Key studies and key terms

External approval versus acceptance consistent with personal values

Public agreement alone does not identify private acceptance.

Social acceptance versus evidence about reality

Both motives can contribute in the same situation.

Authority ↔ participant. Participant ↔ learner.

A change in the authority’s distance and a change in the learner’s distance are separate manipulations.

A feature present in one baseline does not establish its effect.

A controlled comparison must vary that feature. Field-request compliance and harmful-task continuation are different outcomes.

Milgram’s agentic account: carrying out an authority’s wishes with reduced felt responsibility.

Autonomous: consciously directing one’s own behaviour. Following an instruction alone does not identify the state; distress can coexist with obedience.

A recognised right to direct, based on a role or position.

Recognition of authority and felt personal responsibility are separate questions. A clothing effect alone does not measure the proposed psychological mechanism.

Submission: deference. Aggression: punitive hostility. Conventionalism: rigid values.

Submission concerns idealised ingroup authorities; aggression concerns perceived conventional-value violators. These constructs do not provide a clinical diagnosis.

Attitude response ≠ observed task behaviour ≠ demonstrated cause.

Identify how participants were selected, what was measured and when. Later questionnaires and selected groups do not randomly assign personality.

Asch (1955): a correct partner reduced errors. Milgram (1965): after peers refused, most participants defied.

These are different research tasks. Reduced errors or refusal do not directly measure a change in control expectations.

Internal: personal action. External: chance or powerful others.

These are positions on a continuum. Specific expectations and valued outcomes matter; a single refusal does not identify a fixed type.

Consistency: coherence. Commitment: investment. Flexibility: openness.

A central position can be maintained while other arguments are acknowledged. Personal sacrifice does not by itself establish that the position is correct.

Public matching ≠ private adoption ≠ sustained social change.

Measure each outcome and its timescale. A brief individual answer does not establish lasting shared changes.

Encoding: form; capacity: amount; duration: persistence

Keep the property measured separate from the task used to measure it.

Copy into longer-term storage; access retained information

Transfer does not require erasing the earlier copy.

Focus, divide and switch attention

An attentional controller rather than a dedicated content store.

Limited temporary store for integrated multidimensional information

Links subsystems, perception and long-term information.

Earlier → newer target: proactive; later → earlier target: retroactive

Identify the tested target before naming the effect.

Reduced recall need not mean erased storage

Shared cues with changed responses can create associative competition.

In Godden and Baddeley (1975), word recall was better when learning and test environments matched

Matching differs from a general advantage of either location.

A retained target can be temporarily difficult to retrieve

Useful contextual features can support access.

Matching reports may share post-event information

Trace the source of the detail rather than counting agreement alone.

A changed report does not uniquely identify an altered trace

Separate observed errors from explanations of how they arose.

Report remembered partial detail; do not guess; say when you do not know

Changing perspective is not permission to invent unseen content.

Correct, incorrect and confabulated details are distinct

A larger correct count does not itself establish a higher accuracy proportion.

Early encounter → following target

Lorenz (1937): bird following could be directed towards a human encountered early. Similar outward behaviour does not establish identical human attachment mechanisms.

Contact preference can differ from feeding source

Harlow (1958): cloth contact despite wire feeding challenges a feeding-only account in the surrogate study. Animal welfare and human generalisation need separate evaluation.

Food: primary; caregiver: acquired secondary value

In the feeding account discussed by Bowlby (1958), food is unconditioned and a paired, initially neutral caregiver becomes conditioned. Operant reinforcement changes later responding through consequences.

Monotropy and working models

Bowlby (1958) emphasised a preferred protective figure; Ainsworth (1979) described expectations of availability. APA distinguishes nonrecurring critical opportunities from sensitive periods with harder later acquisition.

Available caregiver → supported exploration

A secure reunion includes seeking or accepting caregiver contact. Separation crying alone does not establish a classification.

Avoidance ≠ seeking contact while resisting it

Avoidant and resistant patterns describe different organisations of behaviour. Neither is a psychiatric diagnosis or proof of no bond.

Country average ≠ every local sample

In van IJzendoorn and Kroonenberg (1988), average within-country variation exceeded average between-country variation. This is an average comparison, not a rule for every pair.

Same procedure ≠ proven same meaning

Standardisation helps comparison. Equivalent meaning and representative sampling still need evidence; national labels are not genetic explanations.

Time apart ≠ insufficient emotional care

A familiar responsive substitute can sustain affectionate care. Biological parenthood and physical co-residence do not alone measure care quality.

Retrospective association ≠ isolated cause

Interpret proposed risks alongside confounding, care quality, recovery and individual variation. A group pattern is not an individual prognosis.

One outcome can recover while another remains elevated

Sonuga-Barke et al. (2017): substantial cognitive recovery coexisted with elevated autism-spectrum symptoms, disinhibited social engagement and inattention/overactivity. Group risks allowed individual variation.

Follow-up ≠ random exposure allocation

The English and Romanian Adoptees project followed Romanian and UK adoptees: a longitudinal natural experiment, not assigned deprivation. Sonuga-Barke et al. (2017) requires considering overlapping timing/duration, prior influences and selective attrition.

Earlier experience → expectations that can be revised

Working models concern self and others. They can inform later relationships without determining a partner’s response. Early influence and later change can coexist.

Adult self-description ≠ direct infant observation

Self-selection and retrospective recall limit representativeness and causal interpretation. Matching category names does not provide an observed infant history.

Frequency / norms / functioning / positive ideals

Identify which perspective supplies a judgment before evaluating its limits.

Specify population, context and decision rule

Rarity is not a value judgment, social conformity is not complete well-being, and practical difficulty does not identify its cause.

Classical cue pairing / operant consequences

In the two-process account, fear can be acquired through pairing and avoidance can be strengthened by relief.

Aversive experience reduced → response strengthened

Negative does not mean that the response is weakened. Persistent avoidance after fear declines limits a complete fear-reduction account.

Systematic desensitisation: relaxation + graded anxiety hierarchy / Flooding: direct high intensity

The procedural contrast does not itself establish comparative efficacy.

Voluntary, informed and with capacity

Compare symptom outcomes and acceptability while keeping valid consent as a distinct requirement.

Self / personal world / future

These are contents of negative views, rather than the A, B and C roles in a different framework.

Activating event / beliefs / emotional and behavioural consequences

Beliefs contribute to responses while events remain part of an interacting account. A rigid demand is different from a flexible preference.

Examine evidence / consider alternatives / practise changes

CBT combines collaborative discussion with applying skills beyond sessions; completing a process is different from measuring its outcome.

Comparator, quality and individual assessment

Adult pooled efficacy informs a question about average benefit. It does not establish equivalence with other treatments, a personal guarantee or suitability without assessment.

Unwanted intrusion → distress → repetitive act or mental ritual

Compulsions may briefly ease distress. Examples illustrate features rather than establish a diagnosis.

Inhibit serotonin reuptake

Treatment benefit does not establish a primary serotonin deficit as the cause of OCD.