3.1 AS Level content - core studies
- Syllabus
- 9990–2028–2029
- Section
- 3.1
- Level
- AS

| Main syllabus assumption | What it directs psychologists to investigate | Bounded inference |
|---|---|---|
| Behaviour, cognitions and emotions can be explained through brain function and effects of hormones, genetics and evolution | Neural activity/structure/localisation, endocrine effects, inherited variation, adaptation and species/sex similarities | A measured association/mechanism can support a biological contribution; it does not show biology is the only cause |
| Similarities and differences between people can be understood through biological factors and their interaction with other factors | Individual/group differences plus experience, learning, culture, social context and environment | Interaction means experience can alter biology and biological predispositions can shape responses to context |
Mechanism examples: EEG patterns index real-time sleep function; hormones/evolution can generate predictions about sex-linked behaviour; MRI can measure structural change after experience. Operational measures must be connected to the construct and alternative explanations.
| Core study | Biological evidence | Interaction/limit |
|---|---|---|
| Dement and Kleitman | REM/EEG and eye-movement patterns relate to dream recall/duration/content | Self-reported dreams and artificial awakenings; association is not a dream-reading device |
| Hassett et al. | Rhesus sex differences in wheeled/plush interaction parallel some child patterns | Monkey social rank/experience and human toy meanings remain; hormones were not directly manipulated |
| Hölzel et al. | Grey-matter concentration changed after an eight-week MBSR course | Experience-linked plasticity itself demonstrates biological–environment interaction; self-selection/bundled course limit specificity |
| Issue/debate | Biological approach question |
|---|---|
| Nature–nurture | Which biological predispositions, experience and interactions explain the result? |
| Reductionism–holism | What is gained/lost by reducing experience to EEG, hormones or grey matter? |
| Determinism–free will | Does a biological mechanism constrain behaviour without making it inevitable? |
| Individual–situational | Is variation located in person biology, current environment or both? |
| Application/ethics | Can findings improve sleep/stress/welfare without overclaiming, stigma or harm? |
Biological does not mean fixed, inherited or purely innate. Brain structure/function can be shaped by experience; correlation, group difference and objective technology each require design-based evaluation before causal/general claims.
Background: sleep cycles through REM and NREM in an ultradian rhythm, but dreaming had relied on retrospective self-report. Aims: test whether dream recall differs after REM versus NREM awakening; whether estimated dream duration corresponds to 5 versus 15 minutes of REM; and whether eye-movement direction relates to dream content.
Opportunity sample: nine adults (seven men, two women). Five were studied intensively; four provided minimal confirmatory data. Participants arrived at/about usual bedtime, ate normally and avoided alcohol/caffeine that day.
| Study feature | Operational detail |
|---|---|
| Sleep recording | EEG electrodes on scalp/head recorded brain activity continuously; electrodes beside eyes recorded eye movements; leads were gathered into one cord at top of head for movement |
| Awakening | Researchers identified REM/NREM and woke participants with a sufficiently loud doorbell at scheduled/random relevant points |
| Dream report | Participant immediately said whether dreaming and described content into a recorder; researcher could ask for detail |
| Duration test | On selected REM awakenings participant judged whether dream had lasted 5 or 15 minutes |
| Eye-content test | Vertical, horizontal, mixed or little eye movement before awakening was compared with reported visual action |
| Design/data | Repeated observations within participants, experimental REM/NREM/5–15 comparisons plus correlations; physiological quantitative and interview qualitative data |
| Question | Result | Inference |
|---|---|---|
| REM/NREM recall | 152 dreams were recalled after REM; dream reports occurred on 79.6% of REM awakenings versus 6.8% of NREM awakenings | Dream recall is much more likely in REM, not exclusive to it |
| Duration | 15-minute estimates were correct on 47 occasions and incorrect on 13 (about 78% correct); 5-minute judgements were more accurate | Dream experience broadly tracks elapsed REM time |
| Eye direction/content | Vertical movements accompanied looking/climbing up/down (cliff hoist, ladder, basketball); horizontal movements accompanied side-to-side scenes; little movement accompanied distant/fixed viewing | Eye-movement patterns can correspond to visual dream activity |
EEG/eye recordings offered objective temporal markers that converged with immediate reports: dreaming is strongly associated with REM, duration estimates broadly reflect REM duration and some eye patterns reflect dream imagery. The results do not show that dreams never occur in NREM or that EEG reveals content directly.
| Strength | Weakness |
|---|---|
| Standardised continuous EEG, timed awakenings and immediate reports support replication and reduce long recall delay | Sleep laboratory, electrodes and repeated bell awakenings reduce ecological validity and may alter sleep/dream recall |
| Physiological records plus qualitative reports triangulate objective state and subjective experience | Tiny sex-imbalanced sample, with only five intensive cases, limits generalisation and individual independence |
| Repeated measures control many participant differences; quantitative comparisons/correlations allow pattern testing | Demand/self-report reconstruction and researcher questioning may affect content; correlational eye-content/duration links do not prove simple cause |
Ethical/practical concerns include disrupted sleep, electrodes and privacy of intimate dream reports; consent, withdrawal, confidentiality and monitoring harm matter. Biological/reductionist strength comes from objective sleep markers, while the subjective report and laboratory context show why a fuller explanation needs cognitive and situational evidence.
Ethical applications include scheduling awakening near REM for dream research/recall, improving sleep/dream education and using objective sleep-stage monitoring in clinical assessment—without claiming to decode thoughts or diagnosing from this small sample alone.
REM predicts a high probability of report, not a necessary/sufficient rule for all dreaming. Keep each conclusion tied to its measure: EEG stage, timed REM, eye pattern and verbal report answer different questions.
Human sex differences in toy preference are often explained by socialisation or biology. Rhesus monkeys share relevant social/hormonal systems but have not received human toy-gender teaching. Hassett et al. tested whether male/female monkeys differed in interaction with wheeled ('masculine') and plush ('feminine') toys in a pattern paralleling children.
The socially housed matrilineal troop at the Yerkes Center contained 135 rhesus monkeys and had lived together for over 25 years. After exclusions (including infants not reliably coded and hormonally treated animals), the analysed sample was 34 monkeys: 11 males and 23 females, ranging from 3 months to over 13 years.
| Trial step | Operational detail |
|---|---|
| Prepare | Troop held indoors while one wheeled and one plush toy were placed 10 m apart in outdoor enclosure |
| Control position | Left/right toy category placement was counterbalanced across trials |
| Observe | Monkeys released outdoors for a 25-minute trial; a separate camera focused on each toy |
| Code | Recorded interactions were coded with behavioural checklist for frequency and duration; examples included touch, carry and drag (moving toy along ground behind monkey) |
| Categories | Wheeled examples: wagon, truck, car, construction vehicle, shopping cart/dump truck; plush examples included familiar soft animal/doll toys |
| Analyses | Sex × category comparisons, individual preference and Spearman correlations with dominance rank; child data from another study supplied cross-species comparison |
| Finding | Accurate interpretation |
|---|---|
| Males | Strong wheeled preference: 73% preferred wheeled toys; wheeled interaction averaged about 10 occasions and plush duration about 4.76 minutes |
| Females | No significant category preference overall; plush interaction frequency was higher than males and plush duration was only slightly greater than wheeled |
| Duration comparison | Males interacted with wheeled toys longer than plush and longer than females; male monkey/child patterns both favoured wheeled/masculine toys |
| Rank | Across sexes rank correlated positively with frequency for both categories; among females rank related to plush frequency/duration and wheeled frequency, but not wheeled duration; male rank correlations were not significant |
A marked male preference for wheeled objects and greater female variability resembles some human child findings, supporting a possible evolved/hormonal contribution to toy-related activity. Because hormones were not manipulated and monkeys have social/rank histories, results do not eliminate learning/socialisation or show that toys are biologically gendered.
| Strength | Weakness |
|---|---|
| Standardised paired trials, counterbalanced position, fixed separation/time and video enable replication and reduce location/order bias | Field-like troop setting leaves participant/social/rank and concurrent-interaction variables difficult to control; causation is limited |
| Operational checklist plus recorded frequency/duration creates objective quantitative comparison and rechecking | Coding/category meaning can involve observer/construct bias; human-designed wheeled/plush categories may not mean the same to monkeys |
| Non-human comparison reduces direct human gender-toy socialisation as an explanation | Small captive single-troop analysed sample and excluded infants/hormonal cases limit generalisation to wild monkeys, other species and children |
Animal ethics should use replacement/species/number and minimise pain/distress. Here monkeys were captive-bred/socially housed, familiar with enclosure/keepers, had food/water/enrichment and were observed remotely; toys created low intrusion, though novel/broken toys and inability to consent require welfare monitoring and removal. Evaluate animal guidelines, not human informed-consent rules.
Nature–nurture and biological–social interaction are central. The nomothetic sex-category analysis finds group patterns but female variability/rank associations caution against deterministic individual prediction. Application to children's toy access should expand opportunities, not reinforce stereotypes.
The strongest result is male wheeled preference; females did not show a simple strong plush preference. Parallel species patterns support—but do not prove—a biological cause, and frequency, duration, preference and rank correlations must not be treated as the same measure.
Mindfulness trains non-judgemental present-moment attention; earlier cross-sectional work could not show whether meditators' brains changed through practice or differed beforehand. Hölzel et al. aimed to test whether an eight-week Mindfulness-Based Stress Reduction (MBSR) course changed regional brain grey-matter concentration/density, especially in regions linked with learning, memory and regulation.
Longitudinal quasi-experiment with independent MBSR and wait-list control groups and repeated pre/post measures. The final MBSR MRI sample was 16 meditation-naïve adults; a separate control group was scanned across a similar interval. MBSR recruits sought stress reduction and had criteria including age 25–55, physical/psychological health, no medication/recent substantial meditation, MRI safety and commitment to all eight classes/homework.
| Component | Operational detail |
|---|---|
| Training | Standard eight-week MBSR: group instruction and home practice including sitting meditation, mindful yoga and body scan |
| Body scan | Attention guided sequentially through body regions, noticing sensations non-judgementally and ending with awareness of body as a whole |
| Structural measure | High-resolution MRI before/after; voxel-based morphometry compared regional grey-matter concentration within and between groups |
| Self-report | 39-item Five Facet Mindfulness Questionnaire (FFMQ), 1–5 from never/rarely true to very often/always true |
| Five facets | Observing; describing; acting with awareness (conscious current action rather than automatic behaviour); non-judging; non-reactivity |
| Practice record | Participants reported home-practice time, allowing correlation with structural change |
| Finding | Interpretation |
|---|---|
| Pre-post MBSR | Grey-matter concentration increased significantly in left hippocampus and clusters including posterior cingulate cortex, temporo-parietal junction and cerebellum |
| Control comparison | Corresponding change was not found in wait-list controls over the interval |
| Practice/self-report | Structural change was not simply/significantly correlated with reported practice amount; FFMQ supplies subjective mindfulness evidence |
Participation in MBSR was associated with measurable regional structural brain change, supporting biological plasticity and interaction between environmental experience/practice and brain. It does not prove every component, benefit or psychological change was caused by meditation alone.
| Strength | Weakness |
|---|---|
| Pre/post MRI and wait-list comparison provide objective quantitative localisation and stronger change evidence than meditator-only correlation | Self-selected stress-reduction participants, small final sample and attrition/MRI intolerance limit generalisation and may bias groups |
| Standard MRI analysis and quantitative FFMQ enable repeatable comparison/triangulation of structure and reported mindfulness | FFMQ/practice logs risk social desirability, response interpretation and inaccurate adherence; MRI grey matter does not directly show felt benefit/function |
| Longitudinal design establishes temporal order and tests neuroplasticity | No random allocation/active control: expectation, teacher/group contact, yoga, stress change or lifestyle could explain differences; bundled MBSR lacks component specificity |
Consent, confidentiality of scans/self-reports, MRI screening/claustrophobia, withdrawal and fair treatment of controls matter. Ethical applications include offering evidence-informed MBSR in stress-management, school/work/clinical support with trained delivery and choice—not scanning people to claim diagnosis or promising guaranteed brain improvement.
The study supports biological explanation and localisation while directly illustrating nature–nurture interaction/plasticity. MRI is objective/reductionist; lived mindfulness, group context and self-report add holistic/subjective dimensions. Group averages are nomothetic and do not predict every individual's outcome.
MRI measures structural concentration, not thoughts or proof of improved functioning. Association after a bundled non-randomised course supports—but does not isolate—a mindfulness causal mechanism; more grey matter is not universally 'better'.
| Main syllabus assumption | Research implication | Boundary |
|---|---|---|
| Information follows a broadly shared input → process → output route, analogous to a computer | Manipulate/present input, control task conditions and measure response/recall as output to infer attention, memory, perception or thinking | Processing is not directly observed and is influenced by emotion, motivation, biology and context |
| Individuals differ in attention, language, thinking and memory; these processes help explain behaviour and emotion | Compare people/ages/groups and correlate cognitive tests with characteristics | A group mean does not define every person or make a test diagnostic |
Example chain: input (recording, eye photograph or witnessed face) → selective attention/encoding/interpretation/storage/retrieval → output (name/place recall, mental-state word or line-up choice). Operational definitions make the hidden process testable but can capture only part of it.
| Core study | Cognitive process/inference |
|---|---|
| Andrade | Doodling condition changed attention/encoding and later monitored/incidental recall during a dull message |
| Baron-Cohen et al. | Revised Eyes Test operationalised advanced theory of mind/social sensitivity; group/sex/AQ differences show individual variation |
| Pozzulo et al. | Familiarity, age and target presence affected face-memory outputs; familiar cartoon memory helped separate cognitive ability from pressure to choose |
Cognitive research often uses controlled, standardised, quantitative tasks (reliable/causal comparison) but risks reductionism, low ecological/construct validity and cultural/language bias. Individual–situational, idiographic–nomothetic, development, application and ethics should be connected to the exact task/sample.
The shared route does not predict identical processing or output. A laboratory response is evidence about an operational task, not transparent access to a person's mind; infer the construct only as far as controls and validity allow.
Attention selects/focuses limited cognitive resources; boredom/low arousal can prompt daydreaming, while a demanding second task can compete with processing. Andrade asked whether a simple, low-demand doodling task would maintain attention and improve monitored and incidental memory for a monotonous telephone message.
Opportunity sample of 40 paid general-population adults (18–55; 35 women, 5 men) from a research participant panel who had just completed an unrelated experiment and agreed to help for five more minutes. Independent groups: doodling versus control.
| Stage | Standardised detail |
|---|---|
| Setting/input | Visually dull room; participants listened to the same monotonous recorded telephone message about a party |
| Doodling IV | A4 sheet with alternating rows of 1 cm circles/squares (10 per row) and margin; shade shapes without concern for speed/neatness |
| Monitoring task | Both groups wrote down names of people attending; they were told the message was dull and did not need to remember it |
| Delay/tests | After a one-minute conversation, unexpected recall requested party-goer names then places; places measured incidental memory |
| Scoring | Monitoring/name and incidental/place correct responses counted; false alarms were deducted from total memory score; suspicion was checked/debriefed |
| Outcome | Doodling versus control |
|---|---|
| Monitoring names | Doodling group recalled slightly more target names (about 7.8 versus 7.1) |
| Surprise memory | Doodling group recalled more names/places overall (about 7.5 versus 5.8 after scoring), roughly 29% more information |
| Doodling compliance | Participants shaded the provided shapes rather than producing free, meaningful drawings |
Structured doodling did not drain attention enough to impair the message and may have reduced daydreaming/maintained arousal, improving monitoring and incidental recall under boredom. The study tested shape shading during one dull auditory task—not every form of drawing, note-taking, learning or divided attention.
| Strength | Weakness |
|---|---|
| Same tape, room, instructions, paper/task and scoring support reliability/replication and internal comparison | Small opportunity panel, strongly female and just fatigued by another study, limits representativeness and may create unusual attentional state |
| Independent groups avoid practice/carryover; objective correct/false-alarm scores support quantitative analysis | Participant variables may differ between groups; no direct measure of daydreaming/arousal, so proposed mechanism is inferred |
| Surprise incidental test reduces deliberate rehearsal for places | Dull laboratory message/shading has limited ecological/construct validity for real lessons/work and free doodling; demand/suspicion remains |
Low-risk participation still requires consent for added time, withdrawal, debriefing about unexpected memory testing and confidentiality. Ethical application: offer simple repetitive shading during dull listening and compare recall; do not require it for everyone or assume benefit in visually demanding/safety-critical tasks.
The experiment supports bounded structured doodling under low stimulation, not unrestricted drawing or a universal memory technique. Improved output does not prove the exact arousal/daydreaming mechanism without a direct measure.
Theory of mind/social sensitivity is attributing mental states to self/others and using them to explain/predict behaviour. Earlier tests could show ceiling effects or weak discrimination in typical adults. The study revised the Reading the Mind in the Eyes Test and tested whether adults with AS/HFA scored lower than comparison groups, plus predicted sex differences and relationships with Autism Spectrum Quotient (AQ).
| Group | Key sample features/purpose |
|---|---|
| 1: AS/HFA | 15 adult men, mean age about 29.7, diagnosed AS/HFA, typical IQ (mean about 115), volunteer response to National Autistic Society magazine advert |
| 2: general adult comparison | 122 adults from adult education/community/library settings, broad occupations/education; socioeconomic/education comparison with Group 1 |
| 3: student comparison | Cambridge University students, enabling sex/AQ analyses but selective high-education sample |
| 4: IQ-matched comparison | Adults selected to match Group 1's IQ, testing whether group difference reflected general intelligence |
| Measure/control | Procedure/logic |
|---|---|
| Revised Eyes Test | 36 cropped eye-region photographs; choose one of four mental-state words; one correct answer per item, producing quantitative score |
| Glossary | Participants could inspect it, flag unfamiliar words and consult definitions throughout, reducing vocabulary confound |
| Gender-recognition control | Judge sex/gender of the person in each eye photograph, checking basic visual face/eye processing |
| AQ | Self-report autistic-trait score used in relevant groups; relation with Eyes score/sex examined |
| Improvements | More items, four options, balanced target terms, revised stimuli/word checks and control task aimed to reduce ceiling/ambiguity and improve discrimination |
| Result | Bounded inference |
|---|---|
| Group 1 scored lower on revised Eyes Test than comparison/IQ-matched groups while succeeding on gender control | Relative difficulty concerns advanced mental-state inference rather than general IQ/basic gender recognition |
| Typical-group women tended/scored higher than men on Eyes Test; sex comparisons depended on the specific comparison group | Supports individual/group variation, not a deterministic sex rule |
| AQ was higher in AS/HFA and related inversely to Eyes performance in tested comparisons | Convergent relationship between autistic traits and this operational theory-of-mind measure, not diagnosis/causation |
The revised test was more sensitive to individual/group differences in adult social cognition and supported a specific advanced theory-of-mind difficulty in AS/HFA. It does not show that autistic people lack empathy, cannot infer any mental states or can be diagnosed from eyes alone.
| Strength | Weakness |
|---|---|
| Standardised 36-item forced-choice test, glossary and gender control support reliability, replication and internal/construct checks | Static cropped eyes and four supplied words lack everyday multimodal/social context; language/cultural familiarity and guessing affect construct/ecological validity |
| Quantitative scores enable four-group, sex and AQ comparisons; IQ control tests a plausible confound | Scores reduce nuanced mental-state reasoning to totals and correlations do not establish causal cognitive mechanism |
| Revised items reduce original ceiling/ambiguity and can be reused for research | Group 1 was only 15 volunteer men; community/student groups were selective, limiting sex/cultural/clinical generalisation |
The paper supports informed consent/confidentiality and generally low physical risk, but anticipated poor performance, labels and unfamiliar mental-state words could cause psychological discomfort/stigma—especially for Group 1. Minimise harm, explain non-diagnostic interpretation and protect individual scores.
The Eyes Test is an operational psychometric task, not mind-reading or an autism diagnostic. A lower group average describes this task and sample; do not convert it into a deficit claim about every autistic person's empathy or real-life social ability.
A false positive is selecting someone/foil when the target is absent while believing the choice is correct. Young children's errors may reflect weaker face memory (cognitive) or social demand from an adult/authority and the mere presentation of a line-up implying someone must be chosen. The study tested whether age/presence patterns remained when target familiarity was manipulated using well-known cartoons versus unfamiliar humans.
Participants included young children and adults. The young-child group was 59 (38 boys, 21 girls), aged 4–7 (mean about 5), from three private pre-kindergarten/kindergarten schools in Eastern Ontario and familiar with Dora/Diego. Adult volunteers came from an Eastern Ontario university pool. Targets were Dora, Diego and two unfamiliar humans, including a 22-year-old white female student brushing her hair.
Cartoon foils came from internet images judged similar to Dora/Diego by facial structure/hair. Three judges rated ten candidates per target; the four highest-similarity cartoon characters were used. Human foils were also matched to target appearance. Participants watched four target videos in random order on a laptop.
| Condition | Simultaneous photo-array logic | Correct response |
|---|---|---|
| Target present | Target plus three foils, random/counterbalanced target position, plus silhouette option representing 'not present' | Select target |
| Target absent | Target replaced by similar foil in same/counterbalanced position; four foils, none from video, plus silhouette | Select silhouette / reject all faces |
| Outcomes | Correct identification for target-present; correct rejection versus false-positive foil selection for target-absent | Separates recognition from willingness/ability to say absent |
| Finding | Meaning |
|---|---|
| Children correctly identified about 99% of familiar cartoon targets when present | Familiar-face cognitive representation/recognition was strong |
| Adults correctly identified about 66% of unfamiliar human faces in the cited comparison | Face type/familiarity and age/task produce individual/group processing differences |
| For child target-absent trials, correct rejection was higher for cartoon faces (about 74% overall; Dora about 80%) than human faces | Familiarity improved absence decisions, yet errors remained even with strong memory |
| Young children still made target-absent false positives | Supports social demand/pressure to choose as an explanation in addition to cognition |
Young children can recognise familiar targets very accurately, but have difficulty rejecting a line-up when the target is absent. This pattern supports social demand—not faulty memory alone—as a cause of false-positive choosing. It does not mean every child error is social or that target-present-only line-ups are acceptable police practice.
| Strength | Weakness |
|---|---|
| Controlled familiarity, target presence, simultaneous format, foil matching, random order/counterbalancing and standard instructions support internal validity/replication | Cartoon/video/silhouette computer task lacks crime stress, delay, live police context and diverse faces, limiting ecological validity |
| Familiar cartoons create a strong cognitive-memory check; quantitative correct-ID/rejection rates distinguish output types | Familiarity, age, face type and social demand may not be fully separable; forced arrays/authority remain and percentages simplify reasoning/confidence |
| Child/adult comparison addresses developmental individual differences | Young children from three private schools and university-pool adults limit socioeconomic/cultural/general generalisation |
Parental/guardian consent and demographic/cartoon form, low-threat familiar videos and confidentiality support ethics; children may still feel stress/pressure or limited withdrawal with an adult questioner, so assent, reassurance and stopping are needed. Application: explicitly state the culprit may be absent, use child-friendly non-authoritative interviewers, open/non-leading questions, avoid challenging a rejection and monitor distress.
Correct identification (target present) and correct rejection (target absent) are different outcomes. The silhouette offered an absence response but did not eliminate social pressure; familiar cartoons diagnose the explanation, not a literal replacement for real criminal line-ups.
Two linked assumptions organise the approach: experience and environmental interaction shape directly observable behaviour from an initially blank slate; learning occurs through classical conditioning, operant conditioning and social learning, often represented as a stimulus-response relation.
| Mechanism | What changes behaviour? | Diagnostic example | Do not confuse it with |
|---|---|---|---|
| Classical conditioning | Association between stimuli makes a previously neutral stimulus elicit a response | Buttons become linked with fear/disgust after an aversive incident | A consequence delivered after behaviour |
| Operant conditioning | Consequences alter the future probability of a behaviour | Banana after a correct trunk action increases repetition | Positive = morally good; it means something is added |
| Social learning | Attention to a model followed by retention/reproduction; direct reward is not required | Children reproduce a model's aggressive acts after the model leaves | Simple stimulus-response association without representation |
| Core study | Environmental input | Observable output | Best-supported mechanism |
|---|---|---|---|
| Bandura et al. | Aggressive/non-aggressive/no model | Coded imitative and other aggression | Social learning |
| Fagen et al. | Whistle-banana contingencies and successive approximations | Criterion trunk behaviours and completed wash | Operant conditioning |
| Saavedra & Silverman | Aversive history plus exposure/reinforcement treatment | Distress ratings and button contact | Evaluative/classical and operant learning |
To apply the approach, name the mechanism, identify the environmental event, operationalise the observable response, and state how the evidence changes the probability/association/modelled behaviour. A label such as 'nurture' is not enough without this chain.
Blank slate is an explanatory emphasis, not proof that genes, brain processes or cognition never contribute. Similar outcomes can arise by different mechanisms, and a learned behaviour can still involve attention, emotion and biological constraints.
Aim: test whether children imitate an aggressive adult when the model is absent, and whether imitation varies with the model's and child's sex. The sample was 72 Stanford University Nursery School children (36 girls, 36 boys), aged 37-69 months (mean 52 months).
| Design element | Operational detail | Why it matters |
|---|---|---|
| Conditions | Aggressive model, non-aggressive model, or no-model control; male/female models and children | Separates exposure, model sex and child sex |
| Matching | Children were rated for prior aggression and matched across conditions | Reduces pre-existing aggression as an alternative explanation |
| Modelling phase | Adult used a Bobo doll and standardised aggressive acts/phrases or quiet non-aggressive play | Makes behaviour available for direct imitation |
| Aggression arousal | Attractive toys were removed after about two minutes and reserved for other children | Creates comparable frustration before the test |
| Test | Child observed through a one-way mirror with aggressive/non-aggressive toys | Records delayed imitation after the model has left |
Observers coded imitative physical aggression, imitative verbal aggression, mallet aggression, non-imitative physical/verbal aggression and aggressive gun play. Gun play meant shooting darts, aiming the gun at an object or firing imaginary shots. Inter-rater agreement and standard categories support reliability, but codes reduce the context and meaning of play.
| Comparison | Pattern | Inference |
|---|---|---|
| Aggressive vs non-aggressive/control | Aggressive-model children reproduced substantially more modelled aggression | Observation can produce new aggressive responses without direct reinforcement |
| Same-sex vs opposite-sex model | Same-sex imitation was generally stronger | Identification with a model can affect reproduction |
| Boys vs girls | Boys showed more physical aggression; girls were relatively more likely to imitate verbal aggression | Sex differences depend partly on aggression type and model |
| Model absent at test | Distinct acts/phrases appeared after the adult left | Supports retention and delayed reproduction, not immediate copying only |
| Strength | Limitation |
|---|---|
| Standardised model, toys, arousal and coding improve replication and comparison | Bobo-doll play is artificial; ecological validity and transfer to harmful real aggression are uncertain |
| Matched baseline aggression and no-model control improve internal validity | Children from one university nursery limit population/cultural generalisation |
| Quantitative frequencies permit direct condition/sex comparisons | Counts omit motives, emotion and whether behaviour persists |
| Young children may show fewer hypothesis-demand effects | Exposure to aggression, induced frustration, covert observation, weak withdrawal/consent/debrief and no aggression-reduction phase create serious ethical concerns |
Bounded application: reduce children's exposure to aggressive models and show salient, preferably relatable models behaving calmly and tidying/playing cooperatively. This follows modelling evidence; rewarding the child directly would instead be an operant-conditioning intervention.
The study shows short-term reproduction in a controlled play setting, not that every observed aggressive act causes lasting violence. Distinguish imitative responses from other aggression and a result (more acts in a condition) from the conclusion that observation can be sufficient for acquisition.
Aim: evaluate positive-reinforcement training as a welfare-oriented alternative for teaching working elephants in Nepal to cooperate with a trunk wash used in tuberculosis management. Five female elephants took part: four juveniles and one older adult, selected for factors including docility, no current pregnancy/calf, mahout willingness, traditional training history and no prior secondary-positive-reinforcement experience.
| Learning component | Exact role in training |
|---|---|
| Primary reinforcer | Chopped banana directly satisfies a biological need |
| Secondary reinforcer | A whistle, consistently paired with banana, marks the correct response immediately |
| Positive reinforcement | Whistle/banana follow a desired behaviour, increasing its future probability |
| Shaping | Reinforce successive approximations—for 'trunk up', lure with banana near the forehead and progressively require greater height |
| Behavioural chaining | Join component behaviours such as trunk here, trunk up/down, steady, blow and water into the complete wash sequence |
Training was observed against a behavioural checklist and criterion test. 'Trunk here' required the distal/ventral trunk to rest gently on the trainer's palm. 'Steady' required holding the previously requested trunk position while head, ears or feet could move. Trainers used luring and shaping, reinforced correct approximations, and recorded sessions/trials and task difficulty until the full sequence met criterion.
| Outcome | Exact interpretation |
|---|---|
| Four of five elephants passed the full trunk-wash test | Positive-reinforcement training was feasible for most of this small working-elephant sample |
| Juveniles passed within 25-35 sessions; Elephant 2 passed in 25 | Learning speed varied even under the same broad programme |
| The older fifth elephant did not pass | Age/history/individual differences may constrain transfer; failure is evidence, not an outlier to erase |
| Component behaviours improved through successive criteria | Supports shaping/chaining rather than a single one-step reward account |
| Strength | Limitation |
|---|---|
| Operational checklist, task criteria and session counts allow replication and within/individual comparison | Five selected females in one Nepalese working context limit species/population/setting generalisation |
| Quantitative trials/sessions reveal progress and individual differences | No untreated/traditional-training comparison isolates causality; trainer/elephant histories differ |
| Real management task gives high mundane validity and practical welfare value | Observer/trainer expectations and difficulty judgements can affect validity |
| Positive reinforcement, small necessary sample, relevant species, food/water access and ability to walk away reduce harm | Existing chaining to posts and delayed rewards may still create welfare concerns; animal ethics must be evaluated, not assumed |
Application: pair a precise marker with a meaningful reward, define small observable approximations, reinforce only progress toward the target, then chain mastered components. This can reduce coercive handling, but reward value, health, age, prior training and voluntary participation must be monitored.
Positive reinforcement means adding a consequence that increases behaviour—not simply being kind. The whistle is secondary because pairing gives it value; banana is primary. Four successes support feasibility in this setting, not guaranteed effectiveness for every elephant or proof that training alone caused all change.
The participant was a 9-year-old Hispanic-American boy referred by his mother to a Child Anxiety and Phobic Program. He met DSM-IV criteria for specific button phobia, not OCD; the problem began around age five after a bowl of buttons fell on him at school. Aims included examining disgust/evaluative learning and testing imagery exposure plus positive reinforcement.
| Evidence source | Operational detail | Contribution |
|---|---|---|
| Diagnostic interview/history | DSM criteria, onset incident and avoidance | Establishes clinical case and possible acquisition route |
| Feelings Thermometer | 0-8 subjective distress hierarchy for 11 button stimuli (0 none, 8 highest) | Repeated quantitative self-report comparison |
| Fear/disgust reports and observations | Reactions to imagined and real buttons | Tests whether disgust is distinct from fear |
| Post-treatment/follow-up | Diagnostic status, reported distress and everyday button use | Tests maintenance and practical change |
| Stage | What happened | Learning interpretation |
|---|---|---|
| Acquisition/history | Bowl of buttons fell on him; visual/tactile qualities became aversive | Evaluative/classical association may have conditioned fear/disgust |
| Behavioural exposure | Gradual contact with button stimuli was attempted | Extinction/counterlearning opportunity, but some severity ratings increased |
| Imagery exposure | He repeatedly imagined increasingly distressing button scenarios | Controlled exposure targets fear/disgust without immediate full contact |
| Positive reinforcement | Mother rewarded successful completion of gradual exposure | Operant consequence strengthens approach/coping behaviour |
| Follow-up | Minimal distress, no longer met phobia criteria, wore clear plastic buttons on school uniform | Improvement generalised beyond questionnaire scores in this case |
For imagining hugging his mother while her shirt was full of buttons, distress fell from 7 before exposure to 4 midway and 3 after the final imagery session. After treatment he reported minimal distress, no longer met diagnostic criteria and could wear plastic buttons daily. These converging measures support meaningful change, but the combined treatment prevents attribution to one component.
| Strength | Limitation |
|---|---|
| Rich interviews, observation, scales and history provide triangulation and ecological clinical detail | One unusual case cannot represent other ages, genders, cultures or phobias |
| Standardised 0-8 hierarchy and repeated ratings show change over time; follow-up tests maintenance | Ratings are subjective and vulnerable to demand/social desirability; numbers do not make them objective |
| Everyday outcome (wearing buttons) supports practical validity | No control, randomisation or component comparison; maturation, attention and expectancy remain alternatives |
| Parent involvement and graded imagery can be adapted ethically | Deliberately evoking distress requires consent/assent, monitoring and stopping safeguards; behavioural exposure sometimes raised distress |
Ethical application: assess a diagnosed person's individual fear/disgust hierarchy and acquisition history, obtain consent/assent, use carefully graded imagery exposure, monitor distress and pair completed approach steps with appropriate support/reinforcement. Do not prescribe from this case alone or force high-intensity contact.
A falling-button event is a plausible learned origin, not definitive causal proof. Reduced thermometer ratings plus daily button use strengthen the improvement claim, but this single case cannot establish universal effectiveness or show whether imagery exposure, reinforcement, relationship factors or their combination produced the change.
Two assumptions define the approach: behaviour, cognition and emotion are influenced by social contexts, environments and groups; influence can come from the actual, implied or imagined presence of other people.
| Social source | Meaning | Core-study example |
|---|---|---|
| Actual presence | Another person/group is physically present | Milgram's experimenter gives orders; subway passengers witness a collapse |
| Implied presence | Rules, roles or authority carry expected social consequences | Yale setting and experimenter role legitimise continued shocks |
| Imagined presence | A represented other changes judgement without a live person | Perry participants imagine themselves as animated figures approach |
| Social context/group | Situation and surrounding people change response options/costs | Victim condition and nearby bystanders alter helping |
A strong application names the social source, the context/role/group cue, the measured response and the comparison that shows influence. 'People conform because society' is too vague; specify authority pressure, target familiarity, perceived need, group composition or responsibility.
| Study | Social factor | Interacting factor | Bounded lesson |
|---|---|---|---|
| Milgram | Legitimate authority and escalating commitment | Personal conscience/distress | Situation can overpower predicted disposition for many, not all |
| Perry | Social identity of approaching figure | Empathy and oxytocin | Social salience depends on person and target |
| Piliavin | Public emergency and nearby bystanders | Victim condition/race and group composition | Helping reflects perceived costs, need and social expectations |
Social influence is probabilistic, not total control. Biological variables (oxytocin), individual empathy/conscience and prior experience can interact with context; an effect of actual/implied/imagined others does not erase agency or prove one universal social rule.
Aim: measure how far ordinary people obey a legitimate authority when orders conflict with conscience and appear to harm another person, testing situational against dispositional expectations. Forty volunteer men aged 20-50 from New Haven and nearby areas responded to newspaper/direct-mail recruitment, represented varied occupations/education and were paid.
This baseline 1963 study is a controlled laboratory observation/interview with one main condition: obedience was called a dependent variable, but there was no independent variable in this main study. Do not import later variations. In this syllabus version the learner did not give voice feedback.
| Stage | Standardised event | Psychological function |
|---|---|---|
| Role allocation | Both slips said 'teacher'; confederate became learner | Deception fixes participant role |
| Credibility | Learner strapped to chair/electrode; teacher received real 45 V sample | Makes apparatus and possible pain believable |
| Paired-associate task | Teacher read word pairs and punished errors with increasing shock | Creates gradual, rule-bound escalation |
| Generator | 30 switches, 15-450 V in 15 V steps, labels from Slight Shock to Danger/XXX, lights/buzz | Quantifies maximum obedience and intensifies commitment |
| Resistance | Learner pounded wall at 300/315 V then stopped responding; experimenter used ordered prods | Pits conscience/ambiguity against authority |
| End/debrief | Stop after refusal following prods or 450 V; interview, reconciliation/dehoaxing | Collects qualitative response and attempts ethical repair |
| Evidence | Result | Meaning |
|---|---|---|
| Maximum shock | 26/40 (65%) pressed 450 V | Obedience greatly exceeded prior expectations |
| Minimum stopping point | All 40 continued to at least 300 V; 14 eventually refused | Situation was powerful but not irresistible |
| Tension | Sweating, trembling, stuttering, lip-biting, groaning, nail-digging, nervous laughter; one seizure | Obedience coexisted with severe conflict rather than sadistic enjoyment |
| Comments/interviews | Participants voiced concern and denied enjoying harm | Qualitative data expose conscience/responsibility conflict |
| Strength | Limitation |
|---|---|
| Highly standardised apparatus, script, 45 V sample and prods permit replication | No IV means the baseline cannot by itself establish which situational feature caused obedience |
| Behavioural voltage plus observations/interviews provide quantitative and qualitative triangulation | Volunteer male New Haven sample and prestigious Yale context limit population/cultural/setting generalisation |
| Credible setup reveals behaviour unavailable through hypothetical prediction | Laboratory learning/shock task may not represent every real authority relation, despite strong involvement |
| Debrief/reconciliation and follow-up attempted restoration | Deception, lack of informed consent, pressure against withdrawal and intense psychological harm are major ethical costs |
Application: in hospitals or organisations, make individual responsibility explicit, require independent checking, normalise questioning, and provide a clear escalation/withdrawal route when authority orders conflict with safety. This transfers the responsibility and legitimacy mechanisms, not the electric-shock procedure.
A result is '26 reached 450 V'; a conclusion is that legitimate authority and displaced responsibility can lead ordinary people to harm despite distress. Do not say everyone obeyed fully, that participants were sadistic, or that one condition proves authority alone caused each decision.
Personal space is an invisible, context-sensitive boundary used to regulate interaction. Perry et al. tested whether oxytocin (OT) changes preferred interpersonal distance depending on empathy. The recruited sample was 54 male University of Haifa undergraduates, aged 19-32 (mean about 25.3), with normal/corrected vision and no psychiatric/neurological history; empathy grouping used the Interpersonal Reactivity Index (high group around score 40 or ≥0.5 SD above mean).
| Control/design | Exact feature | Purpose |
|---|---|---|
| Repeated measures crossover | OT and saline placebo sessions one week apart | Each participant acts as own control |
| Administration | 24 IU intranasally, three drops per nostril, supervised | Standardises treatment |
| Timing | About 45 minutes before testing | Allows OT level to plateau |
| Double blind/placebo | Participant/administrator condition knowledge controlled | Reduces expectancy and observer effects |
| Individual difference | IRI empathy score/group | Tests interaction rather than universal OT effect |
| Experiment | Task | Operational distance measure |
|---|---|---|
| 1: computerised CID | Imagine self at centre of circular room; friend, stranger, authority or ball approaches from one of eight entrances for up to 3 s | Press spacebar to stop figure, or trial ends at collision |
| 2: room choice | Choose among coloured rooms with chairs/objects at varying spacing, believing a later personal conversation would use a computer-calculated room | Selected chair/interpersonal distance; cover story reduces obvious demand |
| Result | Correct interpretation |
|---|---|
| Friend allowed closer than stranger; stranger needed greatest distance | Imagined social identity affects personal space |
| Under OT, high-empathy participants preferred closer distances than placebo/low-empathy participants | OT's effect depends on empathy; it is not a universal closeness hormone |
| Low-empathy trend could move in the opposite direction | Social-salience effects differ by person/context |
| Experiment 2 showed the same conditional closer-distance pattern; average object angle was unaffected | Convergence across different indirect tasks supports construct interpretation |
| Strength | Limitation |
|---|---|
| Placebo, double blind, crossover, fixed dose/timing improve internal validity and reliability | Animations and room pictures are indirect/artificial; real invasion may produce different behaviour |
| Two tasks and IRI permit convergent and interaction evidence | Male Israeli student volunteers limit gender/age/cultural generalisation |
| Continuous/choice quantitative data support precise comparison | IRI self-report is subjective and group cut-offs simplify empathy |
| Controlled OT administration and debriefing support safety/ethics | OT risks require screening/consent; Experiment 2 deceived participants about a personal conversation and could cause stress |
Application must preserve interaction: do not administer OT as a generic social aid. In space design or support, consider relationship, empathy, culture and individual preference, and use consensual distance choices rather than assuming one optimal interpersonal distance.
The claim is OT × empathy × social target, not 'OT makes people friendly'. Pressing a spacebar or choosing a room operationalises preferred distance but does not directly measure every real encounter, trust, attraction or empathy itself.
Aim: observe whether real subway passengers help a collapsed stranger and how victim condition/race, model timing/location and group composition affect helping, testing bystander apathy and diffusion of responsibility. It was a covert field experiment/structured observation on New York subway trains with naturally occurring passenger samples.
| Element | Operational detail |
|---|---|
| Team | Four students per trial: two male actors (victim/model) and two female covert observers; teams boarded through different doors |
| Victim | Male aged about 26-35, white or Black, casual clothes; either ill with cane or smelled of alcohol and carried bottle in bag |
| Collapse | Victim stood by a pole in the critical area and collapsed after about 70 seconds |
| Model | White male aged 24-29 in casual clothes; helped after set delay/from set area if no passenger had helped |
| Measures | Latency/frequency and identity/race/sex of helpers, number in critical/adjacent areas, movement and spontaneous comments |
| Comparison | Result | Interpretation |
|---|---|---|
| Ill/cane vs drunk | Ill victim received spontaneous help on 62/65 trials (about 95%); drunk victim on 19/38 (50%) and more slowly | Perceived need/deservingness and possible cost/risk strongly shape helping |
| Group composition | More males in the critical area predicted faster help | No simple diffusion-of-responsibility effect here; trapped context may raise arousal/cost of non-help |
| Race | Limited overall race effect, but same-race helping was stronger especially for the drunk condition | Race interacted with ambiguous/high-cost need; do not claim broad prejudice proof |
| Sex/comments | Men provided most first help; comments such as 'it's for men to help' showed social role expectations | Qualitative evidence explains some observed choices |
The cost-reward account: witnessing need creates arousal; a bystander weighs costs of helping (risk, effort) against costs of not helping (distress, blame). A cane victim appears genuinely ill and low-risk; a drunk victim may seem self-caused or risky. In a subway car, leaving is difficult, so larger groups can increase available helpers rather than diffuse responsibility.
| Strength | Limitation |
|---|---|
| Real passengers/emergency context gives high ecological validity and low demand characteristics | Train composition, crowding, repeated routes and uncontrolled events reduce internal validity/replicability |
| Behavioural latency/frequency plus comments provide quantitative/qualitative triangulation | Passenger demographics were estimated; individuals may appear on more than one trial and cannot be fully sampled |
| Standard victim collapse/model/observer roles support comparison in the field | New York subway and historical context constrain cultural/time generalisation |
| Practical insight into emergency design/helping | No informed consent, deception, covert observation, no practical withdrawal/debrief, and possible distress/privacy harms are substantial |
Application: make emergencies unambiguous, identify a specific helper ('you in the blue coat'), reduce perceived danger and clarify the needed action. This can lower uncertainty/diffused responsibility, but Piliavin itself found plentiful helping—especially for a clearly ill victim—so do not begin from 'bystanders never help'.
Bystander apathy means witnesses are less likely to intervene, not that no one helps. The group result contradicted a simple larger-group diffusion prediction in this setting. Separate victim condition, race, number of males and model intervention; each addresses a different comparison.
| Application move | Required content |
|---|---|
| 1 Evidence | State the exact result/conclusion that matters |
| 2 Mechanism | Explain why that result predicts change in the target setting |
| 3 Action | Specify who changes what, how and where |
| 4 Outcome | Name the behaviour/experience to improve and how it could be checked |
| 5 Boundary/ethics | Restrict transfer to comparable people/tasks and protect consent, welfare and autonomy |
| Core study evidence | Ethical prospective application | Boundary |
|---|---|---|
| Bandura: children imitated modelled aggression | Teachers/media can model and reinforce specific prosocial speech/actions, then observe whether those actions increase | Laboratory imitation does not prove lasting violence; do not expose children to harm |
| Andrade: simple doodling improved monitoring/recall in a monotonous task | Permit low-demand doodling during comparable listening situations and compare attention/recall | Do not assume all drawing or complex tasks help every learner |
| Milgram: legitimate authority/prods produced high obedience | Build clear whistleblowing, challenge and stop-work routes so unsafe authority can be questioned | Use evidence to reduce harmful obedience, not manipulate people into unethical acts |
A retrospective explanation (‘authority may explain obedience’) is not yet an application. The application must add a feasible intervention or policy. Evaluate ecological similarity, effectiveness evidence, cost, unintended effects and whether the action remains ethical.
‘This study helps us understand everyday life’ earns no mechanism or action. A laboratory result can inspire a cautious pilot, not a universal recommendation; an application that reproduces the original harm is not justified.
| Explanation level | Causal focus | Good evidence form |
|---|---|---|
| Individual/dispositional | Relatively stable or person-specific biology, personality, cognition, history, skill or state | Measured participant characteristic predicts different behaviour under comparable conditions |
| Situational | External setting, instructions, authority, victim/model, reward, group or task changes behaviour | Manipulated/contextual condition predicts different behaviour across participants |
| Interaction | A person characteristic changes sensitivity to a situation, or context expresses/suppresses a tendency | Person × condition pattern rather than either side alone |
| Core study | Individual reading | Situational reading |
|---|---|---|
| Dement & Kleitman | Dream content/recall varied by participant | Laboratory sleep, waking schedule and EEG attachment may alter sleep/recall |
| Piliavin | Passengers differed in willingness/experience/position | Ill victim was helped more readily than drunk victim; victim/model/group context altered cost–reward |
| Fagen elephants | Training speed and success varied with age, health and prior capacity; one elephant did not complete | Shaping, banana reinforcement, trainer cues and tourist distraction altered learning |
To argue a side: define the level without circular wording, cite one exact result/procedure, then explain the causal link. Prefer an interaction conclusion when evidence supports both—situations change probabilities, while individuals enter them with different capacities and histories.
‘People were different’ is description, not an individual explanation; identify the characteristic. ‘It happened in a situation’ is tautological; identify the external feature and how it changed behaviour.
| Side | Includes | Evidence needed |
|---|---|---|
| Nature | Genetic inheritance, evolved predisposition, hormones, neural structure/function and maturation | Biological measure, heritability/species/sex pattern or development evidence—with confounds considered |
| Nurture | Learning, modelling, reinforcement, culture, relationships and other post-birth experience | Manipulated/observed experience predicts change while alternatives are controlled |
| Interaction | Biology shapes what can be learned; experience changes expression and sometimes biology | A mechanism linking predisposition and environment |
| Bandura aggression study | Debate reading |
|---|---|
| Nurture evidence | Children exposed to aggressive adult models produced more imitative aggression than non-aggressive/control conditions despite prior aggression matching—direct support for observational learning |
| Possible nature contribution | Boys showed more physical aggression; hormones/evolved tendencies are possible explanations, but sex also covaries with socialisation and the study did not measure genes or testosterone |
| Best conclusion | Model behaviour and model/child sex affected expression; strong evidence for nurture in imitation, not proof that biology is absent or that sex difference is genetic |
Define each side, cite an operational result, then state what the design can actually infer. Experimental exposure can support learning causally; an unmeasured biological explanation remains a plausible interpretation, not a measured finding.
Nature does not mean unchangeable, nurture is not limited to parenting, and a biological difference may itself reflect experience. Avoid declaring ‘both’ without explaining how evidence or mechanism connects them.
| Scientific value | Method requirement |
|---|---|
| Study development before adult patterns exist | Sample exact age/developmental stage; do not treat children as small adults |
| Test child-specific education, health or protection | Use age-appropriate tasks/scales/language and real settings |
| Follow change over time | Longitudinal design distinguishes age/cohort/history carefully |
| Issue | Safeguard / validity consequence |
|---|---|
| Consent/assent | Guardian consent plus the child's informed, voluntary assent; authority must not coerce |
| Withdrawal/debrief | Explain repeatedly in simple language; child may not recognise or exercise the right |
| Protection | Minimise fear, pain, embarrassment, drugs and scanner/task distress; monitor non-verbal distress |
| Comprehension | Pilot instructions/vocabulary and avoid mistaking language or game interpretation for the construct |
| Privacy | Protect school/health data and avoid labels that follow the child |
| Adult core-study procedure changed to children | Problem |
|---|---|
| Hölzel MRI/mindfulness | Keeping still and understanding homework affect scan validity; developing grey matter creates alternative explanation |
| Perry personal-space tasks | Oxytocin risk, many trials and computer/figure interpretation may produce distress or measure game skill |
| Baron-Cohen Eyes Test | Emotion vocabulary, social experience and test anxiety may lower scores without indicating the same theory-of-mind construct |
Parental permission does not remove the child's right to assent, withdraw or be protected. Child data are necessary for child claims but may not generalise to adults; adult tasks require redesign, not simple reuse.
| Decision | Required question |
|---|---|
| Replace | Can non-animal observation, existing data, simulation or ethical human work answer it? |
| Reduce | What is the smallest sample/design that can produce interpretable evidence rather than waste animals? |
| Refine | How can housing, handling, procedure, enrichment, stopping rules and aftercare minimise distress and support natural behaviour? |
| Benefit–harm | Is the scientific value specific, credible and proportionate to welfare cost? |
| Infer | Which species/setting similarity justifies the conclusion, and where must human transfer stop? |
| Core animal study | Scientific value | Welfare/inference boundary |
|---|---|---|
| Hassett rhesus-monkey toy preferences | Observation of sex-linked interaction with wheeled/plush objects in another primate tests whether human toy patterns may not be solely social instruction | Captivity, object familiarity/social context and small species-specific sample matter; monkey preference does not prove human innate gender roles or justify sex-specific marketing |
| Fagen elephant trunk-wash training | Shaping/reinforcement showed a practical, less forceful route for veterinary cooperation and revealed individual learning/health differences | Training must be voluntary/positive and sensitive to illness/age; zoo distractions and individual elephants limit species/general learning claims |
Animal work can permit close observation, controlled exposure and investigation of behaviour that cannot ethically be manipulated in humans. Comparative evidence is strongest when the mechanism is conserved and naturalistic behaviour/welfare are preserved; it is weakest when captivity or training creates the behaviour being generalised.
‘Animals are similar to humans’ is not enough—name the relevant similarity. Ethical approval does not make harm irrelevant, and absence of human participants does not remove consent-like duties: researchers remain responsible for necessity, humane care and stopping avoidable distress.