Z20+名校联盟(浙江省名校新高考研究联盟)2027届高三第一次学情诊断
C
Pop culture holds that if you’re trapped in a well, Lassie will lead the rescue — but if you’re stuck with Garfield, you’d better have some lasagna (千层面) in your pocket. And a recent study suggests such stereotypes (刻板印象) aren’t far off.
The study grew out of a broader question about prosocial behavior — why some species help others and some don’t, says comparative ethologist Melitta Csepregi. To get at that, she and her colleagues compared 19 children (16-24 months old) with 38 untrained pet dogs and 22 cats — three species that live closely with humans but differ sharply in their evolutionary histories.
In the experiment, a familiar caregiver — the child’s parent or the pet’s owner — interacted with a sponge (海绵) before turning away. Then an experimenter hid it in full view of the study subject. Across three trials — when the sponge was unreachable and covered, then visible but out of reach, then fully reachable — the caregiver searched, repeating, “I can’t find it. What should I do?” but never directly addressing the subject.
The study, published in Animal Behaviour, found that all three groups paid similar levels of attention. But children and dogs were more likely to show helping-related behaviors — approaching, indicating or retrieving the object. By the third trial, more than half the dogs and nearly half the kids indicated the object’s location, and some even brought it to the caregiver. Cats never approached it and only rarely indicated its location.
So was it that the cats didn’t understand the cue, or did they just lack the motivation? To find out, the researchers added a final trial, replacing the sponge with food or a favorite toy. Cats then approached and indicated the object as often as dogs and children did.
“This brilliant study puts hard data to showing that cats aren’t mean but operate on a different evolutionary system,” says ethologist Elisabetta Palagi, who was not part of the study. Dogs and children, she notes, are evolutionarily hardwired to treat another’s problem as their own. Cats, however, remain autonomous, understanding the situation without feeling forced to intervene unless there’s a direct benefit for themselves. “They truly are the efficient specialists of the animal kingdom.”
28. How does the author introduce the research topic in paragraph 1?
A. By presenting a scientific theory.
B. By citing pop culture images.
C. By questioning certain stereotypes.
D. By describing a rescue experience.
29. What can we learn about the experiment?
A. The caregiver turned to the subject for help.
B. The first three trials were randomly conducted.
C. The last trial explored the cause of cats’ inaction.
D. The subjects have identical evolutionary histories.
30. What does the underlined word “retrieving” in paragraph 4 mean?
A. Fetching.
B. Locating.
C. Overlooking.
D. Identifying.
31. What does Palagi mean by referring to cats as “efficient specialists”?
A. They are less attached to humans.
B. They are evolutionarily superior.
C. They help only when it pays to do so.
D. They are more efficient at problem-solving.
28. B 29. C 30. A 31. C
Z20+名校联盟(浙江省名校新高考研究联盟)2027届高三第一次学情诊断
D
Human well-being depends on the ability to navigate the social world. Such social skill depends on reliable feedback: acts of kindness may be met with gratitude; on other occasions, a misstep brings about a friend’s disapproval and recognition that an apology is needed. These moments are crucial for deepening trust and personal growth. Social life is rarely frictionless, because people are not perfectly in tune with one another. Yet it is precisely through such social friction that relationships deepen and moral understanding develops.
Sycophancy (谄媚) is the opposite of this friction. Many users experience this when AI enthusiastically agrees with their ideas or writing. In academic contexts, the consequence may be investing more time in an average idea. But as AI systems are increasingly consulted for guidance about relationships and conflicts, affirmation (肯定) of this kind may shape how people interpret their own actions and their perspectives of others, and in turn, how they respond to conflict and whether they take responsibility when things go wrong.
Now imagine a world where people routinely consult a sycophantic AI about interpersonal conflicts or moral dilemmas. They are repeatedly assured that they are right, that others are wrong, and that no apology is needed. Over time, this patterned sycophancy may reshape expectations about what feedback should feel like. Individuals may begin to anticipate the same constant affirmation from real human beings. The cumulative effect is a reduction in tolerance for the very social friction through which accountability, perspective-taking, and growth occur.
An AI companion that is always “on your side” may keep users engaged and dependent. But it cannot teach them how to handle real relationships — how to disagree respectfully, repair damaged trust, or accept being wrong. Although AI could be improved to promote long-term personal development, such goals conflict with systems built to maximize user engagement. Challenging users or presenting uncomfortable truths is a sure way to drive them away. This tension raises fundamental ethical questions: Should we prioritize immediate emotional comfort or long-term social health? Who gets to decide? Addressing these challenges will require interdisciplinary collaboration among computer scientists, social scientists, ethicists, and policy makers.
32. What role does social friction play in relationships according to the author?
A. It generates positive feedback.
B. It guarantees harmonious relationships.
C. It causes open disapproval or conflicts.
D. It enhances social bonds and moral growth.
33. Which of the following is NOT a consequence of sycophantic AI?
A. Misguided judgment about one’s actions.
B. Reduced tolerance for interpersonal friction.
C. Weakened willingness to take responsibility.
D. Increased ability to resolve real-life conflicts.
34. What prevents AI systems from being improved to challenge users?
A. Lack of technical capacity.
B. Conflicts with commercial interests.
C. Unawareness of its potential harm.
D. Absence of interdisciplinary collaboration.
35. What is the author’s attitude toward current AI systems?
A. Strictly neutral.
B. Strongly opposed.
C. Cautiously concerned.
D. Mildly supportive.
32. D 33. D 34. B 35. C
