AQA心理 GCSE 知觉 深度线索 错觉 文化
AQA GCSE Psychology 的 Paper 1 里，知觉（Perception）只是四个板块之一，却是错觉图形与跨文化研究最集中的一块。这一篇按答题需要把 3.1.2 拆开：感觉与知觉的分界、单眼与双眼深度线索、Gibson 与 Gregory 两套理论、四类错觉解释，以及文化、动机、情绪、期望四个影响因素。8182 的卷面上，这些内容都能落到两分题、四分题与六分题上。
Perception is one of four topics examined in Paper 1 of AQA GCSE Psychology, and it is the part where visual illusions and cross-cultural research carry the most marks. This article follows section 3.1.2 of the specification in the order questions usually arrive: sensation and perception, depth cues, the two theories, the four explanations of illusions, and the four factors that shape what we see. Each idea below can be written straight into a two-mark, four-mark or six-mark answer.
Paper 1 卷面结构与知觉的题型位置
AQA GCSE Psychology 的代码是 8182，2017 年 9 月起施教，2019 年首次考试。Paper 1 Cognition and behaviour 时长 1 小时 45 分钟，满分 100 分，占 GCSE 总成绩的一半，卷面分成四个部分，每部分 25 分，考的都是选择题、短答题与扩展写作。知觉属于 3.1.2，与记忆、发展、研究方法同卷出现，同一张卷子上会反复要求你切换话题。官方 spec 给出的评估目标权重是 AO1 35%、AO2 35%、AO3 30%，折到每张卷子是 AO1 17.5%、AO2 17.5%、AO3 15%。纯背定义的题只占三分之一，其余要靠把知觉概念用进陌生情境。很多学生以为记住定义就能拿分，实际上情境题的分值往往压在六分题上。
AQA GCSE Psychology is coded 8182, taught from September 2017 with the first exams in 2019. Paper 1 is called Cognition and behaviour; it lasts 1 hour 45 minutes, carries 100 marks and is worth half of the GCSE, split into four sections of 25 marks that mix multiple choice, short answer and extended writing. Assessment objectives are weighted AO1 35%, AO2 35% and AO3 30% overall, so about two thirds of the marks depend on applying and evaluating rather than recalling a definition.
感觉与知觉的分界: 眼睛收到的是二维图像
感觉（sensation）是感觉器官接收外界物理刺激的过程，光线落在视网膜上、声音到达耳蜗，都还只是原始信号。知觉（perception）是大脑对这些信号作出解释并赋予意义，把二维的视网膜图像变成有深度、有距离的三维世界。看一页书时，感觉到的是纸上的黑色线条，知觉到的是句子在说什么。Müller-Lyer 错觉能证明两者是分开的步骤：两条线段在视网膜上的成像长度完全相同，感觉没有差别，知觉却报告其中一条更长。常见的错误是答题时把 sensation 与 perception 当成同义词互换，评分方案里这两个词各占独立的给分点。很多学生以为错觉是眼睛看错了，其实眼睛给出的信息没有问题，出错的是解释这一步。
Sensation is the detection of physical energy by a sensory receptor, such as light falling on the retina, while perception interprets that raw signal and gives it meaning. A page of text shows the split clearly: the sensation is a pattern of dark lines, and the perception is the sentence you understand; the Müller-Lyer illusion makes the same point, because two lines of identical length on the retina are still reported as different. Many students lose marks here by treating the two words as interchangeable, whereas the mark scheme awards them separately.
单眼深度线索: 高度、相对大小、遮挡与线性透视
spec 列出的单眼深度线索有四条，只用一只眼睛、甚至只看一张平面照片也能起作用。高度（height in plane）指画面中位置更靠上的物体被认为更远；相对大小（relative size）指同类物体中成像更小的那个更远；遮挡（occlusion）指被另一个物体挡住的部分更远，挡住别人的那个更近；线性透视（linear perspective）指平行的线向消失点收拢，铁轨与公路是最常见的例子。照片与写实绘画靠这四条线索就能让人读出纵深，所以二维图像也能骗过三维经验。容易记反的是遮挡，被遮住的物体在后，不是在前。判断图片题时把四条线索逐条对图找出来，比凭感觉描述更稳。
The specification names four monocular depth cues, and all of them work with one eye as well as inside a flat picture. Height in plane places higher objects further away, relative size makes the smaller of two similar objects the more distant one, occlusion means the covering object is the nearer one, and linear perspective is the convergence of parallel lines towards a vanishing point. Occlusion is the cue students most often reverse, treating the hidden object as the closer one.
双眼深度线索: 视网膜像差与辐辏
双眼线索需要两只眼睛同时工作，精度比单眼线索高。视网膜像差（retinal disparity）来自两眼相距约六到七厘米，同一物体在左右眼的投影略有错位，错位越大物体越近，这个线索在十米之后基本失效，因为远处的投影几乎重合。辐辏（convergence）来自眼肌的反馈，看近物时双眼向内转的角度更大，大脑依据这个角度判断距离。3D 电影正是给左右眼各送一张略有差异的画面，用视网膜像差造出屏幕之外的纵深。阅卷时常见的是把两个概念写反，把辐辏说成两眼图像差异，一分就这样丢掉。
Binocular depth cues need both eyes and are more accurate than the monocular set. Retinal disparity comes from the eyes being about six to seven centimetres apart, so each eye receives a slightly different image; the bigger the difference, the closer the object, although beyond roughly ten metres the two images nearly match. Convergence works through the eye muscles, which turn the eyes further inwards for near objects, and 3D cinema exploits disparity by feeding each eye its own slightly different picture.
Gibson 直接理论: 运动视差与不变量
Gibson 的直接理论属于天性一派，认为知觉是自下而上的过程，环境里的光学阵列本身包含足够信息，不需要推断。他给出的关键证据是运动视差（motion parallax）：观察者移动时，近处的物体在视野里快速后退，远处的物体几乎不动，速度差本身就编码了距离。坐在车里看窗外最直接，近处的护栏一闪而过，远处的山几乎静止。不变量（invariants）指视野里稳定的结构，例如地平线与纹理梯度，它们让运动中的判断保持准确。这套解释的长处在于速度快，运动员与司机不需要逐帧推理就能行动。
Gibson's direct theory sits on the nature side and treats perception as a bottom-up process, because the optic array already carries the information an observer needs. Motion parallax is the everyday evidence behind it: as you move, nearby objects sweep past quickly while distant ones barely shift, and that difference in speed encodes distance without any inference. Invariants such as the horizon and texture gradients keep those judgements stable while the observer is moving, which is why driving and sport feel fast rather than calculated.
Gibson 理论的评价: 婴儿研究与错觉的反证
支持的一面来自发育研究，婴儿在会走路之前就对逼近的物体有反应，说明距离判断至少有一部分是与生俱来的。运动视差也解释了熟练动作为什么不需要思考，节省了有限的注意资源。问题出在错觉上：Müller-Lyer 的两条线在光学阵列里完全等长，直接理论没有理由让它们看起来不同，Gregory 的推断说却能自然解释。静止图片里没有运动视差，我们照样读得出纵深，这说明还有别的机制在参与。跨文化研究进一步说明，同一张图纸在不同背景的人眼里读法不同，这一点也不是直接理论能够容纳的。
Support for Gibson comes from developmental research, in which infants respond to looming objects before they can walk, suggesting at least part of depth perception is innate. The theory also explains the speed of skilled movement, since an observer does not have to reason about every approaching step. Visual illusions are its weakest point, because two equal lines in the optic array have no reason to look different yet they do, and motion parallax cannot account for depth in a still photograph.
Gregory 建构理论: 知觉假设与推断
Gregory 的建构理论属于教养一派，把知觉看成自上而下的过程。感官输入进入大脑后立即触发一个知觉假设（perceptual hypothesis），也就是对眼前事物最可能的解释，再由过去的经验与图式加以修正，最终建构出一个关于现实的模型。海面上露出一个弯曲的鳍，你立刻判断是鲨鱼，这就是用存储的知识填补感官信息的空隙。线索不充分时推断的作用最大，模糊图形可以有两种读法，读到哪一种取决于你先前的经验。这套理论把错觉解释为误用的假设，人把三维世界的经验套到二维图像上，于是作出了错误但合理的猜测。
Gregory's constructivist theory sits on the nurture side and describes perception as top-down, driven by inferences from visual cues and past experience. Sensory input triggers a perceptual hypothesis, a best guess about what is being looked at, which stored knowledge and schemas then confirm or correct until a model of reality is built. The theory explains illusions as mistaken hypotheses, where three-dimensional experience is applied to a flat picture and produces a wrong but reasonable answer.
Gregory 理论的评价: SB 个案与跨文化证据
Gregory 与 Wallace 在 1963 年报告了个案 SB，一位五十年来一直失明的人通过手术恢复视力，术后他难以判断深度与距离，过马路需要重新学习，这些能力随练习逐步改善，说明距离知觉不是天生就绪的。跨文化研究也支持教养一侧，同一张带有深度线索的图画，不同成长环境的人读出的结果不同。批评集中在两点：SB 恢复视力后并不需要重新学习识别每一个物体，所以经验的作用被夸大了；婴儿对图案化刺激注视更久，说明感觉输入在早期就有意义。两套理论更接近互补，简单环境靠直接拾取，复杂或模糊的刺激靠推断。
Gregory and Wallace reported the case of SB in 1963, a man who regained his sight after fifty years of blindness and then struggled with depth and distance, needing practice before he could cross a road safely. Cross-cultural studies point the same way, because people from different backgrounds read the depth cues in one drawing differently, and babies look longer at patterned stimuli, which suggests sensory input carries meaning early on. The fairest conclusion is that direct pick-up and inference both operate, with inference doing more work on ambiguous or artificial images.
四类错觉解释与六种经典图形
spec 要求掌握四类解释，并且能在具体图形上对号入座。歧义（ambiguity）指图形本身可以有两种读法，Rubin 的花瓶与 Necker 立方体都属于这一类，看的人会在图与底之间来回切换。误判深度线索（misinterpreted depth cues）指大脑把二维图上的线条当成三维场景，Ponzo 的两条等长横线因为夹在收拢的直线之间而被看作远处那条更长，Müller-Lyer 的箭头则模拟了墙角与屋角。虚构（fiction）指图形并不完整，Kanizsa 三角的边与白色三角形都是大脑补出来的轮廓。大小恒常（size constancy）解释 Ames 房间，房间本身不规则，两个人看起来相差数倍。分数通常分给写出图形与写出解释两处，只写图形名拿不到解释分。
Four explanations cover the illusions named in the specification, and each figure belongs to one of them. Ambiguity appears in Rubin's vase and the Necker cube, where one picture supports two readings; misinterpreted depth cues explain the Ponzo and Müller-Lyer illusions, because flat lines are treated as converging edges or as corners of a room; fiction covers the Kanizsa triangle, whose edges the brain supplies; and size constancy accounts for the Ames room, where two people of similar height look wildly different. Marks are usually split between naming an illusion and explaining it, so a name by itself earns little.
文化因素: Hudson 1960 图画测试
Hudson 在 1960 年用一张二维图画做了跨文化测试，画面里一个猎人举着长矛，前面是羚羊，后面是大象，纵深靠单眼线索表现。他让南非的学童与成年人看这张图，回答猎人瞄准哪只动物、哪只动物更近。很少接触西式图画或正规学校教育的参与者经常读不出这些线索，有人甚至认为大象更近，或认为猎人瞄准的是远处的大象。他们的视力没有任何问题，差别出在对图画惯例的经验上。这个结果支持教养一侧，也解释了为什么刚来英国不久的学生会在图片题上丢分。答题时把结论写成知觉受经验与文化影响，比只写文化有影响更容易拿到第二分。
Hudson's 1960 picture test used a two-dimensional drawing of a hunter with a spear, an antelope in front and an elephant behind, with depth carried entirely by monocular cues. Adults and schoolchildren in South Africa who had little contact with Western pictures often failed to read those cues, sometimes reporting the elephant as nearer or the spear as aimed at it, even though their eyesight was normal. The results support the nurture side and explain why newcomers to a British classroom can lose marks on picture questions until they learn the conventions.
动机、情绪、期望与知觉定势
知觉定势（perceptual set）是一种心理准备状态，它使人更可能注意某一类线索而忽略其他。Gilchrist 与 Nesberg 在 1952 年做了动机研究，参与者是 26 名大学生，实验组在 20 小时内只喝水不进食，对照组照常吃饭，两组都先看四张餐食照片（牛排、炸鸡、汉堡、意大利面）15 秒，再在屏幕重新亮起时把亮度调回原来的样子，测试安排在午餐之后、间隔 6 小时与 20 小时三个时间点。饥饿程度越高的参与者把照片调得越亮，控制组基本准确，说明动机改变了知觉。期望的作用由 Bruner 与 Minturn 在 1955 年证实，同一个模棱两可的符号，先看字母序列的人读成字母 B，先看数字序列的人读成 13。情绪的方向相似，负面情绪下模糊面孔更容易被解释成有敌意。常见的错误是把两个研究互换对应，把 Bruner 与 Minturn 写成动机研究。
Perceptual set is a readiness to notice certain cues and ignore others, and it is studied through culture, motivation, emotion and expectation. In the study by Gilchrist and Nesberg in 1952, twenty-six university students took part, with the experimental group going twenty hours on water only; both groups adjusted the brightness of four meal photographs at three time points, and the hungrier participants set them brighter than the control group. Expectation was shown by Bruner and Minturn in 1955, where the same ambiguous figure was read as the letter B after a letter sequence and as the number 13 after a number sequence.
三周复习排期与六分题落笔口径
答题时分值决定写法，两分题只要求给出两个独立的点，四分题要求描述加解释，六分题要在情境里应用并给出评价。涉及研究的分值要先写名字与年份，再写参与者数量、时长这类具体细节，Gilchrist 与 Nesberg 的 26 名大学生、20 小时禁食、15 秒呈现都是可以直接写进答案的给分点。评价题需要优缺点各一侧，只写支持某某理论写不满两分。第一周把单眼与双眼线索与六种错觉图形配对，第二周整理两套理论各自的支撑与反证，第三周用两个研究练习情境题，并按 1 小时 45 分钟计时做一份完整 Paper 1。术语全篇统一，一处写 retinal disparity 就不要中途换成双眼视差，评分方案认的是官方术语。
The number of marks decides the shape of an answer: two marks need two separate points, four marks need description plus explanation, and six marks need application to a scenario plus evaluation. When a study is named, add the year and the concrete details such as twenty-six participants, twenty hours without food and a fifteen-second presentation, because those count as separate credit points. A workable three-week plan pairs the depth cues with the six illusions first, then compares the strengths and weaknesses of both theories, and finishes with a full timed Paper 1.
牛津高阶英汉双解词典 第10版 商务印书馆¥168📱 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