 |
| 13 | B, E | bolstered（支持）and definitive（確定的）。數據支持假設，但樣本量太小，不能下確定結論。 |
| 14 | B, D | exacting（嚴格的）and scrutiny（審查）。嚴格審查同事稿件，雖然有科學價值但可能傷害關係。 |
| 15 | B, E, H | rapid（快速）、state-of-the-art（最先進的）、obsolete（過時的）。機器學習進展快到五年前的尖端技術現在已經過時。 |
| 16 | B, E, G | invention（發明）、breakthrough（突破）、theoretical（理論的）。望遠鏡的發明不是突然的突破，而是漸進改良過程，依賴玻璃工藝進步與理論洞察。 |
| 17 | B, D | endorsed & embraced（支持、接受）。多數研究者支持，但少數人批評統計方法。 |
| 18 | B, D | elegance & simplicity（優雅、簡潔）。用一半的組件實現相同功能，這是設計上的 elegance/simplicity。 |
| 19 | B, E | accessible & comprehensible（可理解的）。“using vivid analogies and minimal technical jargon” 是為了讓量子物理對一般讀者變得 accessible。 |
| 20 | D, F | circumspect & cautious（謹慎的）。雖然早期結果令人鼓舞，但強調需要更大型臨床試驗才能下結論——顯示出謹慎的態度。 |
Verbal Reasoning Section 2 答案
| 題號 | 答案 | 詳解（中文） |
|---|---|---|
| 1 | B | 結構題。全文先批判 Apollo model 的侷限，用 historical examples 支持，再應用於當代 AI 辯論。 |
| 2 | B | 功能題。War on Cancer 是 Apollo model 應用在定義不清的問題上效果不如預期的例子。 |
| 3 | 最後一句 | 選句題。最後一句 “design funding ecosystems that sustain a productive interplay between the two” 是作者的解決方案。 |
| 4 | B | 推論題。作者認為傳統觀點是過度簡化，忽略了 placebo 效應背後的真實神經生物機制。 |
| 5 | B | 功能題。內源性類鴉片與多巴胺的提及是用來證明 placebo 效應有真實的神經生物學基礎。 |
| 6 | A, B | 多選題。A 對應 discourages collaborative work；B 對應 imposter phenomenon；C 錯誤（作者認為這些故事是 retrospective constructions）。 |
| 7 | B | 細節題。作者明確稱這些故事為 “retrospective constructions that compress complex, collaborative, and incremental processes”。 |
| 8 | D | 排除題。A、B、C、E 都在文中有提到，唯獨 D（sudden, isolated flashes of individual genius）正是作者在批判的神話。 |
| 9 | A | outpaced（超越、超過）。科技發展速度超過了倫理指南的發展速度。 |
| 10 | B | heretical（異端的）。最初被認為是異端，但證據累積後獲得認可。 |
| 11 | B, E | insightful（有洞見的）and self-serving（自私的）。提供了有用建議（insightful），但也過度自我吹捧（self-serving）。 |
| 12 | B, F | niche（小眾的）and broadened（擴大了）。基因療法曾被認為只適用少數罕見疾病，最近的成功擴大了其潛在應用範圍。 |
| 13 | B, E | emphatic（斷然的）and equivocal（模稜兩可的）。證據不明確，但科學家的語氣卻非常斷然。 |
| 14 | A, D, G | autonomous（自主的）、catalyst（催化劑）、contingent（依賴的）。印刷術不是自主的變革推動者，而是現有社會力量的催化劑；其革命性效果依賴於已有閱讀需求的公眾。 |
| 15 | B, E, G | compromised（受損的）、validity（有效性）、incorporated（被納入）。研究因使用有效性受質疑的技術而受損，她的發現被納入文獻時問題尚未被解決。 |
| 16 | B, D | convoluted & tortuous（費解的、曲折複雜的）。prose so dense 與 struggle to extract 說明文章極度複雜難懂。 |
| 17 | A, C | inflated & exaggerated（誇大的）。分析師學會將科技公司的承諾打折 50%，說明其承諾總是過度誇大。 |
| 18 | B, E | diminution & reduction（減少、縮減）。“Far from” 暗示相反——智慧型手機不是計算能力的減少，而是濃縮。 |
| 19 | B, D | convergent & congruent（趨同的、一致的）。兩個獨立團隊得出相同結論，增加假設的可信度。 |
| 20 | D, E | corroborated & authenticated（證實、驗證）。新天體的發現與預測吻合，證實了她的假設。 |
Quantitative Reasoning 答案
| 題號 | 答案 | 詳解 |
|---|---|---|
| 1 | E | 總 ratio 2+3+5=10 份，每份 75,000 |
| 2 | E | f(-2)=(-2)²-3(-2)+2=4+6+2=12 |
| 3 | A | 8400.15=12652=$6,552 |
| 4 | C | SD=0 表示所有值相同。200/20=10 |
| 5 | C | P(at least one)=1-P(neither)=1-(0.92*0.88)=1-0.8096=0.1904 |
| 6 | C | V=πr²h=3.1495=141.3 |
| 7 | B | (2.410⁶)(2.510³)=(2.42.5)10⁹=6.010⁹ |
| 8 | B | Mean < Median 表示數據向左偏斜（負偏態） |
| 9 | C | C(4,3)(0.5)⁴=41/16=1/4 |
| 10 | B | 100(0.9)ᵗ 表示每年衰減 10%，等價於 (1−0.1)ᵗ。選項 B 的 100(1−0.1)ᵗ 為標準答案；選項 C 的 (1.1)⁻ᵗ 雖數值近似但不嚴格相等。 |
寫作評分標準（0-6 分制）
| 分數 | 描述 |
|---|---|
| 6.0 | Outstanding. Insightful, sophisticated analysis; compelling, well-developed examples; elegant organization; exceptional command of language. |
| 5.0 | Strong. Generally thoughtful analysis; well-chosen examples; clear organization; strong vocabulary and sentence fluency; minor errors. |
| 4.0 | Adequate. Competent analysis; relevant examples; acceptable organization; adequate control of language; some errors. |
| 3.0 | Limited. Superficial analysis; weak or poorly developed examples; inconsistent organization; frequent language errors. |
| 2.0 | Seriously Flawed. Little to no analysis; irrelevant or missing examples; poor organization; pervasive language errors. |
| 1.0 | Fundamentally Deficient. Essentially incoherent; minimal development; severe and persistent language errors. |
| 0.0 | Off-topic, not in English, or no text submitted. |
本回單字表（20 個高難度 GRE 字彙）
| 英文 | 中文意思 | 出處 |
|---|---|---|
| CRISPR-Cas9 | 基因編輯技術 | RC Long Passage |
| biohacker | 生物駭客（業餘生物科技實驗者） | RC Long Passage |
| ostensibly | 表面上；聲稱地 | RC Long Passage |
| gene drive | 基因驅動（強制遺傳的基因改造機制） | RC Long Passage |
| interlocutor | 對話者；交談者 | RC Short A |
| intentionality | 意向性（哲學用語） | RC Short A |
| smog | 煙霧（工業污染與霧的混合） | RC Short B |
| downscaling | 降尺度（從大範圍數據推估地方數據） | RC Short C |
| hyperbolic | 誇張的；雙曲線的 | TC (12) |
| exacting | 嚴格的；苛求的 | TC (14) |
| state-of-the-art | 最先進的 | TC (15) |
| obsolete | 過時的；淘汰的 | TC (15) |
| circumspect | 謹慎的；周到的 | SE (20) |
| sanguine | 樂觀的；血紅色的 | SE (20) |
| heretical | 異端的；邪說的 | TC (S2-10) |
| equivocal | 模稜兩可的 | TC (S2-13) |
| convoluted | 極度複雜的；費解的 | SE (S2-16) |
| tortuous | 曲折的；迂迴的 | SE (S2-16) |
| corroborated | 確證的；證實的 | SE (S2-20) |
| retrospective | 回顧的；追溯的 | RC S2 Short B |
自我評分追蹤表
| 部分 | 滿分 | 得分 | 正確率 |
|---|---|---|---|
| Verbal Section 1 (RC) | 10 | /10 | % |
| Verbal Section 1 (TC) | 6 | /6 | % |
| Verbal Section 1 (SE) | 4 | /4 | % |
| Verbal Section 1 總分 | 20 | /20 | % |
| Verbal Section 2 (RC) | 8 | /8 | % |
| Verbal Section 2 (TC) | 7 | /7 | % |
| Verbal Section 2 (SE) | 5 | /5 | % |
| Verbal Section 2 總分 | 20 | /20 | % |
| Quantitative Reasoning | 10 | /10 | % |
| 全卷總分 | 50 | /50 | % |
威威老師的話： 第二回的主題是科學與科技，這在 GRE 中是非常常見的閱讀主題。注意科學文章的特性——它們常常先介紹某個技術或發現，然後探討其 limitations 或 controversies。閱讀時掌握這個結構，解題會輕鬆很多。加油！