Studies: Classic study — Edexcel A-Level Psychology

The logic here is that the code a store uses shows up in the errors people make, so lists that sound alike are set against lists that mean the same. Volunteers from the Applied Psychology Unit panel in Cambridge were split into groups, each learning one ten word sequence across repeated trials separated by an interference task of copying digits, then recalling the sequence in order after a filled delay of about fifteen minutes. Words that sounded alike hurt the early learning trials but no longer mattered by the delayed test, whereas words sharing meaning were recalled markedly worse than their unrelated control at that delay. Read alongside the companion experiment on immediate recall, it separates an acoustic store from a meaning based one.

This is a quasi-experiment, because nobody assigns a person to the murderer group. Defendants charged with murder who had pleaded not guilty by reason of insanity were compared with controls matched for age and sex, with the participants who had schizophrenia matched against patients with schizophrenia, and all were scanned while working at a continuous performance task so that every brain was doing something comparable. Glucose uptake was lower in prefrontal cortex and corpus callosum in the offender group, and the balance between the hemispheres was abnormal in the amygdala, thalamus and medial temporal lobe, with less activity on the left and more on the right. Structures with no predicted role, such as the caudate and the cerebellum, showed no group difference, which is what makes the pattern look specific.

Watson and Rayner set out to show that a fear response could be built into a healthy infant by pairing a harmless object with something already frightening. Albert B, about eleven months old, showed no fear at baseline of a white rat, a rabbit, a dog, a monkey, masks or burning newspaper, so the researchers struck a steel bar behind his head each time he reached for the rat, seven pairings across two sessions a week apart. The rat alone then made him cry and crawl away, and the reaction spread to a rabbit, a dog, cotton wool, a sealskin coat and a Santa Claus mask. Markers want the conditioning terms mapped onto the procedure, the generalisation and persistence data quoted, and a judgement on what one unusually placid infant, never deconditioned, can prove.

Eight sane confederates got themselves admitted to twelve American hospitals by reporting a voice saying empty, hollow and thud, then behaved entirely normally once inside; all but one were admitted with a diagnosis of schizophrenia, stays ran from seven to fifty two days, and every discharge carried the label schizophrenia in remission. Staff reread ordinary behaviour through the label, recording note taking as writing behaviour, and answered direct questions by moving on with the head averted far more often than by stopping to talk. A second study in which a hospital was warned to expect pseudopatients, and none was sent, left dozens of genuine patients suspected of faking. The teaching point is the validity of diagnosis and the power of a label, not the honesty of psychiatrists.

Two laboratory experiments using students who watched film clips of traffic accidents. In the first, the verb in the critical question changed the speed estimate: those asked how fast the cars were going when they smashed into each other averaged about forty one miles per hour against roughly thirty two for contacted, with hit, bumped and collided in between. The second separated response bias from genuine memory change by waiting a week and then asking whether any broken glass had been seen. There was none, yet more than twice as many participants in the smashed condition as in the hit or control conditions said that there was. The conclusion drawn was that information supplied after an event is integrated with the original memory rather than merely biasing the answer a witness gives.

The aim was to find out whether the proportions of secure, avoidant and resistant infants differ between countries, and whether differences within a country are smaller than those between them. The data were secondary: samples were included only if they used the standard procedure with infants under two years and the original three-way classification, giving thirty-two samples from eight countries and close to two thousand classifications. Secure was the most common outcome in every country; avoidant peaked in West German samples and resistant in Japanese and Israeli ones; and the spread of results between samples from the same country was roughly one and a half times the spread between countries.

Fifteen rats, fine electrodes implanted into the brain under anaesthetic, and a lever that delivered a brief burst of current to whatever site the electrode happened to sit in. The animals then taught themselves. Where the electrode lay in the septal region the rat went back to the corner of the box where current had been delivered and afterwards pressed the lever again and again, in the highest cases hundreds of times within a single hour, and stopped when the current was disconnected. Other placements produced indifference or active avoidance of that corner, and it is this contrast that turns the work into a claim about particular structures rather than a general statement that electricity is rewarding.