AQA GCSE心理学 Memory 多存储模型 重构记忆
AQA GCSE Psychology 的记忆单元在 Paper 1 里占了四分之一的分值。官方考纲 8182 把这一章压成三块内容：记忆的过程、多存储模型、以及记忆作为主动重建的过程。试卷上的题型从一分的选择题一直排到九分的论述题。同一个知识点会用四种不同深浅的方式被问到。所以背下定义与会答题之间的距离，比多数人预想的要远。下面按考纲的三个小节把必考的知识点拆开，每个知识点给出术语、研究证据与最常见的失分方式。
The Memory unit of AQA GCSE Psychology (8182) carries a quarter of the marks on Paper 1. The specification reduces the chapter to three blocks: the processes of memory, the multi-store model, and memory as an active process. Questions on this content run from one-mark multiple choice items up to nine-mark extended writing, so a definition that has been memorised is not the same thing as an answer that earns the marks; the difference is usually terminology used under pressure.
编码、存储、提取：记忆的三个过程
考纲把记忆拆成三个过程：编码、存储、提取。编码是信息进入系统并被转换成可保存的形式，可以按图像编码，按声音编码，也可以按意义编码。存储是把编码后的信息保留一段时间。提取是把信息重新取出来用于回答或行动。短时记忆以声学编码为主，长时记忆以语义编码为主，Baddeley 的相似词实验是支持这一区分的常用证据。考纲额外要求区分三类长时记忆：情景记忆记录个人经历，语义记忆保存事实与概念，程序记忆对应技能动作。很多学生以为记不住就是提取失败，把三个过程当成一件事。常见的错误是只回答"没背熟"，说不出断在编码、存储还是提取哪一步。
AQA splits memory into encoding, storage and retrieval, and expects candidates to use the three terms separately rather than as synonyms for learning. Encoding converts incoming information into a storable form, usually as an image, a sound pattern or a meaning; storage holds it over time; retrieval brings it back when a question demands it. Baddeley's experiments with acoustically and semantically similar word lists are the standard evidence that short-term memory leans on sound while long-term memory leans on meaning. The specification also separates episodic, semantic and procedural long-term memory, which is why answers about "remembering my last exam" and "remembering the meaning of E=mc2" should not use the same label.
多存储模型：三个存储的编码 容量 持续时间
Atkinson 与 Shiffrin 的多存储模型把记忆画成三个依次相连的存储，考纲要求写出每个存储的编码方式、容量与持续时间。感觉记忆容量很大，但只保留极短时间，视觉图像大约半秒就消失。短时记忆以声学编码为主，容量约七个加减两个项目，Miller 的整理研究是常被引用的依据；不复习的情况下，信息在十几秒到三十秒内丢失，Peterson 与 Peterson 用倒数三位数阻止复述，十几秒之后几乎没有被试还能回忆。长时记忆容量在理论上没有上限，持续时间可以覆盖一生，Bahrick 让毕业生辨认老同学的照片，几十年后仍保持很高的辨认准确度。信息从短时记忆进入长时记忆靠复述，复述一停，链条就断在中途。这里最容易失分的是把容量与持续时间两个数字写反，写反之后整道题的描述分一起丢。
Atkinson and Shiffrin drew memory as three linked stores, and the specification asks for coding, capacity and duration for each one. Sensory memory is large but extremely brief, with visual traces lasting roughly half a second, while short-term memory codes acoustically, holds about seven items plus or minus two, and loses them within tens of seconds unless they are rehearsed; Peterson and Peterson demonstrated the loss by stopping rehearsal with a counting task. Long-term memory is effectively unlimited in capacity and can last a lifetime, which Bahrick's recognition tests of former classmates support decades after graduation. Rehearsal is the transfer mechanism in this model, so a candidate who reverses capacity and duration figures tends to lose the description marks for the whole question.
系列位置曲线：首因效应与近因效应
考纲点名了 Murdock 的系列位置曲线研究，这是支撑多存储模型的核心证据。Murdock 给被试一串词，要求按任意顺序回忆，再把回忆率对词在表里的位置作图，得到一条两端高、中间低的曲线。开头的词回忆得好，这叫首因效应，它们在有限时间里被复述进了长时记忆。结尾的词回忆得好，这叫近因效应，它们还停在短时记忆里，没有被打断。中间部分的词成绩最差，它们既来不及被复述，又被后面进来的词顶出了短时记忆。这条曲线的价值在于两个高点各有独立的解释，说明系统里存在两个功能不同的存储。答题时要把效应名称、所在位置与原因三件事对应写清，只写"中间记不住"拿不到分析分。
Murdock's serial position curve is named in the specification and is the reason the multi-store model has survived in the classroom. Participants heard a list of words and recalled them in any order; plotting recall against position produced a curve that rises at the start, dips in the middle and rises again at the end. The primacy effect at the start reflects rehearsal into long-term memory, the recency effect at the end reflects items still sitting in short-term memory, and the middle items suffer on both counts. What makes the evidence strong is that the two peaks need separate explanations, which is difficult to reconcile with a single store.
重构记忆：Bartlett 的 War of the Ghosts 与 effort after meaning
重构记忆理论认为记忆不是录像回放，而是每次提取时借助 schema 重新搭建出来的产物。Bartlett 让英国被试反复复述一个北美原住民故事 War of the Ghosts，复述之后故事里的独木舟变成了船，陌生的细节被删掉或被替换成英国读者熟悉的内容，故事的形状越来越像被试自己文化里的叙事。effort after meaning 说的是被试努力给陌生材料找一个说得通的解释，而这个解释动作本身就在改写材料。考纲要求会描述并且评价这项研究。它的优点是材料来自真实的阅读与复述情境，接近日常记忆的使用方式；局限是复述次数与时间间隔没有严格控制，不同被试走的路不一样，结果很难被精确重复。答题时流程、结果、结论三块都要写，只写结论会丢掉描述分。
Reconstructive memory treats recall as a rebuild rather than a replay, and Bartlett's War of the Ghosts study is the set piece for this idea. British participants retold a North American folk tale, and across retellings the canoe turned into a boat while unfamiliar details were dropped or replaced with something culturally familiar, so the story drifted towards the shape of the reteller's own expectations. Effort after meaning describes the listener working to make sense of strange material, and that work is itself an edit. The study is praised for using realistic reading and retelling rather than artificial word lists, but criticised because retelling intervals and repetition counts were not tightly controlled, which makes exact replication difficult.
前摄干扰与倒摄干扰：相似材料互相顶掉
干扰指一条记忆妨碍另一条记忆的提取，考纲把它与情境、虚假记忆并列，算作影响记忆准确度的因素。前摄干扰是先前学的内容挡住后来学的内容，先学法语再学德语时，法语词汇会不断冒出来干扰新词的提取。倒摄干扰是后来学的内容挡住先前学的内容，考完化学再回头做数学，公式会一时读不顺。McGeoch 与 McDonald 让被试学习两组材料，两组内容相似时第二组的回忆成绩比单独学习时更差，这说明遗忘不能只归因于时间流逝。考场上错得最多的是把两种干扰的名字写反，只写"记混了"不得分。判断方法不复杂，看破坏回忆的是先前学的还是后来学的内容。
Interference describes one memory blocking the retrieval of another, and the specification groups it with context and false memories as a factor affecting accuracy. Proactive interference is old learning disrupting new learning, which is why French vocabulary surfaces while a student is trying to produce German; retroactive interference is the reverse, where a later chemistry revision session makes earlier maths formulas harder to retrieve. McGeoch and McDonald showed that two similar sets of material produced weaker recall than one set alone, which rules out the passing of time as the only cause of forgetting. Examiners regularly see the two labels swapped, and a vague answer about memories "getting mixed up" earns nothing.
情境与提取线索：Godden 与 Baddeley 的水下实验
提取失败的解释认为信息仍然留在长时记忆里，只是当时缺少合适的线索把它调出来。线索可以是地点、气味、音乐，也可以是情绪状态，只要与编码时的情境接近，回忆就更容易。Godden 与 Baddeley 让大学潜水俱乐部的成员在水下或岸上学习词表，再在水下或岸上回忆，四种组合都测了一遍，学习与回忆处在同一环境时成绩最好，换到另一个环境时明显下降。这项结果解释了学生的真实体验：在教室里做熟的题，换到陌生考场会做得慢一些。复习时把练习条件与考试条件对齐，用相同的限时、相同的纸笔方式、相同的座位状态，比换着花样做题更接近考试当天的提取条件。答题时写清情境一致与提取线索这两个关键词，答案才算落在考纲上。
Retrieval failure suggests the information is still in long-term memory but cannot be reached without a suitable cue. A cue can be a place, a smell, a piece of music or a mood, and it works best when it matches the state in which the material was encoded. Godden and Baddeley had university divers learn word lists underwater and on land, then recall them in either environment, and performance was highest when learning and recall took place in the same setting. Students recognise the pattern from their own exam experience, and the practical response is to rehearse under conditions that resemble the exam room rather than under comfortable ones.
虚假记忆：提问措辞如何改写记住的细节
虚假记忆指记得并没有发生的事，或者把发生过的细节记错了。考纲把虚假记忆与干扰、情境并列，并把它接到重构记忆上，因为 schema 在重建事件时会填入合理但并不真实的内容。Loftus 与 Palmer 让被试观看交通事故的影片，随后用不同强度的动词提问，一周后再问影片里是否出现碎玻璃，被问过更激烈动词的被试更常报告并不存在的碎玻璃，说明提问措辞会进入记忆并改写细节。这项研究常被用来讨论目击证词的可靠性。它的局限同样明显，材料是实验室影片而不是真实事故，情绪强度与后果都无法复制。答题时要把记忆被改写与当时没看清两种情况分开，这是两个不同的解释。
False memories are recollections of events that did not happen, or of details that happened differently. The specification links them to reconstructive memory, because a schema fills gaps with plausible material that the person then treats as remembered fact. Loftus and Palmer showed participants films of traffic accidents and questioned them with verbs of different intensity; a week later those who had been asked about a violent collision were more likely to report broken glass that never appeared in the film. The study is widely used to question the reliability of witness statements, although its laboratory films carry less emotional weight than a real crash.
8182 与 7182：GCSE 记忆与 A-Level 记忆的边界
同一章内容在两个资格里并不相同，背了超纲的部分等于白花时间。AQA GCSE 考纲 8182 的记忆部分只需要多存储模型，三个存储就是感觉、短时、长时；工作记忆模型写在 A-Level 的 7182 里，GCSE 不考。目击证词在 8182 中没有独立小节，它只作为虚假记忆的应用场景出现，考纲改版时删掉 eyewitness testimony 一节，这一点可以从 AQA 官网的改动说明里查到。分类方面，情景、语义、程序三类长时记忆是 8182 新增的内容，首因效应、近因效应与 Murdock 的研究也从 2017 年首次施教的版本开始写进考纲。复习时按 8182 的小节逐条核对，能省下相当一部分被无关内容占用的时间。
The exam board's two Psychology specifications overlap but are not identical, and revision aimed at the wrong level wastes time. AQA GCSE 8182 needs the multi-store model with its sensory, short-term and long-term stores; the working memory model belongs to A-Level 7182 and is not examined at GCSE. Eyewitness testimony has no standalone section in 8182 and appears only as an application of false memories, which the board's own summary of changes confirms. Types of long-term memory and the serial position effects entered the GCSE specification with first teaching in 2017, so checking each sub-unit against 8182 is a reliable way to cut irrelevant reading.
把干扰与情境写进复习安排
两个因素都可以直接换成复习动作，不需要额外的时间。前摄干扰的常见来源是把两门相似科目排在同一天，比如西班牙语与法语，或者物理与数学里公式密集的章节，改成隔天交错能明显减少互相顶掉。中间位置的劣势还提示一件事：一份提纲或一节课里最居中的那几条内容最容易被漏掉，复述时把它们单独拎出来过一遍，比从头通读一遍有效。情境线索从考场条件入手，用相同的限时、相同的手写方式，不用计算器的卷子就不要用计算器练。提取练习的做法是合上书，用白纸写出某个知识点的术语、研究者与结论，写不出来的地方当晚补上。分散复习把同样的时间拆成三次，多数学生的反馈是比一次连着做完记得更牢。
Both findings convert into timetable decisions rather than extra hours. Proactive interference usually comes from placing two similar subjects on the same day, so Spanish and French, or the formula-heavy chapters of physics and maths, are better split across alternate days. The weak middle of a list carries a practical warning as well: the items buried in the centre of a revision sheet are the ones most often skipped, and they deserve a separate pass. Context cues are easiest to copy from the exam itself, with the same time limit, the same handwriting and no calculator when the paper forbids one, while retrieval practice means closing the book and writing out terms, researchers and conclusions from memory before repairing the gaps.
9 分题作答：结论 证据 术语
Paper 1 的记忆部分有 extended writing 题，通常要求描述并评价某项研究或某个模型，满分九分。写法上先给一个明确结论，再用研究证据支撑，最后用考纲术语收口。描述部分按参与者、程序、结果、结论四步走，一步一句。评价部分给出优点与局限各一条，并说明为什么这条算优点或局限，例如多存储模型把短时与长时分开是它的解释优势，而把复述当成唯一通道是它的局限，无法解释只靠意义加工就能长期记住的内容。阅卷时见过很多写满一页却没有术语的答案，分数停在描述分上不去。另一种常见丢分是把评价写成个人感受，评价必须落在方法或解释力上。时间上给九分题留十二到十五分钟，动笔前先用三个词写出提纲。
Extended writing on Paper 1 usually asks candidates to describe and evaluate a study or a model for nine marks. The safe shape is a clear point first, evidence second and terminology throughout, with the description following participants, procedure, results and conclusion in separate sentences. Evaluation needs one strength and one limitation, each with a reason, such as the multi-store model's separation of short and long-term storage on the credit side and its reliance on rehearsal as the only route on the debit side. Answers that fill a page without technical terms tend to stall at description marks, and personal reactions are not evaluation; the judgement has to rest on method or explanatory power.
牛津AQA国际GCSE生物?OxfordAQA International GCSE Biology (9201): Onl?进口原版¥399.0📱 去京东购买京东精选需要一对一辅导？Need One-on-One Tutoring?
A-Level / GCSE / IB / KS2-KS3 全科辅导
16621398022 同微信
Visit aleveler.com
或截图后用「扫一扫 → 相册」选择本图