题型04 阅读理解之说明文（解题技巧+题型专练）（北京高考真题+各区模拟）（北京专用）2026年高考英语二轮复习讲练测

题型04 阅读理解之说明文 目录 2025东城期末 1 2025西城期末 2 2025朝阳期末 4 2025海淀期末 5 2025丰台期末 7 2025石景山期末 8 2025东城一模 9 2025西城一模 11 2025朝阳一模 12 2025海淀一模 14 2025丰台一模 15 2025石景山一模 17 2025东城二模 18 2025西城二模 19 2025朝阳二模 21 2025海淀二模 22 2025丰台二模 24 2025东城期末 C In 1993, a ship sailed into the Pacific Ocean carrying nearly 1,000 pounds of iron, then dumped it all into the waves. The next morning, the water turned a little greenish due to newly emerged phytoplankton (浮游生物). The microorganisms, which need iron to grow, draw CO₂ out of the air. Fertilizing the ocean with iron is a form of geoengineering, a set of technologies fascinating for their potential to meaningfully impact Earth’s systems, and controversial for the same reason. As the planet heats, geoengineering is employed to manipulate (控制) it in novel ways, whether researchers mist the skies with sea salt to increase how much sunlight clouds reflect or invent machines to suck greenhouse gases out of the atmosphere. Engineering home-grown plankton sounds like a less dramatic approach. But the field is so new that scientists don’t yet know whether geoengineering with microorganisms would really be a gentler form of climate intervention. Microbes, after all, play enormously consequential roles in the world around us and within us. For the best chance of keeping Earth liveable, scientists need to understand exactly how the microscopic creatures might be useful, and perhaps even preferable to the more sci-fi approaches to cooling off the world. One favored approach centers on methane, a greenhouse gas accounting for 30% of the rise in global temperature. Mary Lidstrom, a professor, is working on genetically modifying bacteria that naturally consume methane, so that the microbes pull even more gas from the air. The bacteria would live inside facilities known as bioreactors. They could be positioned near known methane sources — landfills and wetlands — to minimize the amount of methane that makes it into the air. Scaling up any kind of climate engineering is tricky. Significantly slowing warming caused by methane would require about 300,000 bioreactors to be active for 20 years. And the downstream effects are difficult to predict, let alone contain. For example, the iron put into the Pacific Ocean could have consequences for other regions of the world’s seas. Also there’s nearly as much uncertainty about intended effects. Scientists don’t know for sure how effective these concepts will prove to be. The 1993 ship expeditions, for example, didn’t stick around very long; the journeys were expensive, and scientists couldn’t remain at sea to determine what exactly was happening in the depths. Carbon stolen out of the atmosphere should remain buried for at least a century for maximum impact Yet it’s entirely possible that if the phytoplankton doesn’t sink to deep enough waters, it just gets eaten by the organisms in the surface ocean, and that CO₂ just returns to the atmosphere. The hotter the planet gets, the more attractive geoengineering will likely seem. Short of large-scale behavioral changes globally, we do seem to be committing ourselves to an engineered solution on our current routes. We might have to decide, sooner than we think, which levers we need to pull, large or small. 27. What can we learn about geoengineering? A. Ocean iron-fertilization is on the horizon. B. Sunlight-reflection method is cost-effective. C. Methane-eating bacteria are being improved. D. Gas-sucking machine is a theoretical concept. 28. What can we infer from Paragraph 4 about the geoengineering approaches? A. Their impact on the ecosystem is overlooked. B. Their implications are hard to anticipate. C. They call for decades of preparation. D. They are monitored continuously. 29. The iron-fertilization experiment may achieve its intended effects if __________. A. the target microbes survive B. carbon is stored separately C. the phytoplankton grows faster D. more deep-sea organisms are tested 30. What might be the best title for the passage? A. How Will Geoengineering Better Earth’s Systems? B. How to Cool the World Without Blocking the Sun? C How to Steal Carbon with Sci-fi Approaches? D. How Can the Engineered Solution Deliver? 文章主旨：本文围绕地球工程（geoengineering）展开，介绍了其多种形式（如海洋铁施肥、甲烷吞噬细菌改良等），探讨了这些技术在应对气候变化中的潜力，同时也指出了其规模化应用的困难、下游影响的不可预测性以及预期效果的不确定性。 27. 答案：C 解析：定位原文第三段“Mary Lidstrom, a professor, is working on genetically modifying bacteria that naturally consume methane, so that the microbes pull even more gas from the air”，可知吞噬甲烷的细菌正在被改良，以提升其吸收甲烷的能力。选项A错误，海洋铁施肥是1993年的实验，并非“即将实现”；选项B错误，文中未提及反光法“成本效益高”；选项D错误，吸气机器是地球工程的一种形式，并非“理论概念”，因此选C。 28. 答案：B 解析：定位原文第四段“the downstream effects are difficult to predict, let alone contain”，可知地球工程的下游影响难以预测，甚至无法控制。选项A错误，文中未说“生态系统影响被忽视”，而是“难以预测”；选项C错误，文中提到规模化需要20年运作时间，并非“数十年准备”；选项D错误，1993年的实验并未持续监测，因此选B。 29. 答案：B 解析：定位原文第五段“Carbon stolen out of the atmosphere should remain buried for at least a century for maximum impact. Yet it’s entirely possible that if the phytoplankton doesn’t sink to deep enough waters, it just gets eaten by the organisms in the surface ocean, and that CO₂ just returns to the atmosphere”，可知铁施肥实验要达到预期效果，关键是吸收的碳能长期埋藏，若浮游生物未沉入深海，碳会重新释放。选项A错误，微生物存活只是前提，并非关键；选项C错误，浮游生物生长速度不是核心；选项D错误，文中未提及“测试更多深海生物”，因此选B。 30. 答案：D 解析：本文核心探讨地球工程作为应对气候变化的“工程解决方案”，其潜力、挑战及不确定性，重点关注这些方案能否真正发挥作用。选项A错误，文章未详细说明“如何改善地球系统”，而是聚焦可行性和问题；选项B错误，“不遮挡太阳”并非核心；选项C错误，“科幻手段吸碳”只是部分内容，并非主旨，因此选D。 2025西城期末 C In recent research, my colleagues and I set out to investigate the neural (神经的) mechanisms underlying experiences like solving challenging problems. Specifically, we wanted to understand what happens in the brain while a person engages in abstract and demanding thought — so we designed a study involving math expertise. For our experiment, we selected 22 university students majoring in mathematics, along with 22 students from other disciplines. We asked the students to watch step-by-step presentations that explained how to solve several challenging math problems. Throughout this demonstration, the students wore a cap covered with electrodes (电极) so that we could track electrical activity in their brains. After each presentation, they had to report whether they thought they had understood the presentations and how engaged they felt during the experience. We found that students whose coursework involved little mathematics showed more signs of complex activity in the prefrontal cortex (皮层) indicating how hard they were working to understand the various steps of the complex math demonstrations. However, in the other group, we noted significant activity linking the frontal and parietal (顶骨) regions of their brain. More specifically, these areas exhibited a pattern of activity described as delta waves. Some recent research suggests that these delta waves may play a crucial role in the cognitive (认知的) processing that supports deep internal concentration and the relocation of information between distant brain regions. In fact, we suspect that the long-distance delta waves we observed may play a central role whenever people engage in contextual and complex problem-solving. This suggests that engaging brain networks in this way could be useful for many tasks requiring concentration. It’s likely that when people with extensive experience in a task are deeply engaged in that effort, these same delta waves are involved, even as the specific brain networks vary. It’s also possible — though we’ll need to investigate further to be sure — that this state of deep concentration is generalizable: developing this way of thinking in one domain could help you in others. These findings suggest a trade-off that people should keep in mind—particularly as artificial intelligence and other tools offer shortcuts for various forms of problem-solving. Each time we offload a problem to a calculator or ask Al to summarize an essay, we lose an opportunity to improve our own skills and practice deep concentration. To be clear, technologies can boost our efficiency in important ways, but the seemingly “inefficient” hard work we do can be powerful, too. 28. Based on the study results, how do Math majors process challenging math problems differently from non-Math majors? A. They experience less engagement during problem-solving tasks. B. They use more areas of their prefrontal cortex to solve problems. C. They display delta wave activity linking frontal and parietal regions. D. They show slower overall brain activity during complex problem-solving. 29. What does the underlined word “trade-off” in the last paragraph probably refer to? A. Weighing short-term efficiency against lasting cognitive benefits. B. Choosing between immediate convenience and faster results. C. Sacrificing technology’s advantage for human competition. D. Deciding between human effort and advanced tools. 30. Which of the following best summarizes the author’s view on the role of “inefficient” work? A. It allows the brain to build skills that are transferable to other tasks. B. It wastes time and resources compared to using technology. C. It improves engagement but has no cognitive benefits. D. It helps those who are good at subjects like math. 31. What is the best title for the passage? A. Which to Choose: Efficiency or Hard Work? B. What Benefits Us More: AI Tools or Deep Concentration? C. Employing Complex Math Problems to Enhance Delta Waves D. Developing Expertise Improves the Brain’s Ability to Concentrate 文章主旨：本文介绍了一项关于数学专长与大脑神经机制的研究，通过对比数学专业和非数学专业学生解决复杂数学问题时的大脑活动，发现数学专业学生大脑额叶和顶叶区域会产生δ波，这种脑波与深度专注相关，同时作者强调过度依赖AI等工具会错失提升认知能力的机会，“低效”的努力对大脑发展至关重要。 28. 答案：C 解析：定位原文第三段“however, in the other group, we noted significant activity linking the frontal and parietal regions of their brain. More specifically, these areas exhibited a pattern of activity described as delta waves”，可知数学专业学生的大脑额叶和顶叶区域会产生δ波活动，而非数学专业学生则是前额叶皮层有复杂活动。选项A错误，文中未说数学专业学生“参与度低”；选项B错误，非数学专业学生更多使用前额叶皮层；选项D错误，文中未提及“大脑整体活动更慢”，因此选C。 29. 答案：A 解析：定位原文最后一段“Each time we offload a problem to a calculator or ask AI to summarize an essay, we lose an opportunity to improve our own skills and practice deep concentration. To be clear, technologies can boost our efficiency in important ways, but the seemingly ‘inefficient’ hard work we do can be powerful, too”，可知“trade-off”指的是“短期效率”与“长期认知收益”之间的权衡。选项B错误，未涉及“更快结果”；选项C错误，并非“牺牲技术优势换人类竞争”；选项D错误，“人类努力与先进工具”是表面，核心是两者带来的不同收益，因此选A。 30. 答案：A 解析：定位原文第四段“it’s also possible — though we’ll need to investigate further to be sure — that this state of deep concentration is generalizable: developing this way of thinking in one domain could help you in others”，可知“低效”的努力（如深度专注解决问题）能培养可迁移到其他任务的技能。选项B错误，作者认为这种努力并非“浪费时间”；选项C错误，文中明确提到有认知益处；选项D错误，这种努力对各领域的人都有帮助，并非只针对擅长数学的人，因此选A。 31. 答案：B 解析：本文核心对比了AI工具带来的效率与深度专注带来的认知提升，强调后者的重要性。选项A错误，“效率与努力”范围过广，未聚焦“AI工具”与“深度专注”的核心对比；选项C错误，“通过复杂数学题增强δ波”只是研究细节；选项D错误，“培养专长提升专注力”并非主旨，主旨是“深度专注的价值”与“工具的权衡”，因此选B。 2025朝阳期末 C According to an idea called the “inflationary (暴胀的) multiverse”, the universe has been constantly expanding, which would have created a countless number of other universes. Collectively, these cosmic islands form what scientists call a “multiverse”. However, other universes may lie beyond our scientific reach, physically and philosophically. Consequently, the methodologies by which physicists might validate (验证) the multiverse and whether such validation is even possible remain unresolved inquiries. In science, hypotheses (假说) are used to propose explanations, but from hypotheses to theories, validation of predictions is essential to see whether initial assumptions are supported or disproved. Theoretical physicist Carlo Rovelli says, “It is hard to write theories that survive the proof of reality. Few survive thanks to a basic idea: Do not trust your fancies. Keep only the ideas that can be tested.” Rovelli adds, “Today many physicists are wasting time trying to guess and jumping to conclusions. This approach has never worked previously, and it is not working now.” The multiverse might be one of those wild guesses. Now, physicists are debating whether that problem moves the multiverse from physics to metaphysics, from the world of science to that of philosophy. Philosopher Richard Dawid believes scientists could support the multiverse hypotheses without actually finding physical support. “If we have a theory that seems to work, and have come up with nothing that works better, chances are our idea is right,” he says. Not everyone is so cheerful, though. Sabine Hossenfelder, another theoretical physicist, thinks “Without making contact with observation, a theory isn’t part of the natural sciences, and not physics.” Some supporters of the multiverse claim they have found physical evidence. Andrei Linde insists “Our universe seems fine-tuned (精调的) to be favourable to life, with a four-dimensional space in which we can live. It is certain that in an eternally inflating multiverse, one of the universes should turn out like ours. Each universe can have different physical laws and fundamentals. Given limitless changes, a universe on which humans can be born will be born. The multiverse actually explains why we’re here. And our existence, therefore, helps explain why the multiverse is plausible.” Just as Einstein’s theory of general relativity could only be confirmed recently with a billion-dollar instrument called LIGO, scientists may eventually discover more direct evidence of the multiverse. So far, all of science has relied on testability, but those tests take time, and most theoreticians want to wait it out. They are not ready to shelve an idea like the multiverse — which could actually be the answer to life, the universe and everything — until and unless they can prove to themselves it doesn’t exist. And that day may never come. 28. What can we learn from the passage? A. The multiverse is recognized as a theory. B. The multiverse starts debates on human existence. C. Rovelli criticizes physicists for their baseless fancies. D. Dawid emphasizes the necessity of physical evidence. 29. What does the underlined word “plausible” in Paragraph 4 probably mean? A. Credible. B. Complex. C. Essential. D. Fascinating. 30. As for the existence of multiverse, the author is _________. A. negative B. neutral C. certain D. positive 31. Which would be the best title for this passage? A. Mysteries of Multiverse: Fine-Tuned for Life B Can Physicists Ever Prove the Multiverse Is Real? C. The Inflationary Multiverse: Proven Beyond Doubt D. Have Physicists Lost Their Minds in the Multiverse? 文章主旨：本文围绕“暴胀多元宇宙”假说展开，探讨了物理学家和哲学家对多元宇宙是否存在、能否被验证的争议。部分物理学家认为多元宇宙可解释宇宙的“精调”特性，但也有学者强调科学理论需可测试性，目前多元宇宙的验证仍面临巨大挑战。 28. 答案：C 解析：定位原文第二段“Rovelli says, ‘It is hard to write theories that survive the proof of reality. Few survive thanks to a basic idea: Do not trust your fancies. Keep only the ideas that can be tested.’ Rovelli adds, ‘Today many physicists are wasting time trying to guess and jumping to conclusions’”，可知Rovelli批评物理学家沉迷无根据的幻想，缺乏可测试性的理论。选项A错误，多元宇宙仍是假说，未被认可为“理论”；选项B错误，多元宇宙假说解释了人类存在，但未“引发关于人类存在的争论”；选项D错误，Dawid认为无需物理证据也可支持假说，因此选C。 29. 答案：A 解析：定位原文第四段“Andrei Linde insists ‘Our universe seems fine-tuned to be favourable to life... The multiverse actually explains why we’re here. And our existence, therefore, helps explain why the multiverse is plausible’”，可知Linde认为人类的存在能解释多元宇宙的“合理性”，因此“plausible”意为“可信的”。选项B（复杂的）、C（必要的）、D（迷人的）均不符合语境，因此选A。 30. 答案：B 解析：作者客观呈现了不同学者对多元宇宙的观点，既提到支持方的理由，也指出反对方的依据，最后提到“科学家可能最终发现更多直接证据”“那天可能永远不会到来”，未表达明确的支持或反对态度，保持中立。选项A（否定的）、C（确定的）、D（积极的）均不符合，因此选B。 31. 答案：B 解析：本文核心围绕“多元宇宙能否被物理学家证明”展开，探讨了科学验证的必要性、不同学者的争议及未来可能性。选项A错误，“为生命精调”只是支持方的一个理由，非主旨；选项C错误，多元宇宙尚未被“证实”；选项D错误，文章未指责物理学家“失去理智”，而是客观讨论争议，因此选B。 2025海淀期末 C As global temperatures continue to rise due to human-induced climate change, precise computerized climate models will be crucial in shedding light on how our climate will evolve in the coming years. A study published in Journal of Geophysical Research: Atmospheres by Chloe Clark and her team from UC Irvine highlights a significant issue with a commonly used climate model known as the Energy Exascale Earth System Model (E3SM). The research reveals that this model currently overestimates a key physical property of Earth’s climate system called albedo, which is the degree to which ice reflects planet-warming sunlight into space. Albedo is a measure of the reflectivity of a surface, expressed as the fraction of incoming solar radiation that is reflected by that surface back into space. It is a critical factor in determining Earth’s climate and energy balance. Surfaces with high albedo, such as snow and ice, reflect a large portion of solar energy, whereas darker surfaces like forests or oceans absorb more solar energy. “We found that with old model versions, the ice is too reflective by about five percent,” said Chloe Clarke, “Ice reflectivity was much too high.” The amount of sunlight the planet receives and reflects is important for estimating how much the planet will warm in the coming years. Previous versions of E3SM overestimated albedo because they failed to account for microphysical properties of ice in a warming world. These properties include the effects things like algae (藻类) and dust have on albedo. Dark colored algae and dust can make snow and ice less reflective and less able to reflect sunlight. To do the analysis, Clarke and her team studied satellite data to track the albedo of the Greenland Ice Sheet. They found E3SM overestimates the reflectivity of the ice sheet, meaning the model estimates less melt than what would be expected from the ice microphysical properties. But with the new ice reflectivity incorporated into the model, they discovered the Greenland Ice Sheet is melting at a rate of about six gigatons more than in older model versions. This is based on albedo measurements that are more consistent with satellite observations. Clarke hopes her team’s study stresses the importance of the seemingly minuscule properties that can have far-reaching consequences for the overall climate. “I think our work is going to help models do a much better job of helping us capture snow and ice-related climate feedbacks,” she said. Next, Clarke wants to study different icy parts of the planet to calculate how widespread the albedo discrepancy (差异) is in E3SM. “Our next steps are to get it so the model is functional globally and not just valid over Greenland,” said Clarke, who also intends to compare the new Greenland Ice Sheet melt rates to observations to measure how much more accurate the new ice albedo is. “It would be useful to apply it to glaciers in places like the Andes and Alaska.” 27. According to the passage, albedo ______. A. changes with the lightness of surfaces B. matches up with the shape of the ice or snow C. is a measure of the solar energy arriving at the surface D. varies with the direction of the incoming solar radiation 28. What can be learned from the passage? A The new model is more reliable than satellite observations. B. The previous model versions overestimate Greenland melt. C. Model adaptation improves the accuracy of climate predictions. D. Scientists have used the model globally to reduce the discrepancy. 29 What does the underlined word “minuscule” in Paragraph 7 probably mean? A. Tiny. B. Invisible. C. Plain. D. Stable. 30. The last paragraph is written to ______. A. note the limitations of the finding B. point out further research directions C. illustrate the benefits of a new approach D. sum up the changes made in E3SM 文章主旨：本文介绍了一项关于气候模型E3SM的研究，该模型之前高估了地球反照率（albedo），原因是未考虑变暖环境下冰的微物理特性（如藻类和尘埃的影响）。研究团队通过整合卫星数据修正了模型，提高了对格陵兰冰盖融化速率预测的准确性，并计划进一步拓展研究范围。 27. 答案：A 解析：定位原文第三段“Surfaces with high albedo, such as snow and ice, reflect a large portion of solar energy, whereas darker surfaces like forests or oceans absorb more solar energy”，可知反照率与表面亮度相关，浅色表面反照率高，深色表面反照率低。选项B错误，与“冰或雪的形状”无关；选项C错误，反照率是“反射太阳辐射的比例”，而非“到达表面的太阳能量”；选项D错误，与“太阳辐射的方向”无关，因此选A。 28. 答案：C 解析：定位原文第六段“with the new ice reflectivity incorporated into the model, they discovered the Greenland Ice Sheet is melting at a rate of about six gigatons more than in older model versions. This is based on albedo measurements that are more consistent with satellite observations”，可知修正后的模型预测更符合卫星观测结果，即模型调整提升了气候预测的准确性。选项A错误，新模型与卫星观测更一致，而非“比卫星观测更可靠”；选项B错误，旧模型低估了格陵兰冰盖融化速率；选项D错误，目前模型尚未在全球范围内应用，因此选C。 29. 答案：A 解析：定位原文第七段“Clarke hopes her team’s study stresses the importance of the seemingly minuscule properties that can have far-reaching consequences for the overall climate”，可知“minuscule properties”指看似“微小的”特性，却能产生深远影响。选项B（不可见的）、C（朴素的）、D（稳定的）均不符合语境，因此选A。 30.答案：B 解析：最后一段提到“Next, Clarke wants to study different icy parts of the planet to calculate how widespread the albedo discrepancy is in E3SM”“apply it to glaciers in places like the Andes and Alaska”，明确说明了后续的研究方向。选项A错误，未提及“研究局限性”；选项C错误，未“说明新方法的益处”；选项D错误，未“总结模型的变化”，因此选B。 2025丰台期末 C On May 4, 2017, 395 migratory (迁徙的) birds died after crashing into a tall building in Galveston, Texas. Their migration routes are full of danger — including bright city lights. Every year, millions of birds are thought to be killed in building collisions in the US. The scene of the mass collision in Galveston was a skyscraper which has four corner spotlights and glowing green LED lights around the top. The influence of artificial lights in urban spaces is a well-known phenomenon among birds, but is still not fully understood. Birds tend to migrate by night, and use the stars to navigate. During nocturnal (夜间的) migration, they are often seen gathering around or crashing into lit-up windows or gathering around beams of projected light. The artificial glow of cities appears to attract them for stopovers. One explanation is that they may generally fly towards light sources, but also, that the artificial light confuses, traps and disorients them. Every tragedy is an opportunity to educate. The accident prompted bird conservation organisations to take action and jointly develop a science-based response. Several campaigns encouraged building owners and developers to switch off non-essential lighting to help travelling birds. Soon the first Lights Out for Birds programme was formed, which has now expanded into the statewide. Meanwhile, the Cornell Lab of Ornithology developed a way to forecast bird migration during specific times and locations, using radar. The resulting BirdCast migration forecast maps are freely accessible. Their live migration map allows anyone to check how many birds are migrating now, and where, as well as how many are predicted to pass in the near future. For one night in October, they recorded a “billion bird night”, with that number or more birds passing in a single night. Andrew Farnsworth, a migration ecologist said that early data collection suggests that Light Out is effective, even with the limited data and uncontrollable factors. Moreover, Farnsworth points out protecting species that move presents special challenges for researchers and conservationists. Migrating birds can travel thousands of kilometres, crossing and stopping in many different habitats, traversing borders, and flying through cities — which are continuing to grow around the world. There is no single solution to helping migrating birds. Instead, spreading awareness, empathy and efficient enforcement are all crucial in keeping birds safe and avoiding collisions. “We need to keep our connection to nature through our eyes, ears, nose, whatever to experience the core of all this research and observation.” 28. According to the passage, why do the migrating birds hit the buildings? A. Their routes are blocked by tall buildings. B. They attempt to avoid potential dangers. C. They are affected by the artificial lights. D. Their migration patterns have changed. 29. What can we learn from the passage? A. Light Out is thought to be helpful despite data limitations. B. BirdCast needs further improvement in tracking accuracy. C. Light Out can guide migrating birds away from artificial lights. D. BirdCast can predict the number and species of migrating birds. 30. As for helping migrating birds,which would the author agree with? A. It is worthwhile to improve Light Out. B. It is necessary to make effective regulations. C. It is crucial to work out a one-button solution. D. It is essential to integrate multiple aspects’ effort. 文章主旨：本文讲述了夜间迁徙的鸟类常因城市人工光源的干扰，与高楼发生碰撞死亡的现象。Galveston的鸟类碰撞事件后，环保组织发起“熄灯护鸟”活动，康奈尔鸟类学实验室开发了鸟类迁徙预测工具，作者强调保护迁徙鸟类需要多方努力，包括提高意识、共情和有效执行。 28. 答案：C 解析：定位原文第二段“Birds tend to migrate by night, and use the stars to navigate... The artificial glow of cities appears to attract them for stopovers. One explanation is that they may generally fly towards light sources, but also, that the artificial light confuses, traps and disorients them”，可知鸟类碰撞建筑是因为人工光源的干扰，导致它们迷失方向。选项A错误，路线未被高楼“阻挡”；选项B错误，并非“试图躲避潜在危险”；选项D错误，迁徙模式未改变，因此选C。 29. 答案：A 解析：定位原文第四段“Andrew Farnsworth, a migration ecologist said that early data collection suggests that Light Out is effective, even with the limited data and uncontrollable factors”，可知尽管数据有限，“熄灯护鸟”活动被认为是有帮助的。选项B错误，文中未提及BirdCast“需要提升追踪准确性”；选项C错误，Light Out是“关闭非必要灯光”，而非“引导鸟类远离光源”；选项D错误，BirdCast可预测鸟类迁徙数量和位置，未提及“物种”，因此选A。 30. 答案：D 解析：定位原文最后一段“There is no single solution to helping migrating birds. Instead, spreading awareness, empathy and efficient enforcement are all crucial in keeping birds safe and avoiding collisions”，可知作者认为保护迁徙鸟类没有单一解决方案，需要整合多方努力（提高意识、共情、有效执行）。选项A错误，未强调“改进Light Out”；选项B错误，“制定有效法规”只是其中一部分；选项C错误，与“没有单一解决方案”相悖，因此选D。 2025石景山期末 C Physarum polycephalum, a single-celled slime mold (黏菌), lacks a brain and survives on rotting vegetables, but it may hold valuable insights for city planning, according to the researchers from University of Toronto. This greenish-yellow organism forms a complex network that can transfer nutrients efficiently, which researchers believe could model efficient networks for moving people and goods in urban settings. “Humans aren’t the only ones facing the challenge of designing efficient, tough networks,” says Raphael Kay, a master in engineering. “The slime mold has been shaped by hundreds of millions of years of evolution (进化), so in that sense, it is far more experienced at network design than we humans ever could be.” This idea isn’t new. In 2010, Japanese and U. K. researchers created a nutrient layout resembling the Tokyo subway nodes (节点), and the slime mold grew a network similar to the real system. This experiment sparked biologically inspired adaptive network design, leading to studies of rail and road networks using slime molds. Building on this, Kay and an architecture student Anthony developed a computer model that simulates (模拟) slime molds’ network-building process, aimed at architects, city planners, and other professionals. Their model allows users to specify factors like maximum cost or minimum travel time, generating networks that meet these criteria. To test it, they applied the model to Canada’s Wonderland amusement park and Toronto’s subway system. For Wonderland, their network offered 10% faster travel time at the same cost as the existing setup and was 80% more resilient at maintaining functionality if a part was blocked. For Toronto’ s subway, they found that a network with the same travel time as the real one was 40% less easily affected to disruption. While Kay and his team aren’t suggesting a complete redesign of current networks, their model offers insights for improvements. For instance, it could pinpoint the best places to add nodes or help in planning future layouts. The model can also autonomously generate network nodes based on data like population concentration, adapting to serve specific needs. Hatton, a collaborator, believes this tool could benefit diverse fields, not just those related to the movement of people or goods, but also fields like genetic engineering and energy networks, highlighting the importance of interdisciplinary approaches. 27. What unique feature of the slime mold interests city planners? A. Its reliance on nutrients to survive. B. Its ability to form efficient networks. C. Its nature to grow in rotting vegetables. D. Its adaptive skill in a new environment. 28. Why does the author mention the experiment with Tokyo’s subway nodes? A. To highlight that the Tokyo subway system is efficient. B. To show that slime molds can recognize urban structures. C. To argue that Tokyo subway system needs to be improved. D. To suggest that slime molds can inspire unban network design. 29. What does the underlined word “resilient” probably mean in paragraph 5? A. Reliable. B. Relevant. C. Accurate. D. Advanced. 30. What can we learn from Kay and his team’s model? A. It can be applied beyond transportation. B. It can predict population growth patterns. C. It is limited to networks with simple layouts. D. It is recommended to redesign current networks. 文章主旨：本文介绍了黏菌（Physarum polycephalum）虽无大脑，却能形成高效的营养传输网络，这一特性为城市规划提供了灵感。研究团队基于黏菌的网络构建过程开发了计算机模型，应用于游乐园和地铁系统时表现出更高效、更具韧性的优势，该模型还可拓展到其他领域。 27. 答案：B 解析：定位原文第一段“This greenish-yellow organism forms a complex network that can transfer nutrients efficiently, which researchers believe could model efficient networks for moving people and goods in urban settings”，可知黏菌吸引城市规划师的独特特性是“能形成高效网络”。选项A（依赖营养生存）、C（在腐烂蔬菜中生长）、D（适应新环境的能力）均非核心原因，因此选B。 28. 答案：D 解析：定位原文第三段“In 2010, Japanese and U.K. researchers created a nutrient layout resembling the Tokyo subway nodes, and the slime mold grew a network similar to the real system. This experiment sparked biologically inspired adaptive network design”，可知提及东京地铁实验是为了说明黏菌能为城市网络设计提供灵感。选项A错误，未“强调东京地铁高效”；选项B错误，黏菌并非“识别城市结构”；选项C错误，未“主张东京地铁需要改进”，因此选D。 29. 答案：A 解析：定位原文第五段“For Wonderland, their network offered 10% faster travel time at the same cost as the existing setup and was 80% more resilient at maintaining functionality if a part was blocked”，可知“resilient”指部分被阻断时仍能维持功能，即“可靠的”。选项B（相关的）、C（准确的）、D（先进的）均不符合语境，因此选A。 30. 答案：A 解析：定位原文最后一段“Hatton, a collaborator, believes this tool could benefit diverse fields, not just those related to the movement of people or goods, but also fields like genetic engineering and energy networks”，可知该模型不仅可应用于交通领域，还能拓展到其他领域。选项B错误，不能“预测人口增长模式”；选项C错误，并非“局限于简单布局”；选项D错误，研究团队不建议“彻底重新设计现有网络”，因此选A。 2025东城一模 C Anyone with insomnia knows the impatience and frustration that accompanies sleeplessness. You long for a button that could instantly dampen all that mental activity. The idea of a mental switch is not far-fetched. Most neuroscientists now agree that our wakefulness is coordinated by a tiny bundle of neurons(一小束神经元)known as the “locus coeruleus"(LC),Latin for "blue dot”. It is a literal description: the neurons in the locus coeruleus have the blue colour from the production of a particular neurotransmitter, called norepinephrine. Norepinephrine raises the chance that a neuron will “fire”with an electric current. When they become active, cells in the locus coeruleus pass bundles of this neurotransmitter along their projections to other regions of the brain—enhancing the communication between the neurons in that area. There are slight differences in the process. Depending on the types of receptors they have, some neurons are more sensitive to smaller amounts of norepinephrine, while others only respond to higher thresholds. This means that, as the locus coeruleus activity rises, it will start to affect some brain areas more than others, which can have dramatic effects on things like our focus, concentration and creativity. Given the blue dot's role, it makes sense that it would be quietest at night during sleep. It is not entirely silent, however, but fires occasionally—and recent research by Anita Lüthi at the University of Lausanne suggests that this activity may determine the quality of our sleeps. Across the night, we alternate between different sleep stages. There is “rapid eye movement”(REM)sleep, which is associated with vivid dreaming and is thought to be crucial for processing and consolidating memories. Much of our rest, however, is spent in non-REM(NREM)sleep, during which the brain may engage in a deep clean, clearing away cellular waste. Measuring brain activity in dozing mice, Anita found NREM sleep was associated with temporary bursts of locus coeruleus activity every 50 seconds. As a result, the animal was more sensitive to outside stimuli, like noises—without fully waking. “It’s generating this state of enhanced vigilance(警觉),”Anita says. “It really gives you this idea that wakefulness can be graded in the brain.” The beginning of REM sleep was almost always associated with low locus coeruleus activity. “That transition to REM sleep has to be very well controlled,” says Anita, “because in REM sleep, we have atonia.” That’s the temporary paralysis(麻痹) of our body, which prevents us from physically acting out our dreams. Anita emphasises that her experiments were conducted in mice, so we still need to confirm that the blue dot plays a similar role in human sleep. If so, she suspects that altered locus coeruleus activity could be implicated in conditions—such as anxiety—that may contribute to disordered sleep. She found that exposing her laboratory mice to mild sources of stress—such as knocking on their cage—raised the blue dot's activity and increased their vigilance throughout the night, resulting in fragmented sleep. 27.What does the underlined word “they” in Paragraph 2 refer to? A. Neurons. B. Electric currents. C. Projections. D. Neurotransmitters. 28. According to the passage , what is the role of the LC? A. producing receptors B. Preserving cell sensitivity C. Monitoring brain activity D. Improving neutral connectivity 29.Which of the following may Anita Lüthi agree with? A. The blue dot fires regularly at night B. Stress has an impact on the LC activity C.vLow LC activity can help clean cellular waste D. Atonia results from sudden bursts of brain activity 30.What might be the next step of the research? A. Grading the wakefulness of human brains B. Unlocking the mechanism of sleeo disorder C. Assessing the function of the blue dot on humans D. Identifying approaches to altering the LC activity 文章主旨：本文介绍了大脑中被称为“蓝斑核（LC）”的神经元束，其分泌的去甲肾上腺素能增强神经元间的沟通，影响注意力、专注力等。研究发现LC在夜间睡眠时并非完全静止，其活动会影响睡眠质量，且压力可能导致LC活动异常，进而影响睡眠，目前该研究仍需在人类身上验证。 27. 答案：A 解析：定位原文第二段“The neurons in the locus coeruleus have the blue colour from the production of a particular neurotransmitter, called norepinephrine. Norepinephrine raises the chance that a neuron will ‘fire’ with an electric current. When they become active, cells in the locus coeruleus pass bundles of this neurotransmitter along their projections to other regions of the brain”，可知“they”指代前文的“neurons in the locus coeruleus”（蓝斑核中的神经元）。选项B（电流）、C（突起）、D（神经递质）均不符合指代逻辑，因此选A。 28. 答案：D 解析：定位原文第二段“when they become active, cells in the locus coeruleus pass bundles of this neurotransmitter along their projections to other regions of the brain—enhancing the communication between the neurons in that area”，可知LC的作用是增强神经元之间的沟通。选项A错误，LC不“产生受体”；选项B错误，不“维持细胞敏感性”；选项C错误，不“监测大脑活动”，因此选D。 29. 答案：B 解析：定位原文最后一段“She found that exposing her laboratory mice to mild sources of stress—such as knocking on their cage—raised the blue dot's activity and increased their vigilance throughout the night, resulting in fragmented sleep”，可知Anita Lüthi发现压力会影响LC的活动。选项A错误，LC在夜间是“偶尔放电”，而非“定期”；选项C错误，清理细胞废物与NREM睡眠相关，未提及“低LC活动”的作用；选项D错误，肌强直是REM睡眠的特征，与LC低活动相关，并非“大脑活动突然爆发”导致，因此选B。 30. 答案：C 解析：定位原文最后一段“Anita emphasises that her experiments were conducted in mice, so we still need to confirm that the blue dot plays a similar role in human sleep”，可知目前研究基于小鼠，下一步需要验证蓝斑核在人类身上的功能。选项A错误，未提及“划分人类大脑清醒度等级”；选项B错误，“解锁睡眠障碍机制”是后续可能的应用，而非下一步研究；选项D错误，“确定改变LC活动的方法”并非当前首要任务，因此选C。 2025西城一模 C Recently, I attended a public talk by someone whose views I “knew” I would oppose. And yet, I went. I listened, asked questions, and gave my time. While my core values weren’t transformed in those two hours, I learned something and left with a deeper appreciation for the complexity of other perspectives. In this weeknight activity, I was actively trying to tackle“beliefism,” a divisive phenomenon in which surrounding yourself with people who share your views leads to discrimination against those who disagree. In this way, beliefism deepens division and reinforces polarization—building walls instead of bridges. Indeed, beliefism is widespread in modern society. A significant part of the problem originates from the fact that we live in a world that is both physically and virtually divided. We rarely interact with people from other walks of life. Social media algorithms stoke the fires of division, locking us into echo chambers that reinforce our preexisting beliefs and shut down debate. Ultimately, where many forces are driving us apart, we must think—what can we do to unpack division and restore connection? We can begin by trying to understand the psychology of beliefism, which at its core is a form of bias—a mental shortcut in which we categorise people according to single characteristics or generalising assumptions. Indeed, in a world that is infinitely complex, our minds use these biases to simplify and make sense of the world. The thing is, when we see others only through the perspective of their political or social beliefs, we reduce and flatten them to a single dimension. Further, when people feel they are dismissed or disregarded only for their beliefs, they are left feeling isolated and misunderstood. Secondly, we can understand the tendency for beliefism as part of our evolutionary (进化的) desire to establish a community or group. The problem is that while this instinct (本能) may have once served evolutionary purposes, today, it blocks meaningful dialogue and cooperation. Indeed, research shows that overcoming beliefism has benefits. When we welcome a variety of ideas and perspectives, we are able to overcome group-think and make better decisions and judgements. What’s more, less beliefist people are generally happier, having stronger, more fulfilling relationships and broader horizons. Luckily, there are a number of relevant, research-backed psychological techniques that help build tolerance and break bias. We might exercise individuation, seeing people as diverse-sided individuals and breaking away from reductive ways of thinking. We can practice perspective-taking, building empathy (同理心) by stepping into someone else’s shoes and trying to see the world through their eyes. Ultimately, the way forward is not through divisions, but through a recognition of our shared humanity. Remind yourself that each person exists at the intersection of many identities, experiences, and beliefs. Challenge yourself to practice empathy, and remember that no one is wholly defined by the worst thing they have said or done. 27. What does the underlined word “stoke” in Paragraph 3 probably mean? A. Fuel. B. Keep. C. Put. D. Contain. 28. What can we know about beliefism from the passage? A. Social media algorithms mainly contribute to its wide spread. B. Human evolution proves its harm in establishing group-think. C. It reflects our simplified way of understanding the surrounding world. D. It leads us to make assumptions about others’ political or social beliefs. 29. Which of the following would be most effective in fighting beliefism? A. Facilitating interactions between people from different communities. B. Creating a list of acceptable beliefs for each community to follow. C. Asking people to write about their experiences of being isolated. D. Stressing conflict instead of cooperation between different beliefs. 30. Which would be the best title for the passage? A. Beyond Us and Them B. The Roots of Beliefism C. The Harm of Social Division D. At the Crossroad of Faith 文章主旨：本文提出“信念偏见（beliefism）”这一概念，指人们倾向于与持相同观点的人相处，歧视不同意见者，加剧社会分裂。文章分析了其产生的原因（心理偏见、进化本能、社交媒体影响），并提出通过理解其心理机制、练习同理心等方式来克服这一现象，强调共享人性而非分裂的重要性。 27. 答案：A 解析：定位原文第三段“Social media algorithms stoke the fires of division, locking us into echo chambers that reinforce our preexisting beliefs and shut down debate”，可知社交媒体算法“助长”了分裂之火，将人们困在信息茧房里。“stoke”意为“助长、煽动”。选项B（保持）、C（放置）、D（控制）均不符合语境，因此选A。 28. 答案：C 解析：定位原文第四段“beliefism is a form of bias—a mental shortcut in which we categorise people according to single characteristics or generalising assumptions... in a world that is infinitely complex, our minds use these biases to simplify and make sense of the world”，可知信念偏见反映了人类简化理解世界的方式。选项A错误，社交媒体算法是“重要原因之一”，并非“主要促成因素”；选项B错误，进化本能是信念偏见的来源，而非“证明其在群体思维中的危害”；选项D错误，信念偏见是“根据他人的信念对其分类”，而非“对他人的信念做出假设”，因此选C。 29. 答案：A 解析：定位原文第三段“We rarely interact with people from other walks of life”和第六段“We might exercise individuation, seeing people as diverse-sided individuals... We can practice perspective-taking, building empathy”，可知克服信念偏见的关键是促进不同群体间的互动，打破信息茧房。选项B错误，“制定可接受的信念清单”会加剧分裂；选项C错误，“写下被孤立的经历”无法直接对抗信念偏见；选项D错误，“强调冲突而非合作”会加重分裂，因此选A。 30. 答案：A 解析：本文核心是反对“我们与他们”的分裂思维，倡导超越信念差异，关注共享人性。选项B错误，“信念偏见的根源”只是部分内容；选项C错误，“社会分裂的危害”未突出“信念偏见”这一核心；选项D错误，“信仰的十字路口”与主旨无关，因此选A。 2025朝阳一模 C When people view photographs of frightening spiders, scientists usually observe increased electrical activity deep in their brain, in a region called the periaqueductal gray (PAG). The obvious conclusion is that the PAG controls fight-or-flight responses in threatening situations. But what if brains don’t have dedicated circuits for fighting and flighting? Is threat detection really a primary mode of the brain with its own neural circuitry (神经回路)? This idea of fight-or-flight circuits originates from the “triune brain” model dividing brains into the so-called lizard brain, including the PAG, for survival functions, the “limbic system” for emotions, and the “neocortex” for higher-order reasoning. If you know anything about evolution, you might know that something is spurious. The only animal on this planet with a lizard brain is a lizard. Also, there’s no unified limbic “system” or the “neocortex” for emotions or rationality (理性) respectively. So where does this leave the fight-or-flight story? Recently, changes in PAG activity have been observed in nonthreatening tasks, such as reading a fascinating article about the brain. The PAG has also been proved to regulate bodily systems continuously. Incidentally, this explains why antianxiety medications manage anxiety disorders without healing them. They target brain circuits that aren’t dedicated to anxiety or fight-or-flight but that simply regulate the body. Researches on the brain’s powers also cast new light on our understandings. In daily life, individuals might feel like they perceive and react to various stimuli as either threatening or safe. Actually, the brain, using past experiences, skillfully predicts and formulates action plans in anticipation of events before stimuli occur. The brain’s preparatory mechanism, metabolically (新陈代谢地) expensive, involves complicated neuronal activities for sustaining multiple predictions and corresponding action plans over extended periods. But the experiences in fight-or-flight responses crafted by your brain don’t reveal how it works. Moreover, your brain, which does much more than avoid threats, handles a complex social world full of uncertainty. That’s a recipe for stress. And what is stress? It reflects energy preparation for anticipated challenges. Your brain often makes sense of metabolic cost of uncertainty, if it drags on for long enough, as anxiety, but that’s not a must. People seek uncertainty out—and its positive cousin, novelty—when they try new foods or learn new skills. Uncertainty is common, but these days, with social media and round-the-clock news coverage, it sometimes bombards us and leaves us anxious and exhausted. But these feelings don’t emerge from fight-or-flight circuits. They may just mean, in an ever changing and only partly predictable world, that you’re doing something hard. 28. The author asks the questions in Paragraph 1 to ________. A. evaluate a statement B. justify an assumption C. challenge an argument D. introduce a new concept 29. Which does the underlined word “spurious” in Paragraph 2 probably mean? A. Rare. B. Wrong. C. Limited. D. Unexpected. 30. What can we infer from this passage? A. Hunting for novelty reduces stress and anxiety. B. Medicine targeting PAG cures anxiety disorders. C. Uncertainty is a product of modern media and news cycles. D. The “Fight or Flight” idea overlooks the brain’s predictive power. 文章主旨：本文质疑了“战斗或逃跑”神经回路的传统观点，指出中脑导水管周围灰质（PAG）并非仅负责威胁反应，还参与非威胁任务和身体系统调节。文章强调大脑的核心功能是基于过往经验进行预测和制定行动计划，而非单纯检测威胁，现代社会的焦虑多源于不确定性，而非所谓的“战斗或逃跑”回路。 28. 答案：C 解析：第一段提出“大脑是否有专门的战斗或逃跑回路”“威胁检测是否是大脑的主要模式”等问题，结合第二段对“三位一体大脑”模型的批判，可知作者提出这些问题是为了挑战传统观点。选项A错误，并非“评估一个陈述”；选项B错误，并非“证明一个假设”；选项D错误，并非“引入一个新概念”，因此选C。 29. 答案：B 解析：定位原文第二段“If you know anything about evolution, you might know that something is spurious. The only animal on this planet with a lizard brain is a lizard. Also, there’s no unified limbic ‘system’ or the ‘neocortex’ for emotions or rationality respectively”，可知作者认为“三位一体大脑”模型是错误的，“spurious”意为“错误的”。选项A（罕见的）、C（有限的）、D（意外的）均不符合语境，因此选B。 30. 答案：D 解析：定位原文第四段“Actually, the brain, using past experiences, skillfully predicts and formulates action plans in anticipation of events before stimuli occur... But the experiences in fight-or-flight responses crafted by your brain don’t reveal how it works”，可知“战斗或逃跑”的观点忽视了大脑的预测能力。选项A错误，“追求新奇”并非“减少压力和焦虑”；选项B错误，抗焦虑药物只是调节身体，不能“治愈焦虑障碍”；选项C错误，不确定性并非“现代媒体和新闻周期的产物”，而是普遍存在的，因此选D。 2025海淀一模 C Imagine a not-too-distant future, where we each inhabit our own AI-driven digital filter(过滤) bubble, crafted for us alone and designed to serve corporate interests. This future resembles 1998's movie The Truman Show, where the main character unknowingly lived his entire life within a reality TV show designed by a production studio. One subset of AI, large language models (LLM),won’t turn our lives into reality TV shows Instead, personalized AI agents threaten to cage each of us in an individualized and illusory(虚假的) unreality, profiting from our digital activities and walling us off from genuine connections. Many companies are developing individualized LLM. The underlying principle is that AI will learn about the individual user and adapt accordingly. For example, if you’re a superfan of a football team, you’ll be fed updates, ads, and videos tailored to your interests 24 hours a day. Some algorithms may even learn your schedule, pushing information at you during precisely those times when you're most likely to be looking. This may sound harmless. But the next step is to use LLM to create memes, or even fake articles, feeding you conspiracy theories about rival teams. This is a miserable reality for at least two reasons. For one, there are neither computational methods or ethical incentives(激励)in place to ensure that the information you receive is true. But just as frightening as the lack of regard for the truth is an even scarier element. You will no longer live with an accurate understanding of football team that is fully compatible with anyone else’s. You will run on information generated only for yourself. This vision is unsettling, even in sports and entertainment. But what of institutions that have more direct social consequences? Education? Politics? With the fall of the press and polarization of everything, conversations around holiday table have already become impossible for many extended families. Bad as the status quo might be, stranger times lie ahead that make us long for today's echo chambers(信息茧房).Soon, our bubbles will shrink further and further, until our digital worlds involve only ourselves. In an AI-mediated future, everyone will live in a private Truman Show. As a society, we will be completely incapable of making fruitful collective decisions because we will have no shared understanding of the world. What's the way out? Find your entertainment in spaces with actual people, exchanging thoughts and creations with each other. Even online, we must keep our understanding of the world grounded in human-authored documents and artifacts. Valuing what humans create is not merely a matter of authenticity; it also ensures we focus on arguments that an author cared enough to make, on conservations that speakers cared enough to have Otherwise, The Truman Show’s premise becomes our reality, unknowingly inhabiting a fake world where our every experience is designed for profit. Even more existentially alienating? Living in a Truman Show where the director, producer and the only one watching is an AI. 27.The author mentions The Truman Show in Paragraph 1 to A. make a comparison B. illustrate a situation C. support an argument D. propose a suggestion 28.According to the passage, personalized AI agents may A. isolate individuals in false realities B. confirm the credibility of the content C. discourage the companies’ ethical incentives D. improve user behavior by feeding targeted ads 29.What can be inferred from the passage? A. Algorithms have raised concerns over privacy. B. LLMs are in great need of computational upgrade. C. Echo chambers weaken collective decision-making. D. Technology development results in polarized society. 30.Which would be the best title for this passage? A.AI and the Future of Human Interaction B.AI Will Turn Our Lives into The Truman Show C. The Truman Show Predicts AI's Impact on Media D. Algorithmic Bubbles and the Value of Human Creations 文章主旨：本文警告个性化AI（如大型语言模型LLM）可能会将人们困在个性化的虚假现实中，形成比信息茧房更封闭的“数字泡泡”，导致人们缺乏对世界的共同理解，阻碍集体决策。作者提出解决方案：重视人类创造的内容，在真实人际互动中获取信息和娱乐，避免成为AI设计的“楚门的世界”中的主角。 27. 答案：B 解析：定位原文第一段“Imagine a not-too-distant future, where we each inhabit our own AI-driven digital filter bubble... This future resembles 1998's movie The Truman Show”，可知作者提及《楚门的世界》是为了形象地说明AI可能带来的未来场景。选项A错误，并非“进行比较”；选项C错误，并非“支持一个论点”；选项D错误，并非“提出一个建议”，因此选B。 28. 答案：A 解析：定位原文第二段“personalized AI agents threaten to cage each of us in an individualized and illusory unreality, profiting from our digital activities and walling us off from genuine connections”，可知个性化AI代理可能将个人孤立在虚假现实中。选项B错误，AI不会“确认内容的可信度”；选项C错误，并非“打击公司的道德激励”；选项D错误，AI推送定向广告是为了盈利，并非“改善用户行为”，因此选A。 29. 答案：C 解析：定位原文第四段“Soon, our bubbles will shrink further and further, until our digital worlds involve only ourselves... we will be completely incapable of making fruitful collective decisions because we will have no shared understanding of the world”，可知信息茧房（数字泡泡）会削弱集体决策能力。选项A错误，文中未提及“隐私担忧”；选项B错误，未提及“LLM需要计算升级”；选项D错误，“社会极化”并非“技术发展直接导致”，而是多种因素作用的结果，因此选C。 30. 答案：D 解析：本文核心围绕“算法泡泡（个性化AI带来的封闭环境）”的危害展开，强调人类创造内容的价值，呼吁避免被AI主导的虚假现实束缚。选项A错误，“AI与人类互动的未来”范围过广；选项B错误，AI不会完全将生活变成《楚门的世界》，只是形成类似的封闭环境；选项C错误，《楚门的世界》并非“预测AI对媒体的影响”，而是作为类比，因此选D。 2025丰台一模 C During WWII, mathematician Alan Turing gathered a team of logic experts to decode enemy messages, a key step toward victory. If there was one branch of mathematics which could be said they were using systematically, it was symbolic logic. However, the curriculum designs that methodically develop students ’ symbolic logic abilities are relatively rare. Symbolic logic is a form of mathematics that lets us check whether certain conclusions correctly follow from a set of facts. Consider a library book-theft case. If a suspect wasn’t in the library on the day the theft happened, they can’t be the thief. If present, they may or may not be. Being in the library doesn’t mean they stole the book. Symbolic logic turns this kind of argument into algebra(代数学), readily shared, solved and generalized for solving mysteries. For careers such as electrical engineering and computer science, the value of symbolic logic is undeniable. The fundamental building blocks of modern digital computers are circuits representing “AND” and “OR”, which are common in symbolic logic. Almost every computer program contains similar “if-then-else” logical conditions. Professional programmers must routinely write and troubleshoot such statements. But symbolic logic is not solely limited to technical fields. Rutgers University mathematician Gray Wenger argues that teaching responsible citizenship requires providing students with mathematical reasoning skills. For example, consider a debate over a study finding that most adults without high-school certificates earn less. A responsible citizen must think through the implications along the lines of the book-theft case example above. No certificate likely means lower income, but low income doesn’t mean no certificate. Training in symbolic logic makes these sorts of reasonings sink into their cognitive processes and become second nature to citizens. In addition, symbolic logic equips individuals to counter false claims by distinguishing between separate incidents and systematic evidence. The most effective strategy for this lies in preventive education—teaching people to recognize faulty reasoning at its source. Consider, for instance, some global-warming doubters use one day of cold weather as proof that global warming is a hoax(骗局). But a logical thinker can quickly spot that weather is short-term, while climate is long-term change. Some doubt the need for a separate symbolic logic course, believing other classes can do the job. In writing courses, for example, students learn how to spot illogical arguments, and elementary proofs are covered in trigonometry(三角学). Much can be gained by teaching people logical thinking without using any symbols. But only symbolic logic ensures logical strictness. A piecemeal approach isn’t enough; a full-semester course is needed for comprehensive learning. Symbolic logic is essential for intellectual development. By integrating symbolic logic into educational frameworks, we equip future generations with the intellectual tools to navigate complex challenges. Training for it must be included in the curriculum and cannot be left to chance. 28. According to the passage, symbolic logic can play a primary role when . A. a chef is creating a new recipe by combining different materials B. a student is deciding which club to join based on personal interests C. a doctor is identifying one’s illness by using signs and test findings D. a journalist is writing a feature article by interviewing local people 29. What does the phrase “second nature” underlined in Paragraph 4 probably mean? A. A routine practice. B. An automatic ability. C. An essential tool. D. A necessary quality. 30. From the passage, we learn that symbolic logic . A. tells the difference between separate incidents B. is mainly applied in solving mathematical puzzles C. questions the basic principles of modern computers D. enables people to detect and challenge false arguments 31. The author may agree that current symbolic logic education is . A. unsatisfying for the lack of symbolic logic courses B. problematic and ignores logical reasoning practice C. acceptable if alternative courses are available D. manageable and will improve naturally 文章主旨：本文介绍了符号逻辑的重要性，它是一种将逻辑推理转化为代数形式的数学分支，不仅在电子工程、计算机科学等技术领域不可或缺，还能培养公民的理性思维、帮助识别虚假主张。作者强调当前系统培养符号逻辑能力的课程较少，呼吁将其纳入正式课程。 28. 答案：C 解析：定位原文第二段“Symbolic logic is a form of mathematics that lets us check whether certain conclusions correctly follow from a set of facts”，可知符号逻辑的核心是根据一系列事实判断结论是否成立。医生通过症状和检测结果诊断疾病，符合这一逻辑应用场景。选项A（厨师创新食谱）、B（学生选择俱乐部）、D（记者采访写文章）均不涉及“事实→结论”的逻辑推导，因此选C。 29. 答案：B 解析：定位原文第四段“Training in symbolic logic makes these sorts of reasonings sink into their cognitive processes and become second nature to citizens”，可知符号逻辑训练能让理性推理融入认知过程，成为“本能反应”。“second nature”意为“习惯性的、自动的能力”。选项A（常规练习）、C（核心工具）、D（必要品质）均不符合语境，因此选B。 30. 答案：D 解析：定位原文第五段“symbolic logic equips individuals to counter false claims by distinguishing between separate incidents and systematic evidence”，可知符号逻辑能让人们识别并反驳虚假主张。选项A错误，符号逻辑是“区分个别事件和系统证据”，而非仅“区分个别事件”；选项B错误，并非“主要应用于解决数学难题”；选项C错误，符号逻辑是现代计算机的基础，并非“质疑其基本原则”，因此选D。 31. 答案：A 解析：定位原文第一段“However, the curriculum designs that methodically develop students’ symbolic logic abilities are relatively rare”和最后一段“A piecemeal approach isn’t enough; a full-semester course is needed for comprehensive learning”，可知作者认为当前符号逻辑教育缺乏系统课程，不够令人满意。选项B错误，未说“忽视逻辑推理练习”；选项C错误，作者认为零散的替代课程不够；选项D错误，认为需要主动纳入课程，而非“自然改善”，因此选A。 2025石景山一模 C ( zebrafish )Researchers from the NeuroMind Institute have developed a new system that uses predator (捕食者) robots to chase (追逐) larval (幼体的) zebrafish in an open water. This innovative approach is helping scientists study how the young fish rapidly learn and adapt in real-world conditions. Larval zebrafish are a valuable tool for neuroscientists because their transparency enables easy study of the brain and behaviour. However, it’s been difficult for scientists to study learning in these developing vertebrates (脊椎动物)—an important part of understanding how the brain works. Previous research found young zebrafish could learn simple associations. But this type of learning happens slowly and often unreliably, and it was still unclear whether days-old zebrafish can learn fast enough to use their memory in natural situations, like recognizing and avoiding new predators. The researchers thought that traditional ways of testing learning in larval zebrafish in the lab, where the conditions were far from what the fish would encounter in the wild, might not be effective for uncovering how the fish learn. To model a real-life situation, the researchers used small robotic cylinders (圆柱体), with some programmed to show predator-like characteristics. The researchers created the dynamics: they first placed a robot that stayed still with a free-swimming zebrafish; after the robot chased the fish for a minute, the fish began avoiding the robot’s area for more than an hour—a big change from the non-avoidant behaviour before the chase experience. When a second robot was introduced that did not chase the fish, the zebrafish only avoided the chasing robot, showing that they could distinguish between a threat and a non-threat. Using this system, the researchers made an unexpected discovery that not only could larval zebrafish learn extremely quickly in a more natural context, but they could also do so just five days after beginning their lives as single cells. This was particularly surprising given the fact that a developing zebrafish larva contains just one percent or so of the neurons (神经元) in its adult form. The findings suggest that some essential learning abilities, like recognizing predators, emerge early in life and are critical for survival. Further brain imaging reveals that different regions of the zebrafish brain are involved in this rapid learning: the hindbrain, a region controlling essential functions, responds to the approaching predator; the forebrain, a region associated with learning and planning, encodes the presence of the predator robot; and the habenula, another brain area, signals avoidance outcomes. All these regions are necessary for learning, and silencing any of them removes the ability of the fish to learn. It is believed that the new work could offer insights into how other brains process real-world threats. 27. Why did the researchers develop a new system with larval zebrafish? A. To observe their hunting behaviours. B. To identify their simple associations. C. To examine their brain characteristics. D. To uncover their learning in natural settings. 28. What does the underlined word “dynamics” in Paragraph 4 probably mean? A. Interaction. B. Function. C. Structure. D. Standard. 29. What can we learn about larval zebrafish? A. They can distinguish between robots and fish. B. They learn fast through a multi-regional brain network. C. They develop learning abilities when reaching adulthood. D. They can recognise predators with much neurons needed. 30. What will the author most probably discuss in the paragraph that follows? A. Explaining the robot design used in the experiment. B. Exploring danger processing in other species’ brains. C. Analysing the connection between learning and planning. D. Describing zebrafish behaviours in different surroundings. 文章主旨：本文介绍了研究人员开发的一种新系统，利用捕食者机器人追逐幼体斑马鱼，以模拟自然环境，研究其学习能力。研究发现，仅出生五天的斑马鱼就能快速学习识别并躲避威胁，这一过程涉及大脑多个区域的协同作用，该研究为理解其他物种的大脑威胁处理机制提供了参考。 27. 答案：D 解析：定位原文第二段“it’s been difficult for scientists to study learning in these developing vertebrates... Previous research found young zebrafish could learn simple associations. But this type of learning happens slowly and often unreliably, and it was still unclear whether days-old zebrafish can learn fast enough to use their memory in natural situations”，可知研究人员开发新系统是为了在自然场景下探究斑马鱼的学习能力。选项A错误，并非“观察捕猎行为”；选项B错误，“识别简单关联”是之前的研究结果；选项C错误，并非“检查大脑特征”，因此选D。 28. 答案：A 解析：定位原文第四段“they first placed a robot that stayed still with a free-swimming zebrafish; after the robot chased the fish for a minute, the fish began avoiding the robot’s area... when a second robot was introduced that did not chase the fish, the zebrafish only avoided the chasing robot”，可知“dynamics”指的是机器人与斑马鱼之间的“互动”。选项B（功能）、C（结构）、D（标准）均不符合语境，因此选A。 29. 答案：B 解析：定位原文最后一段“Further brain imaging reveals that different regions of the zebrafish brain are involved in this rapid learning: the hindbrain... responds to the approaching predator; the forebrain... encodes the presence of the predator robot; and the habenula... signals avoidance outcomes”，可知斑马鱼通过大脑多个区域的协同网络快速学习。选项A错误，斑马鱼是区分“追逐的机器人和不追逐的机器人”，而非“机器人和鱼”；选项C错误，学习能力在幼体阶段就已具备；选项D错误，幼体斑马鱼神经元数量极少但仍能识别捕食者，因此选B。 30. 答案：B 解析：定位原文最后一段“It is believed that the new work could offer insights into how other brains process real-world threats”，可知该研究为理解其他物种的大脑威胁处理机制提供了参考，因此接下来可能探讨其他物种的大脑危险处理过程。选项A错误，“解释机器人设计”并非后续重点；选项C错误，“分析学习与规划的关联”是当前研究已涉及的内容；选项D错误，“描述不同环境下的斑马鱼行为”与最后一段的研究延伸方向不符，因此选B。 2025东城二模 C Neuroscientists usually investigate one brain at a time. They observe how neurons(神经元)fire as a person reads certain words, for example, or plays a video game. As social animals, however, those same scientists do much of their work together — brainstorming hypotheses, puzzling over problems and fine-tuning experimental designs. Increasingly, researchers are bringing that reality into how they study brains. Collective neuroscience, as some practitioners call it, is a rapidly growing field of research. An early, consistent finding is that when people converse or share an experience, their brain waves synchronize. Neurons in corresponding locations of the different brains fire at the same time, creating matching patterns, like dancers moving together. The experience of “being on the same wavelength” as another person is real, and it is visible in the activity of the brain. Such work is beginning to reveal new levels of richness and complexity in sociability. In classrooms where students are engaged with the teacher for example, their patterns of brain processing begin to synchronize with that teacher’s — and greater synchrony may mean better learning. Couples exhibit higher degrees of brain synchrony than non-romantic pairs, as do close friends compared with more distant acquaintances. But much about the phenomenon remains mysterious — even scientists occasionally use the word “magic” when talking about it. One straightforward explanation could be that synchrony between brains is a result of shared experience or simply a sign that we are hearing or seeing the same thing as someone else. But the newest research suggests that synchrony is more than that — or can be. Researchers are discovering synchrony in humans and other species, and they are mapping its choreography — its rhythm, timing and undulations(波动) — to better understand what benefits it may give us. Given that synchronized experiences are often enjoyable, researchers suspect this phenomenon is beneficial: it helps us interact and may have facilitated the evolution of sociality. This new kind of brain research might also cast light on why we don’t always “click” with someone or why social isolation(孤立) is so harmful to physical and mental health. With synchrony and other levels of neural interaction, humans teach and learn, forge friendships and romances, and cooperate and converse. We are driven to connect, and synchrony is one way our brains help us do it. 28. According to the passage, collective neuroscience ______. A. collects and refines research on neurons B. analyses activities of one brain at a time C. promotes connections among neuroscientists D. focuses on studying brains in interactive groups 29. What can we infer about brain synchrony from Paragraph 4? A. It can benefit other species. B. Its mechanism has changed. C It demands further investigations. D. Findings about it are contradictory. 30. Which of the following may result from brain synchrony? A. Increasing popularity among peers. B. Better cooperation among teammates. C. Improved techniques for conversations. D. More shared experiences between a couple. 文章主旨：本文介绍了“集体神经科学”这一新兴领域，研究发现人们在交流或共享体验时，大脑电波会同步，这种同步与学习效果、亲密关系等相关。文章探讨了大脑同步现象的意义，认为它有助于人类互动和社会性进化，同时也指出这一现象仍有许多未解之谜。 28. 答案：D 解析：定位原文第一段“Neuroscientists usually investigate one brain at a time... Increasingly, researchers are bringing that reality into how they study brains”和第二段“Collective neuroscience... is a rapidly growing field of research. An early, consistent finding is that when people converse or share an experience, their brain waves synchronize”，可知集体神经科学聚焦于研究互动群体中的大脑活动。选项A错误，并非“收集和完善神经元研究”；选项B错误，“一次分析一个大脑”是传统神经科学的做法；选项C错误，并非“促进神经科学家之间的联系”，因此选D。 29. 答案：C 解析：定位原文第四段“But much about the phenomenon remains mysterious — even scientists occasionally use the word ‘magic’ when talking about it... the newest research suggests that synchrony is more than that — or can be. Researchers are discovering synchrony in humans and other species, and they are mapping its choreography to better understand what benefits it may give us”，可知大脑同步现象仍有许多未知之处，需要进一步研究。选项A错误，文中未说“能造福其他物种”；选项B错误，“机制已改变”无依据；选项D错误，未提及“研究结果矛盾”，因此选C。 30. 答案：B 解析：定位原文最后一段“Given that synchronized experiences are often enjoyable, researchers suspect this phenomenon is beneficial: it helps us interact and may have facilitated the evolution of sociality”，可知大脑同步有助于互动和社会性进化，队友间的良好合作可能源于此。选项A错误，“在同龄人中更受欢迎”无依据；选项C错误，“提升沟通技巧”并非同步直接导致；选项D错误，“情侣间更多共享体验”是同步的原因而非结果，因此选B。 2025西城二模 C When people hear “artificial intelligence,” many envision “big data.” There’s a reason for that: some of the most important AI breakthroughs in the past decade have relied on enormous data sets. But AI is not only about large data sets, and research in “small data” approaches has grown extensively over the past decade—with so-called transfer learning as an especially promising example. Also known as “fine-tuning,” transfer learning is helpful in settings where you have little data on the task of interest but abundant data on a related problem. The way it works is that you first train a model using a big data set and then retrain slightly using a smaller data set related to your specific problem. Research in transfer learning approaches has grown impressively over the past 10 years. In a new report for Georgetown University’s Center for Security and Emerging Technology (CSET), we examined current and projected progress in scientific research across “small data” approaches. Our analysis found that transfer learning stands out as a category that has experienced the most consistent and highest research growth on average since 2010. This growth has even outpaced the larger and more established field of reinforcement learning, which in recent years has attracted widespread attention. Small data approaches such as transfer learning offer numerous advantages over more data-intensive methods. By enabling the use of AI with less data, they can bolster progress in areas where little or no data exist, such as in forecasting natural disasters that occur relatively rarely or in predicting the risk of disease for a population set that does not have digital health records. Another way of thinking about the value of transfer learning is in terms of generalization. A recurring challenge in the use of AI is that models need to “generalize” beyond their training data. Because transfer learning models work by transferring knowledge from one task to another, they are very helpful in improving generalization in the new task, even if only limited data were available. Moreover, by using pretrained models, transfer learning can speed up training time and could also reduce the amount of computational resources needed to train algorithms (算法). This efficiency is significant, considering that the process of training one large neural (神经系统的) network requires considerable energy. Despite the growth in research, transfer learning has received relatively little visibility. The existence of techniques such as transfer learning does not seem to have reached the awareness of the broader space of policy makers and business leaders in positions of making important decisions about AI funding and adoption. By acknowledging the success of small data techniques like transfer learning—and distributing resources to support their widespread use—we can help overcome some of the common misconceptions regarding the role of data in AI and facilitate innovation in new directions. 27. What does the underlined word “bolster” in Paragraph 3 probably mean? A. Promote. B. Seek. C. Track. D. Monitor. 28. In which of the following settings can transfer learning be best applied? A. Predicting the frequency of floods in Amazon rainforest. B. Designing a program that can read handwritten documents. C. Forecasting the number of people infected with an unknown illness. D. Predicting house prices based on basic features like area and location. 29. What is the writer’s attitude towards transfer learning? A. Doubtful. B. Optimistic. C. Critical. D. Unconcerned. 30. Which would be the best title for the passage? A. Transfer Learning: Where Does It Best Fit? B. Small Data Are Also Crucial for Advancing AI C. Transfer Learning Powers Technological Advances D. Big Data vs. Small Data: Which Is the Future of AI? 文章主旨：本文介绍了“迁移学习（transfer learning）”这一“小数据”AI方法，即先通过大数据训练模型，再用相关小数据微调，以适应特定任务。文章指出迁移学习在研究中增长迅速，具有无需大量数据、提升泛化能力、提高训练效率等优势，但目前尚未被政策制定者和商界领袖广泛关注，作者呼吁重视其价值。 27. 答案：A 解析：定位原文第三段“By enabling the use of AI with less data, they can bolster progress in areas where little or no data exist”，可知迁移学习能在数据稀缺的领域“促进”进展。“bolster”意为“促进、支持”。选项B（寻求）、C（追踪）、D（监测）均不符合语境，因此选A。 28. 答案：C 解析：定位原文第三段“such as in forecasting natural disasters that occur relatively rarely or in predicting the risk of disease for a population set that does not have digital health records”，可知迁移学习适用于数据稀缺的场景。预测未知疾病的感染人数，相关数据较少，符合这一应用场景。选项A错误，亚马逊雨林洪水频率有一定历史数据；选项B错误，手写文档识别有大量训练数据；选项D错误，房价预测有丰富的特征数据，因此选C。 29. 答案：B 解析：作者在文中强调迁移学习的多种优势（适用于小数据场景、提升泛化能力、提高效率），并呼吁政策制定者和商界领袖关注其价值，可见态度是乐观的。选项A（怀疑的）、C（批判的）、D（不关心的）均不符合，因此选B。 30. 答案：B 解析：本文核心围绕“迁移学习”这一“小数据”AI方法展开，强调其在AI发展中的重要性，打破了“AI仅依赖大数据”的误区。选项A错误，“迁移学习的适用场景”只是部分内容；选项C错误，“迁移学习推动技术进步”未突出“小数据”这一核心；选项D错误，文章并非对比“大数据与小数据的未来”，而是强调小数据方法的重要性，因此选B。 2025朝阳二模 C Science begins with mystery. Arguably, the two greatest mysteries are the universe and awareness-the vast world out there and the powerful world within. Scientists attracted by one can be called to study the other, led by the thought that these mysteries are connected. Science writer George Musser’s book Putting Ourselves Back in the Equation (方程) reviews their progress: Can physics unlock the mystery of awareness? Does awareness form the basis of fundamental physics? The result is an ambitious but ultimately disappointing tour, filled with breathless encounters with well-known scientists. Representative of the cast is MIT’s Max Tegmark, who tells Musser, “If you look at the problems that were still difficult to answer in fundamental physics, pretty much all of them trace back to awareness.” The book shows how physicists are contributing to understanding the mind, continuing a long history of physicists exploring other fields. Musser integrates physics with neuroscience, economies, mathematics and more. Yet a key source of local knowledge is obviously absent — psychology. An early example: Musser rightly applauds physicists’ contributions to artificial neural networks but is overly trustful of their implications, declaring that “ChatGPT” is already starting to demonstrate a generalized intelligence like that of humans. Interviewing more psychologists could have exposed the considerable gap remaining. What about awareness itself? Why are some states associated with felt experience, such as the pain of a headache and the sight of a sunset, but others not? Musser’s focus is integrated information theory(IIT). IIT begins with five self-evident principles of awareness: awareness exists, and it is structured, specific, unified, and definite. It then concludes assumptions concerning the causal structure of awareness systems, identifying awareness with integrated information. Finally, IIT offers a mathematical measure of this quantity: an equation for awareness. However, despite its enthusiasts, IIT has significant problems. Its working basis is unfounded, and serious doubts surround its testability and definability. Of course, a theory of awareness must detail when, why, and to what degree we are aware of ourselves. Here, psychology’s absence is most obvious. Over the last century, psychological research has revealed countless phenomena of awareness, from models that alter awareness to methods that stimulate unawareness; from extraordinary disorders of awareness to careful studies of metacognition (元认知). Although questionable, such phenomena are the data that any scientific theory of awareness must account for. Yet these phenomena and ideas are almost nowhere in the book. Of course, psychology has not solved awareness, but one cannot hope to explain awareness without confronting these data. Awareness is genuinely mysterious. So is fundamental physics. But hoping that physics can solve awareness while excluding other approaches is only a recipe for more mystery, not less. 28. What can we learn from Musser’s book? A. IIT is acknowledged for its testability and definability. B. Musser underestimates the impact of artificial intelligence. C. Using psychology is a trend for physicists exploring other fields. D. Musser highlights physicists’ efforts to uncover the secret of awareness. 29. What can be inferred from the passage? A. The link between physics and awareness has been established. B. There is no doubt about the principles of an equation for awareness. C. A multidisciplinary approach is essential to understanding awareness. D Study of awareness needs to ignore related psychological phenomena. 30. Regarding Musser’s view in his book, the author is . A. neutral B. disapproving C. supportive D. uncertain 文章主旨：本文评论了科学作家George Musser的著作《Putting Ourselves Back in the Equation》，该书探讨了物理学与意识的关联，但作者认为该书存在明显缺陷——忽视了心理学的重要性，而意识研究需要多学科方法，仅靠物理学无法解开意识之谜。 28. 答案：D 解析：定位原文第二段“Representative of the cast is MIT’s Max Tegmark, who tells Musser, ‘If you look at the problems that were still difficult to answer in fundamental physics, pretty much all of them trace back to awareness’”和第三段“The book shows how physicists are contributing to understanding the mind”，可知Musser的书重点突出了物理学家在探索意识奥秘方面的努力。选项A错误，书中的整合信息理论（IIT）存在可测试性和可定义性问题；选项B错误，Musser高估了AI的智能，而非“低估”；选项C错误，书中未体现“物理学家探索其他领域时重视心理学”，反而忽视了心理学，因此选D。 29. 答案：C 解析：定位原文第三段“Yet a key source of local knowledge is obviously absent — psychology”和第五段“Of course, psychology has not solved awareness, but one cannot hope to explain awareness without confronting these data”，可知作者认为意识研究不能缺少心理学，需要多学科方法。选项A错误，物理学与意识的关联尚未建立；选项B错误，意识方程的原则存在争议；选项D错误，与“不能忽视心理学现象”相悖，因此选C。 30. 答案：B 解析：作者在文中明确指出Musser的书“ultimately disappointing”，批评其忽视心理学的重要性，认为仅靠物理学无法解释意识，可见态度是不赞同的。选项A（中立的）、C（支持的）、D（不确定的）均不符合，因此选B。 2025海淀二模 C The idea that aging reduces adults’ ability to imagine, a common theme in children’s literature, is contradicted by psychological research. While children are often portrayed as more imaginative, research indicates that adults not only keep this ability but sometimes surpass children in imaginative thinking. Children are frequently celebrated for boundless imagination. Yet, research reveals that their make-believe games often center around realistic scenarios, such as cooking and cleaning, as demonstrated in a 2020 study published in Journal of Cognition and Development. Another study, lasting for four decades, also suggests that children are not naturally more imaginative than adults; their limitations result from a lack of knowledge and expertise to effectively use their imaginative capacity as adults. Imagination may have evolved for considering alternatives to reality, but we use it most naturally to explore close alternatives, like preparing a different meal, rather than far alternatives, like riding on clouds. When we use imagination to envision far alternatives — to innovate or invent — we’re not digging into an inborn appreciation of the extraordinary; we’re using a tool designed to explore the ordinary. When considering alternatives to reality, we fix our attention on possibilities that are physically reasonable, statistically probable, socially conventional and morally permissible. When told about possibilities that violate such regularities, we usually deny they could happen. Generally speaking, our ideas about what could happen are firmly rooted in what we expect to happen. This mindset is also particularly apparent in young children. In a 2018 study I co-designed with psychologist Jonathan Phillips, 4-year-olds were asked to help a distressed girl who disliked going to school due to missing her mother. Among all the solutions given, they perceived the only possible solution was for her mother to do something special after school to ease her concerns. Unexpected alternatives, such as snapping fingers and making it Saturday, wearing pajamas to school or lying about school being closed, were all regarded impossible. From this, we can conclude that children’s earliest intuitions (直觉力) about possibility confuse what could happen with what should happen. Historically, the improbable event of traveling faster than a horse was considered impossible, as was traveling by air or traveling into space. Before the arrival of trains and planes, there were good reasons to think that people could travel only so far and only so fast. But these reasons were empirical (经验主义的), not logical. Imagination, on its own, lumps the improbable with the impossible, but we can combine imagination with other abilities — namely, knowledge and reflection — to separate the two. While imagination in children often subjects to expectation, adults can control their imaginative capacity for innovation by integrating it with accumulated knowledge and reflective thinking. 28. According to the first two paragraphs, we know that _________. A. children develop imagination through games B. children face limitations in acquiring knowledge C. adults are as good as children in imaginative thinking D. adults’ imaginative ability is likely to stay constant with age 29 We can infer from the passage that _________. A. expectation results from imaginative capacity B. certain practical concerns can limit imagination C. breaking regularities may lead to close alternatives D. far alternatives are more important than close alternatives 30. The 2018 study shows that children _________. A. came up with a wide range of alternatives B. were quicker to figure out solutions C. took what should happen as possibilities D. used imagination in a reasonable way 31. The underlined word “lumps” in the last paragraph probably means _________. A. mix B. match C. compare D. replace 文章主旨：本文反驳了“年龄增长会降低想象力”的观点，研究表明成年人的想象力并未衰退，甚至在某些方面超过儿童。儿童的想象力常受限于现实和认知，而成年人能结合知识和反思，控制想象力进行创新，将“不可能”与“不太可能”区分开。 28. 答案：C 解析：定位原文第一段“research indicates that adults not only keep this ability but sometimes surpass children in imaginative thinking”和第二段“another study, lasting for four decades, also suggests that children are not naturally more imaginative than adults”，可知成年人的想象力与儿童相当，甚至更强。选项A错误，文中未说儿童“通过游戏发展想象力”；选项B错误，“儿童获取知识有限”是其想象力受限的原因，而非段落核心；选项D错误，文中未说成年人想象力“随年龄保持不变”，而是“未衰退，甚至更强”，因此选C。 29. 答案：B 解析：定位原文第三段“When we use imagination to envision far alternatives — to innovate or invent — we’re not digging into an inborn appreciation of the extraordinary; we’re using a tool designed to explore the ordinary... our ideas about what could happen are firmly rooted in what we expect to happen”，可知实际考量（如物理合理性、社会常规等）会限制想象力。选项A错误，“预期源于想象力”因果倒置；选项C错误，打破常规可能带来“遥远的替代方案”，而非“接近的”；选项D错误，文中未比较“遥远替代方案”与“接近替代方案”的重要性，因此选B。 30. 答案：C 解析：定位原文第四段“we can conclude that children’s earliest intuitions about possibility confuse what could happen with what should happen”，可知2018年的研究表明儿童会将“应该发生的事”当作“可能发生的事”。选项A错误，儿童并未提出“多种替代方案”；选项B错误，文中未说儿童“更快找到解决方案”；选项D错误，儿童的想象力受限于“应该发生的事”，并非“合理使用”，因此选C。 31. 答案：A 解析：定位原文最后一段“Imagination, on its own, lumps the improbable with the impossible, but we can combine imagination with other abilities — namely, knowledge and reflection — to separate the two”，可知想象力本身会将“不太可能”与“不可能”混为一谈，而成年人能通过知识和反思将两者区分。“lumps”意为“混合”。选项B（匹配）、C（比较）、D（替代）均不符合语境，因此选A。 2025丰台二模 C In economic theories, people are typically represented as analytical agents who learn from past experiences to optimise (优化) their performance, eventually reaching a stable state in which they know how to maximise their earnings. This assumption surprised Garnier Brun at École Polytechnique in France because, as a physicist, he knew that interactions in nature—such as those between atoms—often result in chaos rather than stability. To test whether economists are correct to assume that learning from the past can help people avoid chaos, Brun and his colleagues devised a mathematical model for a game featuring hundreds of simulated players. Each player can choose between two actions, like buying or selling a stock. They also interact with each other, and the players’ decision-making is influenced by what they have done before—meaning they are able to learn from experience. The researchers could adjust the precise extent to which a player’s past experiences influenced their subsequent decision-making. They could also control the interactions between the players to make them either cooperate or compete with each other more. With all these control knobs available to them, Brun and his colleagues used methods from statistical physics to simulate different game situations on a computer. In some situations, the researchers expected that the game would always result in chaos, with players unable to learn how to optimise their performance. Economic theory would also suggest that, given the right set of parameters (参数), the players would settle into a stable state where they had mastered the game—but the researchers found this wasn’t really the case. The most likely outcome was a state that never settled. Team member Jean Martin, also at École Polytechnique, says that, in the absence of one centralised and all-knowing player who could coordinate everyone, regular players could only learn how to reach “satisficing” states. That is a level that satisfied minimum expectations, but not much more. Players gained more than they would have done by playing at random, so learning wasn’t useless, but they still gained less than they would have if past experience had allowed them to truly optimise their performance. Martin says the game model is too simple to be immediately adopted for making real-world predictions, but she sees the study as a challenge to economists to drop many assumptions that currently go into theorising processes, like merchants choosing suppliers or banks setting interest rates. The finding could also be important for simulating processes like foraging decisions by animals or for some machine-learning applications, says Toby Galla at the Institute for Cross-Disciplinary Physics and Complex Systems in Spain. 27. According to the passage, what inspired Brun to carry out the study? A. The belief that chaos is a common outcome in natural systems. B. The conflicts between economic theories and interactions in nature. C. The similarities in interaction patterns between atoms and humans. D. The inability of economic theories to predict individual behaviours. 28. What is Paragraph 2 mainly about? A. The research purpose. B. The research subjects. C. The research methods. D. The research direction. 29. What can we infer from Martin’s words? A. The “satisficing” state is an ideal outcome for players in the game. B. Players can reach optimal performance with enough past experiences. C. Centralised control is essential for players to reach their full potential. D. Learning from past experiences isn’t sufficient to maximise one’s gains. 30. What can be the best title of the passage? A. Economic Theories: Are They Reliable in Predicting Stability? B. Learning from Mistakes: The Key to Optimal Decision-Making. C. A Mathematical Model: Breaking Economic Stability Assumptions. D. Game Model Findings: Spreading Influence Across Multiple Disciplines. 文章主旨：本文介绍了一项研究，物理学家Brun团队对“人类通过学习过往经验能优化表现并达到稳定状态”这一经济学假设提出质疑。他们通过数学模型模拟游戏发现，在缺乏集中协调的情况下，参与者只能达到“满意状态”，而非最优状态，学习过往经验虽有帮助，但不足以实现收益最大化。 27. 答案：B 解析：定位原文第一段“In economic theories, people are typically represented as analytical agents who learn from past experiences to optimise their performance, eventually reaching a stable state... as a physicist, he knew that interactions in nature—such as those between atoms—often result in chaos rather than stability”，可知Brun受经济学理论与自然互动规律的冲突启发，开展了这项研究。选项A错误，“混沌是自然系统的常见结果”只是自然规律的一部分，并非核心启发点；选项C错误，文中未提及“原子与人类互动模式的相似性”；选项D错误，并非“经济学理论无法预测个体行为”，而是质疑其“稳定状态”假设，因此选B。 28. 答案：C 解析：第二段详细描述了研究的具体方法：设计数学模型、模拟数百名参与者、设定两种选择、调整过往经验的影响程度、控制参与者间的合作与竞争关系。选项A错误，“研究目的”在第一段；选项B错误，“研究对象”是模拟参与者，并非核心；选项D错误，“研究方向”未在该段体现，因此选C。 29. 答案：D 解析：定位原文第四段“players gained more than they would have done by playing at random, so learning wasn’t useless, but they still gained less than they would have if past experience had allowed them to truly optimise their performance”，可知从过往经验中学习虽有帮助，但不足以实现收益最大化。选项A错误，“满意状态”并非“理想结果”；选项B错误，文中表明即使有足够过往经验，参与者也未达到最优状态；选项C错误，“集中控制”并非“发挥全部潜力”的必要条件，只是研究中的一个变量，因此选D。 30. 答案：C 解析：本文核心围绕Brun团队的数学模型展开，该模型挑战了经济学中“人类通过学习能达到稳定最优状态”的假设，揭示了参与者实际只能达到“满意状态”。选项A错误，“经济学理论是否能预测稳定性”范围过广，未聚焦“数学模型”这一核心；选项B错误，与“学习过往经验能实现最优决策”的观点相悖；选项D错误，“模型发现的跨领域影响”只是结尾提及的延伸，并非主旨，因此选C。 学科网（北京）股份有限公司 $ 题型04 阅读理解之说明文 目录 2025东城期末 1 2025西城期末 2 2025朝阳期末 4 2025海淀期末 5 2025丰台期末 7 2025石景山期末 8 2025东城一模 9 2025西城一模 11 2025朝阳一模 12 2025海淀一模 14 2025丰台一模 15 2025石景山一模 17 2025东城二模 18 2025西城二模 19 2025朝阳二模 21 2025海淀二模 22 2025丰台二模 24 2025东城期末 C In 1993, a ship sailed into the Pacific Ocean carrying nearly 1,000 pounds of iron, then dumped it all into the waves. The next morning, the water turned a little greenish due to newly emerged phytoplankton (浮游生物). The microorganisms, which need iron to grow, draw CO₂ out of the air. Fertilizing the ocean with iron is a form of geoengineering, a set of technologies fascinating for their potential to meaningfully impact Earth’s systems, and controversial for the same reason. As the planet heats, geoengineering is employed to manipulate (控制) it in novel ways, whether researchers mist the skies with sea salt to increase how much sunlight clouds reflect or invent machines to suck greenhouse gases out of the atmosphere. Engineering home-grown plankton sounds like a less dramatic approach. But the field is so new that scientists don’t yet know whether geoengineering with microorganisms would really be a gentler form of climate intervention. Microbes, after all, play enormously consequential roles in the world around us and within us. For the best chance of keeping Earth liveable, scientists need to understand exactly how the microscopic creatures might be useful, and perhaps even preferable to the more sci-fi approaches to cooling off the world. One favored approach centers on methane, a greenhouse gas accounting for 30% of the rise in global temperature. Mary Lidstrom, a professor, is working on genetically modifying bacteria that naturally consume methane, so that the microbes pull even more gas from the air. The bacteria would live inside facilities known as bioreactors. They could be positioned near known methane sources — landfills and wetlands — to minimize the amount of methane that makes it into the air. Scaling up any kind of climate engineering is tricky. Significantly slowing warming caused by methane would require about 300,000 bioreactors to be active for 20 years. And the downstream effects are difficult to predict, let alone contain. For example, the iron put into the Pacific Ocean could have consequences for other regions of the world’s seas. Also there’s nearly as much uncertainty about intended effects. Scientists don’t know for sure how effective these concepts will prove to be. The 1993 ship expeditions, for example, didn’t stick around very long; the journeys were expensive, and scientists couldn’t remain at sea to determine what exactly was happening in the depths. Carbon stolen out of the atmosphere should remain buried for at least a century for maximum impact Yet it’s entirely possible that if the phytoplankton doesn’t sink to deep enough waters, it just gets eaten by the organisms in the surface ocean, and that CO₂ just returns to the atmosphere. The hotter the planet gets, the more attractive geoengineering will likely seem. Short of large-scale behavioral changes globally, we do seem to be committing ourselves to an engineered solution on our current routes. We might have to decide, sooner than we think, which levers we need to pull, large or small. 27. What can we learn about geoengineering? A. Ocean iron-fertilization is on the horizon. B. Sunlight-reflection method is cost-effective. C. Methane-eating bacteria are being improved. D. Gas-sucking machine is a theoretical concept. 28. What can we infer from Paragraph 4 about the geoengineering approaches? A. Their impact on the ecosystem is overlooked. B. Their implications are hard to anticipate. C. They call for decades of preparation. D. They are monitored continuously. 29. The iron-fertilization experiment may achieve its intended effects if __________. A. the target microbes survive B. carbon is stored separately C. the phytoplankton grows faster D. more deep-sea organisms are tested 30. What might be the best title for the passage? A. How Will Geoengineering Better Earth’s Systems? B. How to Cool the World Without Blocking the Sun? C How to Steal Carbon with Sci-fi Approaches? D. How Can the Engineered Solution Deliver? 2025西城期末 C In recent research, my colleagues and I set out to investigate the neural (神经的) mechanisms underlying experiences like solving challenging problems. Specifically, we wanted to understand what happens in the brain while a person engages in abstract and demanding thought — so we designed a study involving math expertise. For our experiment, we selected 22 university students majoring in mathematics, along with 22 students from other disciplines. We asked the students to watch step-by-step presentations that explained how to solve several challenging math problems. Throughout this demonstration, the students wore a cap covered with electrodes (电极) so that we could track electrical activity in their brains. After each presentation, they had to report whether they thought they had understood the presentations and how engaged they felt during the experience. We found that students whose coursework involved little mathematics showed more signs of complex activity in the prefrontal cortex (皮层) indicating how hard they were working to understand the various steps of the complex math demonstrations. However, in the other group, we noted significant activity linking the frontal and parietal (顶骨) regions of their brain. More specifically, these areas exhibited a pattern of activity described as delta waves. Some recent research suggests that these delta waves may play a crucial role in the cognitive (认知的) processing that supports deep internal concentration and the relocation of information between distant brain regions. In fact, we suspect that the long-distance delta waves we observed may play a central role whenever people engage in contextual and complex problem-solving. This suggests that engaging brain networks in this way could be useful for many tasks requiring concentration. It’s likely that when people with extensive experience in a task are deeply engaged in that effort, these same delta waves are involved, even as the specific brain networks vary. It’s also possible — though we’ll need to investigate further to be sure — that this state of deep concentration is generalizable: developing this way of thinking in one domain could help you in others. These findings suggest a trade-off that people should keep in mind—particularly as artificial intelligence and other tools offer shortcuts for various forms of problem-solving. Each time we offload a problem to a calculator or ask Al to summarize an essay, we lose an opportunity to improve our own skills and practice deep concentration. To be clear, technologies can boost our efficiency in important ways, but the seemingly “inefficient” hard work we do can be powerful, too. 28. Based on the study results, how do Math majors process challenging math problems differently from non-Math majors? A. They experience less engagement during problem-solving tasks. B. They use more areas of their prefrontal cortex to solve problems. C. They display delta wave activity linking frontal and parietal regions. D. They show slower overall brain activity during complex problem-solving. 29. What does the underlined word “trade-off” in the last paragraph probably refer to? A. Weighing short-term efficiency against lasting cognitive benefits. B. Choosing between immediate convenience and faster results. C. Sacrificing technology’s advantage for human competition. D. Deciding between human effort and advanced tools. 30. Which of the following best summarizes the author’s view on the role of “inefficient” work? A. It allows the brain to build skills that are transferable to other tasks. B. It wastes time and resources compared to using technology. C. It improves engagement but has no cognitive benefits. D. It helps those who are good at subjects like math. 31. What is the best title for the passage? A. Which to Choose: Efficiency or Hard Work? B. What Benefits Us More: AI Tools or Deep Concentration? C. Employing Complex Math Problems to Enhance Delta Waves D. Developing Expertise Improves the Brain’s Ability to Concentrate 2025朝阳期末 C According to an idea called the “inflationary (暴胀的) multiverse”, the universe has been constantly expanding, which would have created a countless number of other universes. Collectively, these cosmic islands form what scientists call a “multiverse”. However, other universes may lie beyond our scientific reach, physically and philosophically. Consequently, the methodologies by which physicists might validate (验证) the multiverse and whether such validation is even possible remain unresolved inquiries. In science, hypotheses (假说) are used to propose explanations, but from hypotheses to theories, validation of predictions is essential to see whether initial assumptions are supported or disproved. Theoretical physicist Carlo Rovelli says, “It is hard to write theories that survive the proof of reality. Few survive thanks to a basic idea: Do not trust your fancies. Keep only the ideas that can be tested.” Rovelli adds, “Today many physicists are wasting time trying to guess and jumping to conclusions. This approach has never worked previously, and it is not working now.” The multiverse might be one of those wild guesses. Now, physicists are debating whether that problem moves the multiverse from physics to metaphysics, from the world of science to that of philosophy. Philosopher Richard Dawid believes scientists could support the multiverse hypotheses without actually finding physical support. “If we have a theory that seems to work, and have come up with nothing that works better, chances are our idea is right,” he says. Not everyone is so cheerful, though. Sabine Hossenfelder, another theoretical physicist, thinks “Without making contact with observation, a theory isn’t part of the natural sciences, and not physics.” Some supporters of the multiverse claim they have found physical evidence. Andrei Linde insists “Our universe seems fine-tuned (精调的) to be favourable to life, with a four-dimensional space in which we can live. It is certain that in an eternally inflating multiverse, one of the universes should turn out like ours. Each universe can have different physical laws and fundamentals. Given limitless changes, a universe on which humans can be born will be born. The multiverse actually explains why we’re here. And our existence, therefore, helps explain why the multiverse is plausible.” Just as Einstein’s theory of general relativity could only be confirmed recently with a billion-dollar instrument called LIGO, scientists may eventually discover more direct evidence of the multiverse. So far, all of science has relied on testability, but those tests take time, and most theoreticians want to wait it out. They are not ready to shelve an idea like the multiverse — which could actually be the answer to life, the universe and everything — until and unless they can prove to themselves it doesn’t exist. And that day may never come. 28. What can we learn from the passage? A. The multiverse is recognized as a theory. B. The multiverse starts debates on human existence. C. Rovelli criticizes physicists for their baseless fancies. D. Dawid emphasizes the necessity of physical evidence. 29. What does the underlined word “plausible” in Paragraph 4 probably mean? A. Credible. B. Complex. C. Essential. D. Fascinating. 30. As for the existence of multiverse, the author is _________. A. negative B. neutral C. certain D. positive 31. Which would be the best title for this passage? A. Mysteries of Multiverse: Fine-Tuned for Life B Can Physicists Ever Prove the Multiverse Is Real? C. The Inflationary Multiverse: Proven Beyond Doubt D. Have Physicists Lost Their Minds in the Multiverse? 2025海淀期末 C As global temperatures continue to rise due to human-induced climate change, precise computerized climate models will be crucial in shedding light on how our climate will evolve in the coming years. A study published in Journal of Geophysical Research: Atmospheres by Chloe Clark and her team from UC Irvine highlights a significant issue with a commonly used climate model known as the Energy Exascale Earth System Model (E3SM). The research reveals that this model currently overestimates a key physical property of Earth’s climate system called albedo, which is the degree to which ice reflects planet-warming sunlight into space. Albedo is a measure of the reflectivity of a surface, expressed as the fraction of incoming solar radiation that is reflected by that surface back into space. It is a critical factor in determining Earth’s climate and energy balance. Surfaces with high albedo, such as snow and ice, reflect a large portion of solar energy, whereas darker surfaces like forests or oceans absorb more solar energy. “We found that with old model versions, the ice is too reflective by about five percent,” said Chloe Clarke, “Ice reflectivity was much too high.” The amount of sunlight the planet receives and reflects is important for estimating how much the planet will warm in the coming years. Previous versions of E3SM overestimated albedo because they failed to account for microphysical properties of ice in a warming world. These properties include the effects things like algae (藻类) and dust have on albedo. Dark colored algae and dust can make snow and ice less reflective and less able to reflect sunlight. To do the analysis, Clarke and her team studied satellite data to track the albedo of the Greenland Ice Sheet. They found E3SM overestimates the reflectivity of the ice sheet, meaning the model estimates less melt than what would be expected from the ice microphysical properties. But with the new ice reflectivity incorporated into the model, they discovered the Greenland Ice Sheet is melting at a rate of about six gigatons more than in older model versions. This is based on albedo measurements that are more consistent with satellite observations. Clarke hopes her team’s study stresses the importance of the seemingly minuscule properties that can have far-reaching consequences for the overall climate. “I think our work is going to help models do a much better job of helping us capture snow and ice-related climate feedbacks,” she said. Next, Clarke wants to study different icy parts of the planet to calculate how widespread the albedo discrepancy (差异) is in E3SM. “Our next steps are to get it so the model is functional globally and not just valid over Greenland,” said Clarke, who also intends to compare the new Greenland Ice Sheet melt rates to observations to measure how much more accurate the new ice albedo is. “It would be useful to apply it to glaciers in places like the Andes and Alaska.” 27. According to the passage, albedo ______. A. changes with the lightness of surfaces B. matches up with the shape of the ice or snow C. is a measure of the solar energy arriving at the surface D. varies with the direction of the incoming solar radiation 28. What can be learned from the passage? A The new model is more reliable than satellite observations. B. The previous model versions overestimate Greenland melt. C. Model adaptation improves the accuracy of climate predictions. D. Scientists have used the model globally to reduce the discrepancy. 29 What does the underlined word “minuscule” in Paragraph 7 probably mean? A. Tiny. B. Invisible. C. Plain. D. Stable. 30. The last paragraph is written to ______. A. note the limitations of the finding B. point out further research directions C. illustrate the benefits of a new approach D. sum up the changes made in E3SM 2025丰台期末 C On May 4, 2017, 395 migratory (迁徙的) birds died after crashing into a tall building in Galveston, Texas. Their migration routes are full of danger — including bright city lights. Every year, millions of birds are thought to be killed in building collisions in the US. The scene of the mass collision in Galveston was a skyscraper which has four corner spotlights and glowing green LED lights around the top. The influence of artificial lights in urban spaces is a well-known phenomenon among birds, but is still not fully understood. Birds tend to migrate by night, and use the stars to navigate. During nocturnal (夜间的) migration, they are often seen gathering around or crashing into lit-up windows or gathering around beams of projected light. The artificial glow of cities appears to attract them for stopovers. One explanation is that they may generally fly towards light sources, but also, that the artificial light confuses, traps and disorients them. Every tragedy is an opportunity to educate. The accident prompted bird conservation organisations to take action and jointly develop a science-based response. Several campaigns encouraged building owners and developers to switch off non-essential lighting to help travelling birds. Soon the first Lights Out for Birds programme was formed, which has now expanded into the statewide. Meanwhile, the Cornell Lab of Ornithology developed a way to forecast bird migration during specific times and locations, using radar. The resulting BirdCast migration forecast maps are freely accessible. Their live migration map allows anyone to check how many birds are migrating now, and where, as well as how many are predicted to pass in the near future. For one night in October, they recorded a “billion bird night”, with that number or more birds passing in a single night. Andrew Farnsworth, a migration ecologist said that early data collection suggests that Light Out is effective, even with the limited data and uncontrollable factors. Moreover, Farnsworth points out protecting species that move presents special challenges for researchers and conservationists. Migrating birds can travel thousands of kilometres, crossing and stopping in many different habitats, traversing borders, and flying through cities — which are continuing to grow around the world. There is no single solution to helping migrating birds. Instead, spreading awareness, empathy and efficient enforcement are all crucial in keeping birds safe and avoiding collisions. “We need to keep our connection to nature through our eyes, ears, nose, whatever to experience the core of all this research and observation.” 28. According to the passage, why do the migrating birds hit the buildings? A. Their routes are blocked by tall buildings. B. They attempt to avoid potential dangers. C. They are affected by the artificial lights. D. Their migration patterns have changed. 29. What can we learn from the passage? A. Light Out is thought to be helpful despite data limitations. B. BirdCast needs further improvement in tracking accuracy. C. Light Out can guide migrating birds away from artificial lights. D. BirdCast can predict the number and species of migrating birds. 30. As for helping migrating birds,which would the author agree with? A. It is worthwhile to improve Light Out. B. It is necessary to make effective regulations. C. It is crucial to work out a one-button solution. D. It is essential to integrate multiple aspects’ effort. 2025石景山期末 C Physarum polycephalum, a single-celled slime mold (黏菌), lacks a brain and survives on rotting vegetables, but it may hold valuable insights for city planning, according to the researchers from University of Toronto. This greenish-yellow organism forms a complex network that can transfer nutrients efficiently, which researchers believe could model efficient networks for moving people and goods in urban settings. “Humans aren’t the only ones facing the challenge of designing efficient, tough networks,” says Raphael Kay, a master in engineering. “The slime mold has been shaped by hundreds of millions of years of evolution (进化), so in that sense, it is far more experienced at network design than we humans ever could be.” This idea isn’t new. In 2010, Japanese and U. K. researchers created a nutrient layout resembling the Tokyo subway nodes (节点), and the slime mold grew a network similar to the real system. This experiment sparked biologically inspired adaptive network design, leading to studies of rail and road networks using slime molds. Building on this, Kay and an architecture student Anthony developed a computer model that simulates (模拟) slime molds’ network-building process, aimed at architects, city planners, and other professionals. Their model allows users to specify factors like maximum cost or minimum travel time, generating networks that meet these criteria. To test it, they applied the model to Canada’s Wonderland amusement park and Toronto’s subway system. For Wonderland, their network offered 10% faster travel time at the same cost as the existing setup and was 80% more resilient at maintaining functionality if a part was blocked. For Toronto’ s subway, they found that a network with the same travel time as the real one was 40% less easily affected to disruption. While Kay and his team aren’t suggesting a complete redesign of current networks, their model offers insights for improvements. For instance, it could pinpoint the best places to add nodes or help in planning future layouts. The model can also autonomously generate network nodes based on data like population concentration, adapting to serve specific needs. Hatton, a collaborator, believes this tool could benefit diverse fields, not just those related to the movement of people or goods, but also fields like genetic engineering and energy networks, highlighting the importance of interdisciplinary approaches. 27. What unique feature of the slime mold interests city planners? A. Its reliance on nutrients to survive. B. Its ability to form efficient networks. C. Its nature to grow in rotting vegetables. D. Its adaptive skill in a new environment. 28. Why does the author mention the experiment with Tokyo’s subway nodes? A. To highlight that the Tokyo subway system is efficient. B. To show that slime molds can recognize urban structures. C. To argue that Tokyo subway system needs to be improved. D. To suggest that slime molds can inspire unban network design. 29. What does the underlined word “resilient” probably mean in paragraph 5? A. Reliable. B. Relevant. C. Accurate. D. Advanced. 30. What can we learn from Kay and his team’s model? A. It can be applied beyond transportation. B. It can predict population growth patterns. C. It is limited to networks with simple layouts. D. It is recommended to redesign current networks. 2025东城一模 C Anyone with insomnia knows the impatience and frustration that accompanies sleeplessness. You long for a button that could instantly dampen all that mental activity. The idea of a mental switch is not far-fetched. Most neuroscientists now agree that our wakefulness is coordinated by a tiny bundle of neurons(一小束神经元)known as the “locus coeruleus"(LC),Latin for "blue dot”. It is a literal description: the neurons in the locus coeruleus have the blue colour from the production of a particular neurotransmitter, called norepinephrine. Norepinephrine raises the chance that a neuron will “fire”with an electric current. When they become active, cells in the locus coeruleus pass bundles of this neurotransmitter along their projections to other regions of the brain—enhancing the communication between the neurons in that area. There are slight differences in the process. Depending on the types of receptors they have, some neurons are more sensitive to smaller amounts of norepinephrine, while others only respond to higher thresholds. This means that, as the locus coeruleus activity rises, it will start to affect some brain areas more than others, which can have dramatic effects on things like our focus, concentration and creativity. Given the blue dot's role, it makes sense that it would be quietest at night during sleep. It is not entirely silent, however, but fires occasionally—and recent research by Anita Lüthi at the University of Lausanne suggests that this activity may determine the quality of our sleeps. Across the night, we alternate between different sleep stages. There is “rapid eye movement”(REM)sleep, which is associated with vivid dreaming and is thought to be crucial for processing and consolidating memories. Much of our rest, however, is spent in non-REM(NREM)sleep, during which the brain may engage in a deep clean, clearing away cellular waste. Measuring brain activity in dozing mice, Anita found NREM sleep was associated with temporary bursts of locus coeruleus activity every 50 seconds. As a result, the animal was more sensitive to outside stimuli, like noises—without fully waking. “It’s generating this state of enhanced vigilance(警觉),”Anita says. “It really gives you this idea that wakefulness can be graded in the brain.” The beginning of REM sleep was almost always associated with low locus coeruleus activity. “That transition to REM sleep has to be very well controlled,” says Anita, “because in REM sleep, we have atonia.” That’s the temporary paralysis(麻痹) of our body, which prevents us from physically acting out our dreams. Anita emphasises that her experiments were conducted in mice, so we still need to confirm that the blue dot plays a similar role in human sleep. If so, she suspects that altered locus coeruleus activity could be implicated in conditions—such as anxiety—that may contribute to disordered sleep. She found that exposing her laboratory mice to mild sources of stress—such as knocking on their cage—raised the blue dot's activity and increased their vigilance throughout the night, resulting in fragmented sleep. 27.What does the underlined word “they” in Paragraph 2 refer to? A. Neurons. B. Electric currents. C. Projections. D. Neurotransmitters. 28. According to the passage , what is the role of the LC? A. producing receptors B. Preserving cell sensitivity C. Monitoring brain activity D. Improving neutral connectivity 29.Which of the following may Anita Lüthi agree with? A. The blue dot fires regularly at night B. Stress has an impact on the LC activity C.vLow LC activity can help clean cellular waste D. Atonia results from sudden bursts of brain activity 30.What might be the next step of the research? A. Grading the wakefulness of human brains B. Unlocking the mechanism of sleeo disorder C. Assessing the function of the blue dot on humans D. Identifying approaches to altering the LC activity 2025西城一模 C Recently, I attended a public talk by someone whose views I “knew” I would oppose. And yet, I went. I listened, asked questions, and gave my time. While my core values weren’t transformed in those two hours, I learned something and left with a deeper appreciation for the complexity of other perspectives. In this weeknight activity, I was actively trying to tackle“beliefism,” a divisive phenomenon in which surrounding yourself with people who share your views leads to discrimination against those who disagree. In this way, beliefism deepens division and reinforces polarization—building walls instead of bridges. Indeed, beliefism is widespread in modern society. A significant part of the problem originates from the fact that we live in a world that is both physically and virtually divided. We rarely interact with people from other walks of life. Social media algorithms stoke the fires of division, locking us into echo chambers that reinforce our preexisting beliefs and shut down debate. Ultimately, where many forces are driving us apart, we must think—what can we do to unpack division and restore connection? We can begin by trying to understand the psychology of beliefism, which at its core is a form of bias—a mental shortcut in which we categorise people according to single characteristics or generalising assumptions. Indeed, in a world that is infinitely complex, our minds use these biases to simplify and make sense of the world. The thing is, when we see others only through the perspective of their political or social beliefs, we reduce and flatten them to a single dimension. Further, when people feel they are dismissed or disregarded only for their beliefs, they are left feeling isolated and misunderstood. Secondly, we can understand the tendency for beliefism as part of our evolutionary (进化的) desire to establish a community or group. The problem is that while this instinct (本能) may have once served evolutionary purposes, today, it blocks meaningful dialogue and cooperation. Indeed, research shows that overcoming beliefism has benefits. When we welcome a variety of ideas and perspectives, we are able to overcome group-think and make better decisions and judgements. What’s more, less beliefist people are generally happier, having stronger, more fulfilling relationships and broader horizons. Luckily, there are a number of relevant, research-backed psychological techniques that help build tolerance and break bias. We might exercise individuation, seeing people as diverse-sided individuals and breaking away from reductive ways of thinking. We can practice perspective-taking, building empathy (同理心) by stepping into someone else’s shoes and trying to see the world through their eyes. Ultimately, the way forward is not through divisions, but through a recognition of our shared humanity. Remind yourself that each person exists at the intersection of many identities, experiences, and beliefs. Challenge yourself to practice empathy, and remember that no one is wholly defined by the worst thing they have said or done. 27. What does the underlined word “stoke” in Paragraph 3 probably mean? A. Fuel. B. Keep. C. Put. D. Contain. 28. What can we know about beliefism from the passage? A. Social media algorithms mainly contribute to its wide spread. B. Human evolution proves its harm in establishing group-think. C. It reflects our simplified way of understanding the surrounding world. D. It leads us to make assumptions about others’ political or social beliefs. 29. Which of the following would be most effective in fighting beliefism? A. Facilitating interactions between people from different communities. B. Creating a list of acceptable beliefs for each community to follow. C. Asking people to write about their experiences of being isolated. D. Stressing conflict instead of cooperation between different beliefs. 30. Which would be the best title for the passage? A. Beyond Us and Them B. The Roots of Beliefism C. The Harm of Social Division D. At the Crossroad of Faith 2025朝阳一模 C When people view photographs of frightening spiders, scientists usually observe increased electrical activity deep in their brain, in a region called the periaqueductal gray (PAG). The obvious conclusion is that the PAG controls fight-or-flight responses in threatening situations. But what if brains don’t have dedicated circuits for fighting and flighting? Is threat detection really a primary mode of the brain with its own neural circuitry (神经回路)? This idea of fight-or-flight circuits originates from the “triune brain” model dividing brains into the so-called lizard brain, including the PAG, for survival functions, the “limbic system” for emotions, and the “neocortex” for higher-order reasoning. If you know anything about evolution, you might know that something is spurious. The only animal on this planet with a lizard brain is a lizard. Also, there’s no unified limbic “system” or the “neocortex” for emotions or rationality (理性) respectively. So where does this leave the fight-or-flight story? Recently, changes in PAG activity have been observed in nonthreatening tasks, such as reading a fascinating article about the brain. The PAG has also been proved to regulate bodily systems continuously. Incidentally, this explains why antianxiety medications manage anxiety disorders without healing them. They target brain circuits that aren’t dedicated to anxiety or fight-or-flight but that simply regulate the body. Researches on the brain’s powers also cast new light on our understandings. In daily life, individuals might feel like they perceive and react to various stimuli as either threatening or safe. Actually, the brain, using past experiences, skillfully predicts and formulates action plans in anticipation of events before stimuli occur. The brain’s preparatory mechanism, metabolically (新陈代谢地) expensive, involves complicated neuronal activities for sustaining multiple predictions and corresponding action plans over extended periods. But the experiences in fight-or-flight responses crafted by your brain don’t reveal how it works. Moreover, your brain, which does much more than avoid threats, handles a complex social world full of uncertainty. That’s a recipe for stress. And what is stress? It reflects energy preparation for anticipated challenges. Your brain often makes sense of metabolic cost of uncertainty, if it drags on for long enough, as anxiety, but that’s not a must. People seek uncertainty out—and its positive cousin, novelty—when they try new foods or learn new skills. Uncertainty is common, but these days, with social media and round-the-clock news coverage, it sometimes bombards us and leaves us anxious and exhausted. But these feelings don’t emerge from fight-or-flight circuits. They may just mean, in an ever changing and only partly predictable world, that you’re doing something hard. 28. The author asks the questions in Paragraph 1 to ________. A. evaluate a statement B. justify an assumption C. challenge an argument D. introduce a new concept 29. Which does the underlined word “spurious” in Paragraph 2 probably mean? A. Rare. B. Wrong. C. Limited. D. Unexpected. 30. What can we infer from this passage? A. Hunting for novelty reduces stress and anxiety. B. Medicine targeting PAG cures anxiety disorders. C. Uncertainty is a product of modern media and news cycles. D. The “Fight or Flight” idea overlooks the brain’s predictive power. 2025海淀一模 C Imagine a not-too-distant future, where we each inhabit our own AI-driven digital filter(过滤) bubble, crafted for us alone and designed to serve corporate interests. This future resembles 1998's movie The Truman Show, where the main character unknowingly lived his entire life within a reality TV show designed by a production studio. One subset of AI, large language models (LLM),won’t turn our lives into reality TV shows Instead, personalized AI agents threaten to cage each of us in an individualized and illusory(虚假的) unreality, profiting from our digital activities and walling us off from genuine connections. Many companies are developing individualized LLM. The underlying principle is that AI will learn about the individual user and adapt accordingly. For example, if you’re a superfan of a football team, you’ll be fed updates, ads, and videos tailored to your interests 24 hours a day. Some algorithms may even learn your schedule, pushing information at you during precisely those times when you're most likely to be looking. This may sound harmless. But the next step is to use LLM to create memes, or even fake articles, feeding you conspiracy theories about rival teams. This is a miserable reality for at least two reasons. For one, there are neither computational methods or ethical incentives(激励)in place to ensure that the information you receive is true. But just as frightening as the lack of regard for the truth is an even scarier element. You will no longer live with an accurate understanding of football team that is fully compatible with anyone else’s. You will run on information generated only for yourself. This vision is unsettling, even in sports and entertainment. But what of institutions that have more direct social consequences? Education? Politics? With the fall of the press and polarization of everything, conversations around holiday table have already become impossible for many extended families. Bad as the status quo might be, stranger times lie ahead that make us long for today's echo chambers(信息茧房).Soon, our bubbles will shrink further and further, until our digital worlds involve only ourselves. In an AI-mediated future, everyone will live in a private Truman Show. As a society, we will be completely incapable of making fruitful collective decisions because we will have no shared understanding of the world. What's the way out? Find your entertainment in spaces with actual people, exchanging thoughts and creations with each other. Even online, we must keep our understanding of the world grounded in human-authored documents and artifacts. Valuing what humans create is not merely a matter of authenticity; it also ensures we focus on arguments that an author cared enough to make, on conservations that speakers cared enough to have Otherwise, The Truman Show’s premise becomes our reality, unknowingly inhabiting a fake world where our every experience is designed for profit. Even more existentially alienating? Living in a Truman Show where the director, producer and the only one watching is an AI. 27.The author mentions The Truman Show in Paragraph 1 to A. make a comparison B. illustrate a situation C. support an argument D. propose a suggestion 28.According to the passage, personalized AI agents may A. isolate individuals in false realities B. confirm the credibility of the content C. discourage the companies’ ethical incentives D. improve user behavior by feeding targeted ads 29.What can be inferred from the passage? A. Algorithms have raised concerns over privacy. B. LLMs are in great need of computational upgrade. C. Echo chambers weaken collective decision-making. D. Technology development results in polarized society. 30.Which would be the best title for this passage? A.AI and the Future of Human Interaction B.AI Will Turn Our Lives into The Truman Show C. The Truman Show Predicts AI's Impact on Media D. Algorithmic Bubbles and the Value of Human Creations 2025丰台一模 C During WWII, mathematician Alan Turing gathered a team of logic experts to decode enemy messages, a key step toward victory. If there was one branch of mathematics which could be said they were using systematically, it was symbolic logic. However, the curriculum designs that methodically develop students ’ symbolic logic abilities are relatively rare. Symbolic logic is a form of mathematics that lets us check whether certain conclusions correctly follow from a set of facts. Consider a library book-theft case. If a suspect wasn’t in the library on the day the theft happened, they can’t be the thief. If present, they may or may not be. Being in the library doesn’t mean they stole the book. Symbolic logic turns this kind of argument into algebra(代数学), readily shared, solved and generalized for solving mysteries. For careers such as electrical engineering and computer science, the value of symbolic logic is undeniable. The fundamental building blocks of modern digital computers are circuits representing “AND” and “OR”, which are common in symbolic logic. Almost every computer program contains similar “if-then-else” logical conditions. Professional programmers must routinely write and troubleshoot such statements. But symbolic logic is not solely limited to technical fields. Rutgers University mathematician Gray Wenger argues that teaching responsible citizenship requires providing students with mathematical reasoning skills. For example, consider a debate over a study finding that most adults without high-school certificates earn less. A responsible citizen must think through the implications along the lines of the book-theft case example above. No certificate likely means lower income, but low income doesn’t mean no certificate. Training in symbolic logic makes these sorts of reasonings sink into their cognitive processes and become second nature to citizens. In addition, symbolic logic equips individuals to counter false claims by distinguishing between separate incidents and systematic evidence. The most effective strategy for this lies in preventive education—teaching people to recognize faulty reasoning at its source. Consider, for instance, some global-warming doubters use one day of cold weather as proof that global warming is a hoax(骗局). But a logical thinker can quickly spot that weather is short-term, while climate is long-term change. Some doubt the need for a separate symbolic logic course, believing other classes can do the job. In writing courses, for example, students learn how to spot illogical arguments, and elementary proofs are covered in trigonometry(三角学). Much can be gained by teaching people logical thinking without using any symbols. But only symbolic logic ensures logical strictness. A piecemeal approach isn’t enough; a full-semester course is needed for comprehensive learning. Symbolic logic is essential for intellectual development. By integrating symbolic logic into educational frameworks, we equip future generations with the intellectual tools to navigate complex challenges. Training for it must be included in the curriculum and cannot be left to chance. 28. According to the passage, symbolic logic can play a primary role when . A. a chef is creating a new recipe by combining different materials B. a student is deciding which club to join based on personal interests C. a doctor is identifying one’s illness by using signs and test findings D. a journalist is writing a feature article by interviewing local people 29. What does the phrase “second nature” underlined in Paragraph 4 probably mean? A. A routine practice. B. An automatic ability. C. An essential tool. D. A necessary quality. 30. From the passage, we learn that symbolic logic . A. tells the difference between separate incidents B. is mainly applied in solving mathematical puzzles C. questions the basic principles of modern computers D. enables people to detect and challenge false arguments 31. The author may agree that current symbolic logic education is . A. unsatisfying for the lack of symbolic logic courses B. problematic and ignores logical reasoning practice C. acceptable if alternative courses are available D. manageable and will improve naturally 2025石景山一模 C ( zebrafish )Researchers from the NeuroMind Institute have developed a new system that uses predator (捕食者) robots to chase (追逐) larval (幼体的) zebrafish in an open water. This innovative approach is helping scientists study how the young fish rapidly learn and adapt in real-world conditions. Larval zebrafish are a valuable tool for neuroscientists because their transparency enables easy study of the brain and behaviour. However, it’s been difficult for scientists to study learning in these developing vertebrates (脊椎动物)—an important part of understanding how the brain works. Previous research found young zebrafish could learn simple associations. But this type of learning happens slowly and often unreliably, and it was still unclear whether days-old zebrafish can learn fast enough to use their memory in natural situations, like recognizing and avoiding new predators. The researchers thought that traditional ways of testing learning in larval zebrafish in the lab, where the conditions were far from what the fish would encounter in the wild, might not be effective for uncovering how the fish learn. To model a real-life situation, the researchers used small robotic cylinders (圆柱体), with some programmed to show predator-like characteristics. The researchers created the dynamics: they first placed a robot that stayed still with a free-swimming zebrafish; after the robot chased the fish for a minute, the fish began avoiding the robot’s area for more than an hour—a big change from the non-avoidant behaviour before the chase experience. When a second robot was introduced that did not chase the fish, the zebrafish only avoided the chasing robot, showing that they could distinguish between a threat and a non-threat. Using this system, the researchers made an unexpected discovery that not only could larval zebrafish learn extremely quickly in a more natural context, but they could also do so just five days after beginning their lives as single cells. This was particularly surprising given the fact that a developing zebrafish larva contains just one percent or so of the neurons (神经元) in its adult form. The findings suggest that some essential learning abilities, like recognizing predators, emerge early in life and are critical for survival. Further brain imaging reveals that different regions of the zebrafish brain are involved in this rapid learning: the hindbrain, a region controlling essential functions, responds to the approaching predator; the forebrain, a region associated with learning and planning, encodes the presence of the predator robot; and the habenula, another brain area, signals avoidance outcomes. All these regions are necessary for learning, and silencing any of them removes the ability of the fish to learn. It is believed that the new work could offer insights into how other brains process real-world threats. 27. Why did the researchers develop a new system with larval zebrafish? A. To observe their hunting behaviours. B. To identify their simple associations. C. To examine their brain characteristics. D. To uncover their learning in natural settings. 28. What does the underlined word “dynamics” in Paragraph 4 probably mean? A. Interaction. B. Function. C. Structure. D. Standard. 29. What can we learn about larval zebrafish? A. They can distinguish between robots and fish. B. They learn fast through a multi-regional brain network. C. They develop learning abilities when reaching adulthood. D. They can recognise predators with much neurons needed. 30. What will the author most probably discuss in the paragraph that follows? A. Explaining the robot design used in the experiment. B. Exploring danger processing in other species’ brains. C. Analysing the connection between learning and planning. D. Describing zebrafish behaviours in different surroundings. 2025东城二模 C Neuroscientists usually investigate one brain at a time. They observe how neurons(神经元)fire as a person reads certain words, for example, or plays a video game. As social animals, however, those same scientists do much of their work together — brainstorming hypotheses, puzzling over problems and fine-tuning experimental designs. Increasingly, researchers are bringing that reality into how they study brains. Collective neuroscience, as some practitioners call it, is a rapidly growing field of research. An early, consistent finding is that when people converse or share an experience, their brain waves synchronize. Neurons in corresponding locations of the different brains fire at the same time, creating matching patterns, like dancers moving together. The experience of “being on the same wavelength” as another person is real, and it is visible in the activity of the brain. Such work is beginning to reveal new levels of richness and complexity in sociability. In classrooms where students are engaged with the teacher for example, their patterns of brain processing begin to synchronize with that teacher’s — and greater synchrony may mean better learning. Couples exhibit higher degrees of brain synchrony than non-romantic pairs, as do close friends compared with more distant acquaintances. But much about the phenomenon remains mysterious — even scientists occasionally use the word “magic” when talking about it. One straightforward explanation could be that synchrony between brains is a result of shared experience or simply a sign that we are hearing or seeing the same thing as someone else. But the newest research suggests that synchrony is more than that — or can be. Researchers are discovering synchrony in humans and other species, and they are mapping its choreography — its rhythm, timing and undulations(波动) — to better understand what benefits it may give us. Given that synchronized experiences are often enjoyable, researchers suspect this phenomenon is beneficial: it helps us interact and may have facilitated the evolution of sociality. This new kind of brain research might also cast light on why we don’t always “click” with someone or why social isolation(孤立) is so harmful to physical and mental health. With synchrony and other levels of neural interaction, humans teach and learn, forge friendships and romances, and cooperate and converse. We are driven to connect, and synchrony is one way our brains help us do it. 28. According to the passage, collective neuroscience ______. A. collects and refines research on neurons B. analyses activities of one brain at a time C. promotes connections among neuroscientists D. focuses on studying brains in interactive groups 29. What can we infer about brain synchrony from Paragraph 4? A. It can benefit other species. B. Its mechanism has changed. C It demands further investigations. D. Findings about it are contradictory. 30. Which of the following may result from brain synchrony? A. Increasing popularity among peers. B. Better cooperation among teammates. C. Improved techniques for conversations. D. More shared experiences between a couple. 2025西城二模 C When people hear “artificial intelligence,” many envision “big data.” There’s a reason for that: some of the most important AI breakthroughs in the past decade have relied on enormous data sets. But AI is not only about large data sets, and research in “small data” approaches has grown extensively over the past decade—with so-called transfer learning as an especially promising example. Also known as “fine-tuning,” transfer learning is helpful in settings where you have little data on the task of interest but abundant data on a related problem. The way it works is that you first train a model using a big data set and then retrain slightly using a smaller data set related to your specific problem. Research in transfer learning approaches has grown impressively over the past 10 years. In a new report for Georgetown University’s Center for Security and Emerging Technology (CSET), we examined current and projected progress in scientific research across “small data” approaches. Our analysis found that transfer learning stands out as a category that has experienced the most consistent and highest research growth on average since 2010. This growth has even outpaced the larger and more established field of reinforcement learning, which in recent years has attracted widespread attention. Small data approaches such as transfer learning offer numerous advantages over more data-intensive methods. By enabling the use of AI with less data, they can bolster progress in areas where little or no data exist, such as in forecasting natural disasters that occur relatively rarely or in predicting the risk of disease for a population set that does not have digital health records. Another way of thinking about the value of transfer learning is in terms of generalization. A recurring challenge in the use of AI is that models need to “generalize” beyond their training data. Because transfer learning models work by transferring knowledge from one task to another, they are very helpful in improving generalization in the new task, even if only limited data were available. Moreover, by using pretrained models, transfer learning can speed up training time and could also reduce the amount of computational resources needed to train algorithms (算法). This efficiency is significant, considering that the process of training one large neural (神经系统的) network requires considerable energy. Despite the growth in research, transfer learning has received relatively little visibility. The existence of techniques such as transfer learning does not seem to have reached the awareness of the broader space of policy makers and business leaders in positions of making important decisions about AI funding and adoption. By acknowledging the success of small data techniques like transfer learning—and distributing resources to support their widespread use—we can help overcome some of the common misconceptions regarding the role of data in AI and facilitate innovation in new directions. 27. What does the underlined word “bolster” in Paragraph 3 probably mean? A. Promote. B. Seek. C. Track. D. Monitor. 28. In which of the following settings can transfer learning be best applied? A. Predicting the frequency of floods in Amazon rainforest. B. Designing a program that can read handwritten documents. C. Forecasting the number of people infected with an unknown illness. D. Predicting house prices based on basic features like area and location. 29. What is the writer’s attitude towards transfer learning? A. Doubtful. B. Optimistic. C. Critical. D. Unconcerned. 30. Which would be the best title for the passage? A. Transfer Learning: Where Does It Best Fit? B. Small Data Are Also Crucial for Advancing AI C. Transfer Learning Powers Technological Advances D. Big Data vs. Small Data: Which Is the Future of AI? 2025朝阳二模 C Science begins with mystery. Arguably, the two greatest mysteries are the universe and awareness-the vast world out there and the powerful world within. Scientists attracted by one can be called to study the other, led by the thought that these mysteries are connected. Science writer George Musser’s book Putting Ourselves Back in the Equation (方程) reviews their progress: Can physics unlock the mystery of awareness? Does awareness form the basis of fundamental physics? The result is an ambitious but ultimately disappointing tour, filled with breathless encounters with well-known scientists. Representative of the cast is MIT’s Max Tegmark, who tells Musser, “If you look at the problems that were still difficult to answer in fundamental physics, pretty much all of them trace back to awareness.” The book shows how physicists are contributing to understanding the mind, continuing a long history of physicists exploring other fields. Musser integrates physics with neuroscience, economies, mathematics and more. Yet a key source of local knowledge is obviously absent — psychology. An early example: Musser rightly applauds physicists’ contributions to artificial neural networks but is overly trustful of their implications, declaring that “ChatGPT” is already starting to demonstrate a generalized intelligence like that of humans. Interviewing more psychologists could have exposed the considerable gap remaining. What about awareness itself? Why are some states associated with felt experience, such as the pain of a headache and the sight of a sunset, but others not? Musser’s focus is integrated information theory(IIT). IIT begins with five self-evident principles of awareness: awareness exists, and it is structured, specific, unified, and definite. It then concludes assumptions concerning the causal structure of awareness systems, identifying awareness with integrated information. Finally, IIT offers a mathematical measure of this quantity: an equation for awareness. However, despite its enthusiasts, IIT has significant problems. Its working basis is unfounded, and serious doubts surround its testability and definability. Of course, a theory of awareness must detail when, why, and to what degree we are aware of ourselves. Here, psychology’s absence is most obvious. Over the last century, psychological research has revealed countless phenomena of awareness, from models that alter awareness to methods that stimulate unawareness; from extraordinary disorders of awareness to careful studies of metacognition (元认知). Although questionable, such phenomena are the data that any scientific theory of awareness must account for. Yet these phenomena and ideas are almost nowhere in the book. Of course, psychology has not solved awareness, but one cannot hope to explain awareness without confronting these data. Awareness is genuinely mysterious. So is fundamental physics. But hoping that physics can solve awareness while excluding other approaches is only a recipe for more mystery, not less. 28. What can we learn from Musser’s book? A. IIT is acknowledged for its testability and definability. B. Musser underestimates the impact of artificial intelligence. C. Using psychology is a trend for physicists exploring other fields. D. Musser highlights physicists’ efforts to uncover the secret of awareness. 29. What can be inferred from the passage? A. The link between physics and awareness has been established. B. There is no doubt about the principles of an equation for awareness. C. A multidisciplinary approach is essential to understanding awareness. D Study of awareness needs to ignore related psychological phenomena. 30. Regarding Musser’s view in his book, the author is . A. neutral B. disapproving C. supportive D. uncertain 2025海淀二模 C The idea that aging reduces adults’ ability to imagine, a common theme in children’s literature, is contradicted by psychological research. While children are often portrayed as more imaginative, research indicates that adults not only keep this ability but sometimes surpass children in imaginative thinking. Children are frequently celebrated for boundless imagination. Yet, research reveals that their make-believe games often center around realistic scenarios, such as cooking and cleaning, as demonstrated in a 2020 study published in Journal of Cognition and Development. Another study, lasting for four decades, also suggests that children are not naturally more imaginative than adults; their limitations result from a lack of knowledge and expertise to effectively use their imaginative capacity as adults. Imagination may have evolved for considering alternatives to reality, but we use it most naturally to explore close alternatives, like preparing a different meal, rather than far alternatives, like riding on clouds. When we use imagination to envision far alternatives — to innovate or invent — we’re not digging into an inborn appreciation of the extraordinary; we’re using a tool designed to explore the ordinary. When considering alternatives to reality, we fix our attention on possibilities that are physically reasonable, statistically probable, socially conventional and morally permissible. When told about possibilities that violate such regularities, we usually deny they could happen. Generally speaking, our ideas about what could happen are firmly rooted in what we expect to happen. This mindset is also particularly apparent in young children. In a 2018 study I co-designed with psychologist Jonathan Phillips, 4-year-olds were asked to help a distressed girl who disliked going to school due to missing her mother. Among all the solutions given, they perceived the only possible solution was for her mother to do something special after school to ease her concerns. Unexpected alternatives, such as snapping fingers and making it Saturday, wearing pajamas to school or lying about school being closed, were all regarded impossible. From this, we can conclude that children’s earliest intuitions (直觉力) about possibility confuse what could happen with what should happen. Historically, the improbable event of traveling faster than a horse was considered impossible, as was traveling by air or traveling into space. Before the arrival of trains and planes, there were good reasons to think that people could travel only so far and only so fast. But these reasons were empirical (经验主义的), not logical. Imagination, on its own, lumps the improbable with the impossible, but we can combine imagination with other abilities — namely, knowledge and reflection — to separate the two. While imagination in children often subjects to expectation, adults can control their imaginative capacity for innovation by integrating it with accumulated knowledge and reflective thinking. 28. According to the first two paragraphs, we know that _________. A. children develop imagination through games B. children face limitations in acquiring knowledge C. adults are as good as children in imaginative thinking D. adults’ imaginative ability is likely to stay constant with age 29 We can infer from the passage that _________. A. expectation results from imaginative capacity B. certain practical concerns can limit imagination C. breaking regularities may lead to close alternatives D. far alternatives are more important than close alternatives 30. The 2018 study shows that children _________. A. came up with a wide range of alternatives B. were quicker to figure out solutions C. took what should happen as possibilities D. used imagination in a reasonable way 31. The underlined word “lumps” in the last paragraph probably means _________. A. mix B. match C. compare D. replace 2025丰台二模 C In economic theories, people are typically represented as analytical agents who learn from past experiences to optimise (优化) their performance, eventually reaching a stable state in which they know how to maximise their earnings. This assumption surprised Garnier Brun at École Polytechnique in France because, as a physicist, he knew that interactions in nature—such as those between atoms—often result in chaos rather than stability. To test whether economists are correct to assume that learning from the past can help people avoid chaos, Brun and his colleagues devised a mathematical model for a game featuring hundreds of simulated players. Each player can choose between two actions, like buying or selling a stock. They also interact with each other, and the players’ decision-making is influenced by what they have done before—meaning they are able to learn from experience. The researchers could adjust the precise extent to which a player’s past experiences influenced their subsequent decision-making. They could also control the interactions between the players to make them either cooperate or compete with each other more. With all these control knobs available to them, Brun and his colleagues used methods from statistical physics to simulate different game situations on a computer. In some situations, the researchers expected that the game would always result in chaos, with players unable to learn how to optimise their performance. Economic theory would also suggest that, given the right set of parameters (参数), the players would settle into a stable state where they had mastered the game—but the researchers found this wasn’t really the case. The most likely outcome was a state that never settled. Team member Jean Martin, also at École Polytechnique, says that, in the absence of one centralised and all-knowing player who could coordinate everyone, regular players could only learn how to reach “satisficing” states. That is a level that satisfied minimum expectations, but not much more. Players gained more than they would have done by playing at random, so learning wasn’t useless, but they still gained less than they would have if past experience had allowed them to truly optimise their performance. Martin says the game model is too simple to be immediately adopted for making real-world predictions, but she sees the study as a challenge to economists to drop many assumptions that currently go into theorising processes, like merchants choosing suppliers or banks setting interest rates. The finding could also be important for simulating processes like foraging decisions by animals or for some machine-learning applications, says Toby Galla at the Institute for Cross-Disciplinary Physics and Complex Systems in Spain. 27. According to the passage, what inspired Brun to carry out the study? A. The belief that chaos is a common outcome in natural systems. B. The conflicts between economic theories and interactions in nature. C. The similarities in interaction patterns between atoms and humans. D. The inability of economic theories to predict individual behaviours. 28. What is Paragraph 2 mainly about? A. The research purpose. B. The research subjects. C. The research methods. D. The research direction. 29. What can we infer from Martin’s words? A. The “satisficing” state is an ideal outcome for players in the game. B. Players can reach optimal performance with enough past experiences. C. Centralised control is essential for players to reach their full potential. D. Learning from past experiences isn’t sufficient to maximise one’s gains. 30. What can be the best title of the passage? A. Economic Theories: Are They Reliable in Predicting Stability? B. Learning from Mistakes: The Key to Optimal Decision-Making. C. A Mathematical Model: Breaking Economic Stability Assumptions. D. Game Model Findings: Spreading Influence Across Multiple Disciplines. 学科网（北京）股份有限公司 $