📚 High-Frequency Exam Topics and Common Mistakes Analysis for AS Cambridge Psychology | AS剑桥心理学：高频考点与易错题分析
AS Cambridge Psychology (9990) requires a strong grasp of research methods, four core approaches, and the ability to evaluate studies using evidence and debates. This article highlights the topics that appear most frequently in exams and the typical errors students make, helping you sharpen your revision and boost your marks.
剑桥AS心理学（9990）要求牢固掌握研究方法、四个核心取向，并善于运用证据和争议来评估研究。本文梳理了考试中最高频出现的主题，以及学生常犯的典型错误，帮助同学们高效复习、稳拿高分。
1. Structure of the AS Exam and Command Words | AS考试结构与指令词
The AS Psychology exam has two papers: Paper 1 (Approaches, Issues and Debates) and Paper 2 (Research Methods). High-frequency topics include designing an experiment, evaluating a study’s methodology, and applying the four approaches (biological, cognitive, social, and learning). Many students lose marks because they confuse ‘describe’ with ‘evaluate’ or fail to recognise that questions often combine both. For instance, a question may ask you to ‘Describe the study by Milgram and evaluate it in terms of ethics.’ If you only describe and forget to evaluate, you severely limit your available marks.
AS心理学考试包含两份试卷：试卷一（取向、议题与争论）和试卷二（研究方法）。高频考点涉及设计实验、评估研究的方法论，以及应用四个取向（生物、认知、社会、学习）。很多学生失分是因为混淆了“描述”和“评价”，或没有意识到题目常常同时要求这两项。例如，一道题可能要求“描述米尔格拉姆的研究并从伦理角度评价它”。如果只描述而忘记评价，得分将大幅受限。
2. Research Methods: Experiments, Observations, and Self-reports | 研究方法：实验、观察与自我报告
You must know the key features of each method and when to use them. A laboratory experiment has high control of extraneous variables, allowing you to infer cause and effect, but it may lack ecological validity. A naturalistic observation captures spontaneous behaviour in a real setting, yet observer bias and lack of control over extraneous variables can weaken internal validity. A common mistake is claiming that a correlational study proves causation. Correlation shows a relationship, but a third variable may be responsible, so you must never write ’causes’ when describing a correlational finding.
你必须掌握每种方法的关键特征及适用时机。实验室实验对外部变量有高度控制，可以推断因果关系，但可能缺乏生态效度。自然观察能捕捉真实环境中的自发行为，但观察者偏见和对额外变量缺乏控制会削弱内部效度。一个常见错误是声称相关研究证明了因果关系。相关性只显示关系，可能存在第三变量，因此在描述相关发现时绝不能用“导致”。
Also, when asked to ‘suggest an alternative method’ for a specific scenario, students often propose a method without justifying why it improves the design. For example, replacing a lab experiment with a covert observation may reduce demand characteristics, but you need to explicitly state how demand characteristics are relevant and how the new method addresses them.
此外，当题目要求为某一情景“建议一种替代方法”时，学生常常提出方法但不说明它为何能改善设计。例如，用隐蔽观察替代实验室实验可以减少需求特征，但你需要明确解释需求特征在此处的关联，以及新方法如何解决这个问题。
3. Experimental Designs and Their Vulnerabilities | 实验设计及其缺陷
Independent groups, repeated measures, and matched pairs designs appear in almost every exam series. The most common pitfall is failing to link a design to a specific threat. For independent groups, participant variables (e.g., individual differences in IQ) can become confounding variables. For repeated measures, order effects like practice or fatigue can distort results, but these can be controlled by counterbalancing. Matched pairs attempts to control participant variables, but matching is never perfect and can be time-consuming.
独立组、重复测量和匹配对设计几乎出现在每场考试中。最常见的陷阱是未能将设计与具体威胁联系起来。独立组设计中，被试变量（如智商上的个体差异）可能成为混淆变量。重复测量设计中，顺序效应如练习或疲劳会扭曲结果，但可以通过平衡法控制。匹配对设计试图控制被试变量，但匹配从不完美且耗时。
When evaluating a study, students often write ‘it used repeated measures, therefore order effects were controlled’ — which is incorrect unless counterbalancing is mentioned. The examiner wants to see that you understand the design is vulnerable to order effects, not that it automatically eliminates them. Always specify what control was actually applied.
在评价一项研究时，学生常写“它使用了重复测量，因此顺序效应得到了控制”——除非提及平衡法，否则这句话是错误的。考官希望看到你明白该设计容易受到顺序效应影响，而不是自动消除了它们。务必指明实际采取了什么控制措施。
4. Ethical Guidelines: Applying BPS Principles | 伦理准则：运用BPS原则
The British Psychological Society (BPS) code includes informed consent, deception, protection from harm, right to withdraw, privacy, and debrief. Students often recite these in a list but fail to apply them concretely. For a question about how Milgram’s study might be improved ethically, a weak answer would just say ‘obtain informed consent’. A strong answer explains that genuine informed consent was difficult because telling participants the true purpose would create demand characteristics, so Milgram could have used presumptive consent or prior general consent.
英国心理学会（BPS）的伦理准则包括知情同意、欺骗、免受伤害、退出权、隐私和事后解释。学生常将这些条文罗列出来，却不能具体应用。对于“如何从伦理上改进米尔格拉姆的研究”这类问题，较弱的答案会简单说“获得知情同意”。而强势的答案会解释真正的知情同意很难，因为告知参与者真实目的会产生需求特征，因此米尔格拉姆本可以使用推定同意或事先一般同意。
Another high-frequency error is confusing deception with lack of informed consent. Deception involves actively misleading participants; informed consent is about allowing them to make a knowledgeable choice about participation. A study can have no deception but still lack full informed consent if details are withheld.
另一个高频错误是将欺骗与缺乏知情同意混为一谈。欺骗是指主动误导参与者；而知情同意是关于让他们在知情的情况下做出参与选择。一项研究可能没有欺骗却没有完全的知情同意，如果隐瞒了细节。
5. Biological Approach: Twin Studies and Brain Plasticity | 生物取向：双生子研究与大脑可塑性
Key assumptions of the biological approach — that behaviour has a genetic basis, brain structures and neurotransmitters influence thoughts and actions — are frequently examined. Twin studies (e.g., Gottesman & Shields on schizophrenia) show higher concordance rates in monozygotic (MZ) twins than in dizygotic (DZ) twins, supporting a genetic component. However, the common mistake is to conclude that a high MZ concordance proves a disorder is entirely genetic. MZ twins share most of their environment as well, so the difference from DZ twins may not be purely genetic, and the concordance rate is never 100%.
生物取向的基本假设——行为有遗传基础，大脑结构和神经递质影响思维与行动——是常考内容。双生子研究（如高特斯曼和希尔兹关于精神分裂症的研究）显示同卵双生子的同病率高于异卵双生子，支持遗传因素。但常见错误是，认为高同卵双生子同病率就证明疾病完全是遗传的。同卵双生子也共享大部分环境，因此它们与异卵双生子的差异可能不纯属遗传，且同病率从不会达到100%。
Brain plasticity studies, such as Maguire et al.’s research on London taxi drivers, show that the posterior hippocampus can change in volume due to spatial navigation experience. Students often misunderstand the direction of causality: the study does not prove that a larger hippocampus causes better navigation; it suggests that extensive navigation experience may alter brain structure. The correlation-causation confusion is very common here.
大脑可塑性研究，如马圭尔等人对伦敦出租车司机的研究，显示后海马体的体积可因空间导航经验而改变。学生常误解因果方向：该研究并不能证明更大的海马体导致更佳的导航能力；它表明大量的导航经验可能改变脑结构。相关与因果的混淆在这里非常普遍。
6. Cognitive Approach: Memory Models and Their Evaluations | 认知取向：记忆模型及其评价
The multi-store model (MSM) and the working memory model (WMM) are central. The MSM describes a linear flow from sensory register to short-term memory (STM) to long-term memory (LTM), with attention and rehearsal as key processes. A common error is to describe the model accurately but evaluate it weakly by simply stating ‘it is too simplistic’. Instead, you could use evidence from amnesic patients like HM or KF to argue that the model does not explain different types of LTM or that STM is not a single unitary store, thus supporting the WMM’s multi-component view.
多储存模型（MSM）和工作记忆模型（WMM）是核心。MSM描述了从感觉登记到短时记忆再到长时记忆的线性流程，注意与复述是关键过程。常见错误是描述模型尚可，但评价浅薄，仅仅说“它太简单了”。其实，你可以用健忘症患者如HM或KF的证据论证该模型不能解释不同类型的LTM，或STM并非单一存储，从而支持WMM的多成分观点。
For the WMM, students often forget to mention the episodic buffer when evaluating, or they mislabel the slave systems. The central executive is often described as the ‘most important’ component, but its exact function is vague, which is a valid evaluation point. Also, students frequently mix up which system handles visual vs. verbal information: the visuospatial sketchpad handles visual-spatial, and the phonological loop handles auditory-verbal.
对于WMM，学生在评价时经常忘记提及情景缓冲区，或者错误标注附属系统。中央执行器常被描述为“最重要”的成分，但其确切功能模糊不清，这是一个有效的评价点。另外，学生常混淆哪个系统处理视觉与言语信息：视觉空间画板处理视觉空间信息，语音回路处理听觉言语信息。
7. Social Approach: Conformity and Obedience | 社会取向：从众与服从
Asch’s conformity experiments and Milgram’s obedience study are examined from every angle: methodology, ethics, cultural variations, and situational vs. dispositional explanations. A very frequent mistake is to describe Asch’s line judgement task and then evaluate it only in terms of ethics (deception). That is insufficient because Asch’s study had many other strengths, such as high experimental control and the ability to manipulate group size and unanimity. You must also discuss the ecological validity of the conformity situation: judging lines is a low-stakes task, which may not reflect real-world conformity.
阿希的从众实验和米尔格拉姆的服从研究从各个角度被考查：方法论、伦理、文化差异以及情境性与倾向性解释。一个非常常见的错误是描述阿希的线条判断任务后只从伦理角度（欺骗）评价它。这不够充分，因为阿希的研究还有其他许多优点，如高实验控制以及操纵群体规模和一致性的能力。你还必须讨论从众情境的生态效度：判断线条是一项低风险的任务，可能无法反映现实世界中的从众。
When addressing Milgram’s obedience, many students claim his study was unethical because of the extreme distress observed, but they stop there. A high-scoring answer will weigh the ethical costs against the significant insights gained about destructive obedience. Also, avoid stating that ‘65% of participants obeyed to 450 volts’ without explaining that this percentage refers to the maximum shock level and that the experimenter used prods like ‘the experiment requires that you continue’.
在讨论米尔格拉姆的服从研究时，许多学生声称该研究因观察到极度痛苦而不符合伦理，却停在此处。高分答案会权衡伦理代价与关于破坏性服从的重要洞见。还要避免说“65%的参与者服从到450伏特”却不解释这个百分比指的是最大电击水平，以及实验者使用了诸如“实验要求你继续”的催促语。
8. Learning Approach: Conditioning and Applications | 学习取向：条件作用及应用
The principles of classical conditioning (Pavlov) and operant conditioning (Skinner) are foundational. Students often confuse the unconditional stimulus (UCS) and the conditioned stimulus (CS) in a scenario. In Pavlov’s experiment, food is the UCS, the bell becomes the CS, and salivation to the bell is the conditioned response (CR). A common error is to identify the UCS as ‘the bell’ because the dog eventually responds to it. Learn to ask: ‘Does the stimulus naturally produce the response before any learning?’ If yes, it is a UCS.
经典条件作用（巴甫洛夫）和操作性条件作用（斯金纳）的原理是基础。学生常在一个情境中将非条件刺激（UCS）和条件刺激（CS）混淆。在巴甫洛夫的实验中，食物是UCS，铃声成了CS，听到铃声分泌唾液是条件反应（CR）。一个常见错误是，因为狗最终对铃声产生反应，就把UCS认定为“铃声”。要学会问：“在学习之前，该刺激是否天然地引起反应？”如是，则为UCS。
In operant conditioning, negative reinforcement is frequently confused with punishment. Negative reinforcement increases a behaviour by removing an unpleasant stimulus (e.g., taking a painkiller removes headache, reinforcing pill-taking). Punishment decreases a behaviour. Misidentifying these can lead to losing marks in application questions about therapies like systematic desensitisation or token economies.
在操作性条件作用中，负强化常与惩罚混淆。负强化是通过移除一个不悦刺激来增加行为（如服用止痛药消除头痛，从而强化服药行为）。惩罚则减少行为。将两者混淆会导致在系统脱敏或代币经济法等应用题目中失分。
9. Issues and Debates: Nature vs. Nurture and Beyond | 议题与争论：天性-教养及其他
The AS syllabus requires you to understand nature versus nurture, reductionism versus holism, determinism versus free will, and idiographic versus nomothetic approaches. A typical exam question asks you to explain how a core study illustrates one of these debates. A very common error is using extreme ‘all-or-nothing’ language: for example, saying ‘the biological approach is totally nature and ignores nurture’ when in fact many biological studies acknowledge an interaction (e.g., brain plasticity studies show experience can change biology).
AS大纲要求理解天性-教养、还原论-整体论、决定论-自由意志以及个案法-一般规律法。典型的考题要求你解释一项核心研究如何体现这些争论中的一个。一个非常常见的错误是使用极端的“全有或全无”的语言：例如说“生物取向完全属于天性而忽视教养”，然而事实上很多生物研究承认互动（如大脑可塑性研究表明经验可以改变生物学基础）。
Another mistake is mixing up reductionism and determinism. Reductionism concerns explanation at a lower, more basic level (e.g., reducing love to oxytocin levels), while determinism is about whether behaviour is predictable from prior causes. Make sure you can distinguish them and use appropriate examples from the four approaches.
另一个错误是将还原论与决定论混为一谈。还原论涉及在更低、更基础的层面上进行解释（例如将爱归结为催产素水平），而决定论关乎行为是否可由先前原因预测。请确保你能区分两者并使用四个取向中的恰当例子。
10. Evaluation Skills: Building a PEEL Paragraph | 评价技能：构建PEEL段落
One of the biggest reasons students lose marks on ‘evaluate’ questions is that they describe rather than analyse. A structured answer using Point, Evidence/Example, Explanation, and Link (PEEL) works well. Point: state a strength or weakness. Evidence: cite a specific detail from the study or a counter-study. Explanation: say why this matters for validity, reliability, ethics, or usefulness. Link: connect back to the question. Too many answers give a point (e.g., ‘lacked ecological validity’) with no evidence or explanation, leading to minimal credit.
学生在“评价”类题目上丢分的一大原因是只描述而不分析。用“观点、证据/例子、解释、扣题”（PEEL）结构作答效果好。观点：陈述一个优点或缺点。证据：引用研究中的一个具体细节或反例。解释：说明这为何对效度、信度、伦理或实用性重要。扣题：回联题目。太多答案只给出观点（如“缺乏生态效度”）而没有证据或解释，因此分数极低。
Also, avoid generic evaluative statements like ‘the sample was small, so it cannot be generalised’. Every sample can be criticised for size; what matters is why that specific sample limits the conclusions. For instance, Asch used male American undergraduates; you need to explain how this restricts generalisability to females or collectivist cultures.
还应避免泛泛的评价陈述，如“样本太小，因此不能推广”。任何样本都可能因大小受批评；关键在于为什么这个特定样本限制了结论。例如，阿希使用了美国男大学生；你需要解释这如何限制了向女性或集体主义文化的推广性。
11. Common Mistakes in Research Method Questions | 研究方法题中的常见错误
When designing an experiment in Paper 2, students frequently forget to operationalise the independent variable (IV) and dependent variable (DV). Saying ‘the IV is caffeine’ is insufficient; you must state how you will manipulate it (e.g., ‘participants drink 200 ml of caffeinated coffee (150 mg caffeine) vs. 200 ml of decaffeinated coffee’). Similarly, the DV must be measurable, such as ‘number of words correctly recalled from a list of 20’. Another common error is failing to mention random allocation to conditions, even though it is a fundamental control technique for independent groups designs.
在试卷二中设计实验时，学生常常忘记对自变量和因变量进行操作化定义。只说“自变量是咖啡因”是不够的；你必须说明如何操控它（如“参与者饮用200毫升含咖啡因的咖啡（150毫克咖啡因）对比200毫升脱因咖啡”）。同样，因变量必须可测量，如“从20个词语列表中正确回忆的个数”。另一个常见错误是忘记提及随机分配至条件，尽管它是独立组设计的基本控制技术。
For observation, a frequent oversight is not deciding on behavioural categories before data collection. Vague categories like ‘aggression’ cannot be reliably tallied. You need operationalised categories such as ‘hitting with closed fist’, ‘kicking’, and ‘verbal threats’. Also, many students fail to specify inter-observer reliability procedures, which are crucial for the credibility of an observation.
对于观察法，常见的疏忽是没有在数据收集前确定行为类别。像“攻击性”这样的模糊类别无法被可靠地计数。你需要操作化的类别，如“用拳击打”、“脚踢”和“言语威胁”。此外，很多学生未能说明观察者间信度的程序，而这对于观察的可信度至关重要。
12. Time Management and Reading the Question | 时间管理与审题
Under the pressure of an exam, some candidates dive into writing without breaking down the question. Look for the command term and the content focus. Underline key words such as ‘ethical’, ‘reliability’, or ‘nurture’. If a 6-mark question asks ‘Explain one ethical issue and how it was dealt with’, do not write about two issues — depth over breadth. Also, allocate your time proportionally: a 10-mark essay on Paper 1 deserves about 15 minutes, including planning. Rushing through a high-mark question usually results in superficial answers that miss the evaluation demanded.
在考试压力下，有些考生不加分析就动笔。应该先找出指令词和内容焦点。在关键词如“伦理”、“信度”或“教养”下划线。如果一个6分题要求“解释一个伦理问题及其解决方式”，不要写两个问题——深度优于广度。同时，要按比例分配时间：试卷一的10分论述值得花大约15分钟，包括规划。仓促完成高分值题目通常导致答案肤浅，遗漏要求的评价。
Finally, use the mark allocation as a clue: if a question asks you to ‘Describe and evaluate’ with 12 marks, around 6 marks are for description and 6 for evaluation. Many students spend too long describing and leave no time for evaluation, capping their score at around half the total.
最后，利用分值作为线索：如果一题要求“描述并评价”共12分，大约6分给描述，6分给评价。许多学生在描述上耗时过长，无暇评价，使得分数被限制在总分的一半左右。
Published by TutorHao | Psychology Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导