High-Frequency Topics and Common Mistakes in Year 12 CAIE Psychology | CAIE 心理学高频考点与易错题分析

📚 High-Frequency Topics and Common Mistakes in Year 12 CAIE Psychology | CAIE 心理学高频考点与易错题分析

Mastering the CAIE AS Psychology (9990) syllabus demands not only reciting studies but also understanding methodological nuances and common pitfalls. This guide highlights the most frequently examined topics and the mistakes that repeatedly trip up Year 12 candidates. Use it to focus your revision, sharpen your evaluation skills, and approach the exam with confidence.

掌握 CAIE AS 心理学 (9990) 课程不仅需要熟记各项研究，还要理解方法学上的细微差别和常见的易错点。这篇文章梳理了最高频的考点以及 Year 12 考生反复出现的问题。用它来聚焦复习、提升评价能力，自信地走进考场。

1. Experimental Designs: Independent vs. Repeated Measures | 实验设计：独立组与重复测量设计

A perennial exam favourite is identifying and evaluating experimental designs. Independent measures (IM) use different participants in each condition, while repeated measures (RM) use the same participants for all conditions. Examiners frequently test the ability to match a design to a specific study and to explain its strengths and weaknesses. A common mistake is confusing the design with the method (e.g., lab experiment) or failing to link a named control to the correct design weakness. For IM, individual differences (participant variables) are the primary issue, controlled through random allocation; for RM, order effects (practice, fatigue, boredom) become critical, often controlled by counterbalancing.

实验设计是常考的重点，要求考生识别并评价独立组设计（IM）和重复测量设计（RM）。独立组在每个条件下使用不同的被试，重复测量则让同一批被试参与所有条件。考官经常考查将一个具体研究与设计匹配的能力，以及解释其优缺点。常见的错误是把实验设计与研究方法（如实验室实验）混淆，或未能将特定的控制措施与对应设计的弱点联系起来。独立组的主要问题是个人差异（被试变量），通过随机分配来控制；重复测量的关键问题是顺序效应（练习效应、疲劳效应、厌倦效应），通常用平衡法来控制。

Another error is giving vague control suggestions. Saying “random allocation fixes individual differences” without stating that it randomly assigns participants to conditions to equalise participant characteristics will only attract partial marks. Similarly, for RM, simply stating “counterbalancing” is insufficient; candidates must describe how half of the participants experience condition A then B, and the other half experience B then A, to distribute order effects evenly.

另一个错误是给出的控制措施过于模糊。仅仅说“随机分配解决了个人差异”，却不说明它是通过将被试随机分配到不同条件来平衡个人特征，只会得到部分分数。同样，对于重复测量，仅仅提到“平衡法”是不够的；考生必须描述如何让一半被试先经历条件 A 再经历条件 B，另一半被试先 B 后 A，从而平均分配顺序效应。

2. Ethical Guidelines in AS Studies | AS 研究中的伦理准则

Every AS core study can be examined through the lens of ethics. High-frequency topics include informed consent, deception, protection from harm, privacy, and the right to withdraw. Candidates frequently lose marks by simply listing ethical issues without explaining how they arose in the specific study. For example, in Milgram (1963), saying “there was deception” is not enough; you must specify that participants were deceived about the true purpose of the study (they thought it was about learning and memory) and that the shocks were fake. The crucial link is between the ethical guideline and the concrete detail of the procedure.

每项 AS 核心研究都可以从伦理的角度来考查。高频考点包括知情同意、欺骗、防止伤害、隐私和退出研究的权利。考生常因仅仅罗列伦理问题却未解释其在该研究中的具体体现而失分。例如，在 Milgram (1963) 实验中，只说“存在欺骗”是不够的；必须具体说明被试在研究的真实目的上受到欺骗（他们以为实验是关于学习和记忆的），并且电击是假的。关键是要在伦理准则和具体的程序细节之间建立起联系。

A high-level answer will also suggest credible ways the study could have been modified to overcome the ethical breach, such as presumptive consent, debriefing, or using a pre-screening procedure. A common trap is recommending that the whole study should not be conducted because of ethics – the question expects you to balance ethical considerations with the value of the research, or to propose realistic alternatives.

高水平的答案还会提出可行的修改建议来弥补伦理漏洞，比如假定同意、事后解释或使用预先筛查程序。一个常见的陷阱是建议由于伦理问题该研究根本不应进行——题目期望你平衡伦理考量与研究价值，或者提出切实可行的替代方案。

3. Canli et al. (2000): Brain Scans and Emotional Memory | Canli 研究：脑成像与情绪记忆

Canli’s study is a classic biological approach investigation linking amygdala activation to emotional memory recall. Students must know the aim: to investigate whether an emotionally intense scene produces higher levels of brain activation in the amygdala and whether that activation predicts later memory. The key finding was that higher amygdala activation when viewing emotionally intense scenes correlated with better recall three weeks later. A common mistake is misinterpreting the correlational nature of the fMRI data: the study shows a relationship, not causation. Writing “amygdala activation causes better memory” is inaccurate and will cost marks; always use “linked to” or “associated with”.

Canli 的研究是一项经典的生物取向研究，将杏仁核的激活与情绪记忆联系起来。考生必须掌握其目的：探究情绪强烈的场景是否会使杏仁核产生更高的脑激活水平，以及这种激活是否能预测随后的记忆。关键发现是，观看情绪强烈场景时杏仁核激活程度越高，三周后的回忆成绩越好。常见错误是误解 fMRI 数据的相关性质：这项研究显示的是相关性，而非因果关系。写成“杏仁核激活导致记忆力更好”是不准确的，会扣分；应始终使用“与……有关”或“与……相关”。

Another frequent pitfall is failing to identify the independent variable (emotional intensity of scenes, operationalised by participant ratings of emotional arousal) and the dependent variable (pixel count of amygdala activation and later memory recall). Candidates sometimes mistakenly state the IV was the scene itself, rather than the emotional intensity rating, which is a product of the self-report measure. Being precise with operationalisation distinguishes a top-band answer.

另一个常见陷阱是未能准确识别自变量（场景的情绪强度，通过被试对情绪唤醒度的评分来操作化）和因变量（杏仁核激活的像素计数及随后的记忆回忆）。考生有时错误地认为自变量是场景本身，而非情绪强度评分，后者是自我报告测量得出的结果。对操作化定义的准确把握是区分高分答案的标志。

4. Doodling and Divided Attention (Andrade, 2010) | 涂鸦与分散注意 (Andrade 2010)

Andrade’s study on doodling is frequently assessed because it cleverly tests the cognitive explanation of attention and daydreaming. The hypothesis predicted that doodling would aid concentration on a boring, primary task (monitoring a telephone message) by preventing daydreaming. The experimental group shaded printed shapes while listening; the control group just listened. The doodlers recalled 29% more names and places. A high-frequency mistake is misidentifying the experimental design as independent measures, when in fact it was independent measures (each participant took part in only one condition). More significantly, students often fail to link the result to the theory of working memory: doodling is a low-load task that occupies just enough cognitive resources to stop the mind from wandering, without overloading the central executive.

Andrade 关于涂鸦的研究经常被考查，因为它巧妙地检验了注意力和白日梦的认知解释。该假设预测，涂鸦能通过阻止白日梦来帮助集中注意力在一项枯燥的主要任务（监听电话留言）上。实验组一边听一边给印刷形状涂色，对照组只是听。涂鸦组对名字和地点的回忆量高出 29%。高频错误是把实验设计误认为重复测量，而它实际上是独立组设计（每个被试只参加一种条件）。更关键的是，学生往往不能将结果与工作记忆理论联系起来：涂鸦是一种低负荷任务，正好占用足够的认知资源以防止思维游荡，而不会加重中央执行系统的负担。

When evaluating the study, candidates commonly criticize the sample size (only 40 participants) but then fail to note the link between a small sample and generalisability, or how the use of a standardised, monotonous task improved internal validity. A strong evaluation will also discuss the artificiality of the lab setting and the confounding variable of individual differences in working memory capacity, which could affect the ability to doodle and recall.

评价该研究时，考生常常批评样本量小（只有40名被试），但随后未能指出小样本与推广性的关联，也没说明标准化的单调任务如何提高了内部效度。强有力的评价还会讨论实验室环境的人为性，以及工作记忆容量的个体差异这一混淆变量，它可能影响涂鸦和回忆的能力。

5. Obedience: Milgram’s Baseline Experiment | 服从权威：米尔格拉姆基线实验

Milgram’s (1963) original study is perhaps the most examined study in CAIE AS psychology. Candidates must describe the procedure: 40 male participants acting as “teachers” administered assumed electric shocks to a “learner” under the instruction of an experimenter in a grey lab coat. The voltage increased from 15V to 450V, and the critical measure was the number of participants who administered the maximum shock (65%). A common mistake is failing to clearly identify the three key features that Milgram suggested promote obedience: location (prestigious Yale setting), proximity (learner in another room, only banging at 300V), and the uniform/authority figure. Simply stating that “the experimenter looked authoritative” is too vague.

Milgram (1963) 的原始研究可能是 CAIE AS 心理学中被考察最多的研究。考生必须描述程序：40 名男性被试充当“老师”，在身穿灰色实验袍的实验者指示下对“学习者”施加（他们以为的）电击。电压从 15V 逐步升到 450V，关键测量指标是施加最大电击的被试比例（65%）。常见错误是未能清晰识别 Milgram 提出的促进服从的三大关键因素：地点（耶鲁大学的声望环境）、接近性（学习者在另一个房间，仅在 300V 时敲墙）以及制服/权威人物。仅说“实验者显得权威”太过模糊。

When evaluating ethics, many candidates repeat that participants were distressed, but a better response specifies the ethical violations in detail: lack of informed consent regarding the true nature of the study, deception about the shocks and the learner, and the difficulty some faced in withdrawing due to the prods (“You have no choice; you must go on”). Also, a high-level evaluation will contrast the ethical cost with the ground-breaking findings, concluding that the insight into destructive obedience outweighed the cost.

在评价伦理问题时，许多考生反复说被试很痛苦，但更好的回答会具体说明伦理违规的细节：对实验真实性质缺乏知情同意、对电击和学习者的欺骗，以及由于连续催促（“你别无选择，必须继续”）导致部分被试难以退出。此外，高水平的评价还会权衡伦理成本与突破性的发现，最终得出对破坏性服从的见解超过了成本的结论。

6. Social Learning Theory and Bandura’s Bobo Doll Study | 社会学习理论与班杜拉波波玩偶实验

Bandura et al. (1961) is central to the learning approach. The study aimed to show that children can learn aggressive behaviour through observation, without direct reinforcement. This links directly to the key assumptions of social learning theory: attention, retention, motor reproduction, and motivation. The common mistake is to simply describe the procedure (72 children, aggressive / non-aggressive / control conditions) without explicitly connecting the results to the theoretical concepts. For instance, the finding that boys imitated more physical aggression than girls can be linked to gender-role stereotyping and motivation (vicarious reinforcement), which is examined in 10-mark evaluation questions.

Bandura 等人 (1961) 的研究是学习取向的核心。该研究旨在证明儿童可以通过观察学习攻击行为，无需直接强化。这直接联系到社会学习理论的关键假设：注意、保持、动作再现和动机。常见错误是仅仅描述程序（72 名儿童，攻击性 / 非攻击性 / 控制条件），而没有把结果与理论概念明确联系起来。例如，男孩比女孩模仿了更多身体攻击行为这一发现，可以与性别角色刻板印象和动机（替代强化）挂钩，这正是 10 分评价题要考查的内容。

An area where students often stumble is the difference between imitation and identification. A child imitating the same-sex model’s behaviour demonstrates identification, not just simple imitation. Also, many dismiss the study as lacking mundane realism because hitting a doll is not real aggression; however, a sophisticated evaluation acknowledges this limitation but counters that the controlled environment allowed for clear cause-and-effect conclusions about observational learning, which is a strength.

学生容易困惑的一个地方是模仿与认同的区别。儿童模仿同性别榜样的行为展示的是认同，而不仅仅是简单的模仿。此外，许多人因为攻击对象是玩偶就批评该研究缺乏日常真实性，但更精巧的评价会承认这一局限，同时反驳说，受控环境能让我们对观察学习得出清晰的因果关系结论，这恰恰是一个优点。

7. Theory of Mind: Baron-Cohen et al. (2001) Eyes Test | 心理理论：Baron-Cohen 眼睛测试

The revised Eyes Test (2001) is a core study in the cognitive approach, probing theory of mind (ToM) deficits in adults with high-functioning autism or Asperger syndrome (AS). The study compared three groups: adults with AS/HFA, normal adults, and adults with Tourette syndrome. The AS/HFA group scored significantly lower on the Eyes Test, supporting the notion of a specific neurocognitive deficit in ToM. A classic error is misreporting the aim: the aim was not to diagnose autism but to see if the eyes test would be a valid measure of ToM deficits specific to autism, as distinct from other clinical conditions.

修订版眼睛测试 (2001) 是认知取向的核心研究，探究患有高功能自闭症或阿斯伯格综合征（AS）的成人在心理理论（ToM）上的缺陷。该研究比较了三组被试：AS/HFA 成人、正常成人以及患有抽动秽语综合征的成人。AS/HFA 组在眼睛测试上得分显著较低，支持了 ToM 存在特定神经认知缺陷的观点。经典错误是误报研究目的：目的不是诊断自闭症，而是为了检验眼睛测试能否成为有效测量自闭症特有的 ToM 缺陷的工具，以区别于其他临床病症。

When evaluating validity, many candidates content themselves with saying that the eyes test has low ecological validity because it uses static images. While true, a top answer will also discuss the issue of independent variable manipulation: the test comprises complex mental state terms (e.g., “contemptuous”, “aghast”), and participants’ vocabulary levels could act as an extraneous variable. This was controlled to some extent by using a glossary, but it still presents a construct validity concern – is it measuring ToM or verbal comprehension?

在评价效度时，许多考生满足于说眼睛测试因为使用静态图片所以生态效度低。这固然没错，但高分答案还会讨论自变量操纵的问题：测试包含了复杂的心理状态词汇（如“轻蔑的”、“惊骇的”），被试的词汇水平可能成为额外变量。尽管使用了词汇表在一定程度上加以控制，这仍然构成了构念效度方面的担忧——它到底测量的是心理理论还是言语理解？

8. Validity, Reliability and Generalisability | 效度、信度与推广性

These three methodological concepts are tested across all core studies. A mistake pattern seen in marking is the confusion between reliability and validity. Reliability concerns consistency: if the study were replicated, would the results be similar? Key indicators are standardised procedures, control over extraneous variables, and inter-rater reliability for observations. Validity asks whether the study truly measures what it intends to measure. Internal validity can be threatened by demand characteristics, lack of standardisation, or confounders; external validity includes ecological and population validity. Candidates often write “the study had high validity because it was a lab experiment” – this is conceptually flawed: lab experiments often increase internal validity but may reduce ecological validity.

这三个方法学概念在所有核心研究的考查中都会出现。阅卷中常见的一个错误模式是混淆信度和效度。信度关乎一致性：如果重复该研究，能否得到相似的结果？关键指标有标准化程序、对外部变量的控制以及观察研究中的评分者间信度。效度则问的是研究是否真正测量了它想要测量的东西。内部效度可能受到需求特征、缺乏标准化或混淆变量的威胁；外部效度包括生态效度和族群效度。考生常常写道“该研究因为是实验室实验所以效度高”——这在概念上是错误的：实验室实验往往提高内部效度，却可能降低生态效度。

A smarter approach is to always use the formula: Identify the specific aspect of the procedure (e.g., the mock prison setting in Zimbardo’s study) and then state whether it is a threat or a strength for a specific type of validity. Defend your point with a “because” clause. For example, “The standardised verbal prods in Milgram increased reliability because they ensured each participant was prompted in the same manner, reducing experimenter bias.”

更聪明的做法是始终套用这个公式：识别出程序的某个具体方面（比如 Zimbardo 研究中的模拟监狱环境），然后说明它对某一特定效度类型是威胁还是优势。用“因为”来支撑你的观点。例如，“Milgram 研究中标准化的口头催促提高了信度，因为它确保了每个被试都受到相同方式的催促，减少了实验者偏差。”

9. Quantitative Data Analysis: Mean, Median, Range, SD | 量化数据分析：平均数、中位数、极差与标准差

Calculations and interpretations of descriptive statistics appear regularly in Papers 1 and 2. You must be able to compute the mean, median, mode, range, and standard deviation from a simple data set. Common pitfalls include forgetting to order data before finding the median, or confusing the range (largest minus smallest value) with the interquartile range. More importantly, candidates lose marks in applied questions because they cannot explain why a particular measure was chosen. For example, the median is recommended when the data set contains outliers or is skewed, as it is not affected by extreme scores, whereas the mean is pulled away from the central cluster.

描述统计的计算和解释经常在试卷一和试卷二中出现。你必须能够根据简单的数据集计算出平均数、中位数、众数、极差和标准差。常见的陷阱包括找中位数时忘了先排序，或者将极差（最大值减最小值）与四分位距混淆。更重要的是，考生在应用题中会因无法解释为何选择某一测量指标而失分。例如，当数据集含有极端值或呈偏态分布时，推荐使用中位数，因为它不受极端分数的影响，而平均数则会向极端值方向偏移。

A Grade A candidate can succinctly compare the standard deviation with the range. The range provides a crude measure of spread using only two values; the standard deviation uses all scores and tells you the average distance of each score from the mean. In many AS studies, such as Canli, the means and SDs are presented in tables, and you are expected to interpret them: a smaller SD indicates less variance and more consistent amygdala activation ratings within that condition.

能拿 A 的考生会简洁地比较标准差和极差。极差仅使用两个数值，是一个粗略的离散量指标；标准差则使用了所有分数，告诉你每个分数与平均数的平均距离。在 Canli 等许多 AS 研究中，平均数和标准差以表格形式呈现，你要能够解读：标准差越小，说明该条件下杏仁核激活评分的差异越小，评分越一致。

10. Top Mistakes in 10‑mark Evaluation Questions | 10分评价题的顶级错误

The 10-mark evaluation question (e.g., “Discuss the strengths and weaknesses of …”) is heavily weighted and often the differentiator between B and A grades. The single biggest mistake is writing a purely descriptive answer with little or no evaluation. Listing details of the procedure for six lines and then adding “this was a strength” at the end is not evaluation. Likewise, writing a block of strengths followed by a block of weaknesses in a disembodied manner; each point must be anchored in the study’s aim, procedure, or findings.

10 分评价题（例如“讨论……的优点和缺点”）分值很高，常常是 B 等级和 A 等级的分水岭。最大的错误是写出一篇纯描述性、几乎没有评价的答案。用了六行描述程序然后末尾补一句“这是一个优点”根本算不上评价。同样地，跳跃式地写一堆优点再写一堆缺点也是不行的；每一个观点都必须紧扣研究的目的、程序或发现。

Other top mistakes include: not using methodological terms (e.g., saying “the experiment was in a lab so it was good” instead of “high internal validity due to standardised environment”); making generic evaluation points that could apply to any study (e.g., “the sample was small” without linking to why it matters for the specific study); and failing to offer a balanced conclusion. An exemplary evaluative answer will always weigh the evidence, discuss the implications for theory or real-world application, and include a short but reasoned judgement at the end.

其他顶级错误还包括：不使用方法学术语（例如说“实验在实验室里进行所以很好”，而不是“由于标准化环境而具有高内部效度”）；做出可以套用到任何研究上的泛泛评价（例如只说“样本小”而不联系该具体研究为何重要）；以及未能给出一个平衡的结论。一份模范的评价答案始终会权衡证据，讨论对理论或现实应用的影响，并在结尾给出简短而有推理的判断。

Published by TutorHao | Psychology Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）