Studies: One contemporary study from the following — Edexcel A-Level Psychology

Known as the BBC prison study, this is an experimental case study rather than a true experiment. Fifteen heavily screened British men lived in a purpose built institution for eight days, five randomly made guards and ten prisoners, filmed throughout, with daily psychometric measures of social identification and depression alongside saliva samples for cortisol. Three planned interventions followed in turn. The guards never came to see themselves as a group and shrank from using their power, while the prisoners identified more strongly once promotion was closed off, and on the sixth day they broke the regime. A self governing commune then failed, after which some participants began pressing for a harsh new order and the study was stopped early. Haslam and Reicher read this as evidence that tyranny grows out of failed groups, not out of people sliding automatically into roles.

This is the individual differences answer to prejudice, and its point is methodological. Personality, ideology and prejudice are normally all measured by the same person on the same afternoon, so shared method variance and self presentation could be manufacturing the correlation. Cohrs and colleagues therefore had German adults complete measures of the big five traits, right wing authoritarianism, social dominance orientation and generalised prejudice, and had people who knew them rate the same qualities. Associations were weaker in the peer data but survived: low openness to experience predicted authoritarianism, low agreeableness predicted social dominance, and both ideological attitudes in turn predicted prejudice, sitting between personality and prejudice rather than beside them. That supports a dual process model of how disposition feeds intergroup hostility.

The question behind this comparison is which tissue a person needs in order to know what words and objects mean, as opposed to recording new events. Henry Molaison was tested against patients whose damage was limited to the hippocampal formation, patients with wider medial temporal damage, and patients left with lateral temporal damage after herpes encephalitis, along with healthy controls matched for age and education. Nine tasks probed stored knowledge, among them naming line drawings, defining words, sorting items and answering questions about the properties of animals and objects. Patients whose damage stopped at the hippocampal formation performed like controls, while Henry Molaison and those with the most lateral damage gave thin, imprecise definitions and scored far below them.

This is schema research moved out of the artificial laboratory set up and into scenes people actually know. Participants recruited online viewed photographs of everyday settings such as a kitchen, an office, a hotel room and a city street for either a very brief or a longer exposure, then typed out every object they could remember, while a separate group rated how likely each object was to appear in that kind of room. Objects people expected in a scene were the ones most often recalled, longer viewing produced more objects, and wrong reports were rare but almost always things that would have belonged there. The authors model recall as combining prior expectation with noisy memory for detail, so expectation mostly improves accuracy rather than corrupting it.

Here the phonological loop is measured directly, and across ages, in a language other than English. Around five hundred and seventy schoolchildren and adolescents in Madrid, from about five years old to the late teens, were given a standard forwards digit span task under the same conditions, and the spans of each age band were compared both with one another and with published figures from English speaking samples. Span grew with age, rising steeply through the primary years and flattening out in adolescence, and at every age the Spanish spans sat below the English speaking norms. The explanation offered is the word length effect: Spanish number words take longer to say, so fewer of them fit into the brief rehearsal window the loop allows.

A functional imaging study about craving rather than about being high. Male heroin users recruited through treatment and healthy controls matched for age and sex were scanned twice, once lying still with nothing to do and once while heroin-related pictures alternated with neutral ones. The region singled out in advance was the posterior cingulate cortex, a hub of the default mode network that deals with self-referential thought. At rest the users showed abnormal posterior cingulate function relative to controls, and during the drug-cue blocks the same region responded more strongly in users, with the size of the response tracking how much craving they reported. The argument is that drug cues pull on a self-referential system and not only on reward circuitry, which helps explain why a place or a syringe can reinstate use long after withdrawal is over.

A twin design used to ask whether two kinds of aggression come from the same place. Pairs of six-year-olds from a Canadian birth cohort were rated by their teachers and nominated by their classmates, once for hitting, pushing and fighting and once for gossiping, excluding and spreading rumours, and the identical and non-identical correlations were fitted to models estimating genetic, shared environmental and non-shared environmental contributions. Physical aggression behaved as the genetically influenced trait, with roughly half the variation attributed to genes and the remainder to experiences the twins did not share. Manipulating relationships behaved as the learned one, carried by environment with no reliable genetic contribution, although children who hit were often also the children who excluded, and that overlap was partly genetic.

Animal work that asks what a drug cue does to a synapse. Rats learned to press a lever for heroin while a light and tone signalled delivery, had that responding extinguished, and were later given the cue alone to provoke relapse-like drug seeking. Within about an hour of cue exposure, synapses in the medial prefrontal cortex weakened, and the mechanism was the removal of one class of AMPA receptor from the synaptic membrane, a depression-like change rather than a strengthening one. When that removal was blocked by infusing a peptide into the region, the animals did not resume lever pressing, which is what converts an association into a causal claim. The reading for human addiction is that relapse includes a fast local synaptic step, which is in principle a target for medication.

The arrival of television in Nadroga, Fiji, was used as a natural experiment on whether Western media changes eating attitudes in a culture that had traditionally admired a robust body and treated weight loss as a worry. Two independent samples of ethnic Fijian schoolgirls, one tested soon after broadcasting began and one three years later, completed the Eating Attitudes Test, and a subsample gave narrative interviews. Disordered eating scores roughly doubled, self-induced vomiting to control weight appeared where there had been none, and the interviews had girls describing television characters as models for reshaping their bodies and improving their job prospects. Marks come from tying those numbers to modelling and vicarious reinforcement, then attacking the causal claim.

Bastian et al. (2011) investigated whether playing violent video games diminishes players' perceptions of their own humanity and that of others. In a laboratory experiment, undergraduates were randomly allocated to play either a violent game or a non-violent game in pairs. Afterwards, they rated themselves and their co-player on traits representing human nature (e.g., warmth) and human uniqueness (e.g., rationality). The results showed that playing the violent game significantly reduced ratings of human nature for both the self and the co-player, compared to the non-violent game. However, there was no significant effect on ratings of human uniqueness. A second experiment replicated the self-dehumanisation effect for human nature traits. The study highlights the socially sensitive implications of violent media consumption.

This is the therapy study of the learning options, testing whether systematic desensitisation reduces fear of flying in Spanish adults who answered a media appeal. Everyone was assessed before and after with interviews, fear questionnaires and physiological recordings taken while flight material was presented; the treated group then had roughly a dozen twice-weekly sessions of progressive muscle relaxation, an individually built hierarchy of feared flying scenes and graded exposure in imagination, while the comparison group waited. Self-reported fear fell sharply in the treated group and not in the controls, yet a minority of treated participants did not improve and the physiological measures moved less consistently than the questionnaires. The examiner rewards judging effectiveness against those two caveats rather than announcing a cure.

Bradbury and Williams (2013) conducted an archival, correlational study using records of felony trials from United States jurisdictions to investigate whether the racial makeup of a jury predicted the verdict. They found that a higher proportion of white jurors was associated with a greater likelihood of conviction where the defendant was Black, while no comparable relationship appeared for white defendants. The researchers concluded that jury diversity matters for justice, as juries drawn from a narrow pool may reach systematically different verdicts. The use of real trial outcomes provides high ecological validity but lacks the control of laboratory experiments, shaping evaluation points.

Visitors to a horror attraction wore heart rate monitors and completed a standard state anxiety questionnaire, met a costumed actor in a dark labyrinth, and afterwards described him and tried to pick him out of a video identification parade. Splitting the sample at the median for anxiety, the calmer half identified him correctly about three quarters of the time and the highly anxious half roughly one in six, and the anxious group also produced fewer correct descriptive details and more incorrect ones. Because the fear was genuine and the measures included physiology as well as self report, this is unusually strong evidence that high arousal harms recognition; because the groups were formed by measurement rather than by allocation, it cannot show that anxiety by itself caused the difference.

An evaluation of short cognitive behavioural anger programmes run in Australian prisons, with several hundred male offenders taking the course and a smaller waiting list group for comparison, measured before, after and at follow up on anger knowledge, self reported anger and staff ratings of behaviour. The gains were modest: knowledge about anger improved, but the treated men were not reliably better than the untreated ones on the anger measures themselves. What did predict change was what the offender brought with him, chiefly readiness for treatment, motivation to change and a high level of anger to start with. The practical message is that brief provision is cheap and easy to timetable but is not sufficient on its own, and that selecting participants by readiness may matter more than adding sessions.

The question was whether attachment in one European country follows the international pattern or has a shape of its own, and it was answered by pooling existing evidence rather than by collecting new cases, gathering Italian Strange Situation studies of children and Adult Attachment Interview studies of adults, published and unpublished, and setting their distributions against the worldwide norms. Secure and autonomous classifications were the most common, as they are elsewhere, so the broad picture supports universality, but the Italian samples departed from the norms by carrying a larger share of insecure classifications, with dismissing adults over-represented, and those departures were widest in clinical and at-risk groups rather than in ordinary families.

Children adopted from abroad usually arrive after institutional care, poor nutrition and no stable caregiver, which makes them the natural test of whether very early privation leaves lasting behavioural damage. The researchers followed a group whose health and physical growth had been recorded on arrival and reassessed them at around seven years old, using a standardised behaviour checklist completed by the mothers and a pictorial self-report answered by the children. Most scored within the normal range on the mothers' checklist, and it was the children's own reports of internalising and externalising difficulties that tracked how adverse their circumstances before placement had been, while the mothers' ratings tracked the mothers' own reported stress.

The contemporary option here is a secondary analysis of the American National Institute of Child Health and Human Development cohort, in which around fourteen hundred children were followed from birth and their care was rated by trained observers rather than by parents. The researchers used propensity weighting to compare four groups: children whose care met a high quality threshold only in infancy and toddlerhood, only in the preschool years, in both periods, or in neither. Outcomes at four and a half years were letter and word recognition, applied problems and language. Quality in the preschool period predicted the higher scores, quality confined to the first two years added little on its own, and every effect was modest rather than transformative.

Friends drink alike, and the value of this study is that it refuses to read that similarity as proof of influence. Working from friendship nominations and self-reported drinking collected a year apart from American school students in the Add Health project, Mundt and colleagues used stochastic actor-based simulation to model ties and behaviour as they changed together, so that two processes could be estimated separately: adolescents choosing friends who already drink as they do, and adolescents shifting their own drinking towards that of their friends. Both came out significant, so influence operated on top of the sorting rather than instead of it. The applied message the examiner wants is that a campaign aimed at lone individuals misses the mechanism, because the friendship group is the unit that carries the behaviour.

Epidemiology asks who is drinking and what travels with it, and this survey asked it in one district of northern India rather than in a Western student sample. Interviewers worked through the urban and rural field practice areas attached to a medical college in Aligarh, putting structured questions about alcohol to community members and recording education, occupation, income, family history and the company people drank in. Drinking clustered among men, among the poorer and less educated, and among those whose friends or relatives already drank, with peer company among the reasons most often given for starting, and the rural community returned the higher rate of use. Because the design is cross-sectional, none of this establishes cause, but it maps the social patterning any campaign has to work with.

This option study is a true experiment, as outpatients screening positive for risky drinking were randomly allocated to conditions. Screening used the WHO alcohol use disorders identification test in a South African hospital. Those scoring in the hazardous or harmful band were allocated either to a short motivational counselling session with an information leaflet, or to the leaflet alone, then screened again 3 months later. Scores fell in both groups and fell further in the counselled group, supporting screening and brief advice as something a busy general hospital in a low-resource setting can realistically deliver. The control group also received a leaflet, so the difference understates the intervention's effect.