3.4 Learning and cognition

A cognitive bias is a systematic tendency in attention, memory or judgment that can move decisions away from a relevant standard. Anchoring bias pulls an estimate toward an initial value; confirmation bias favours information that supports an existing belief or hypothesis.

An anchor can set a starting point from which adjustment is insufficient, even when the number is arbitrary. Confirmation bias can shape which evidence is sought, interpreted and remembered, creating a self-reinforcing decision. Both can be efficient shortcuts in some settings rather than universal irrationality.

A first price of $900 can make$650 seem cheap even when market evidence suggests $500. A buyer who then searches only for positive reviews may combine anchoring with confirmation bias. Using independent benchmarks and deliberately testing disconfirming evidence targets the two mechanisms differently.

A poor decision is not automatically biased, and changing one's mind is not proof that bias disappeared. Identify the initial anchor or prior belief, the information-processing route and the comparison needed to show a systematic effect.

Classical conditioning links a cue with an event so that the cue predicts a response. Operant conditioning changes the future probability of behaviour through its consequences: reinforcement increases it and punishment decreases it.

In operant terms, positive means adding and negative means removing; neither means good or bad. Timing, consistency, schedule and the learner's history determine whether a consequence actually changes behaviour.

If checking a phone removes boredom, the behaviour may increase through negative reinforcement. If a tone repeatedly precedes an unpleasant event, the tone can become a conditioned cue even when the event is no longer present.

Conditioning is not just repetition. Name the stimulus or consequence, its timing and the direction of behavioural change before classifying the process.

To apply operant conditioning, define one observable target behaviour and its baseline, identify the antecedent and current consequence, then deliver an immediate and valued reinforcer only after the target response. Begin with an achievable criterion, record frequency, adjust the reinforcement schedule gradually and check whether the behaviour maintains when rewards become less frequent. Punishment may suppress behaviour without teaching an alternative and can create avoidance or ethical harm.

Dual-process theory distinguishes fast, automatic and associative processing from slower, effortful and rule-based reasoning. Both can be adaptive; reliability depends on expertise, feedback, time and the structure of the task.

Fast processing uses learned patterns efficiently, while slow processing can check a tempting intuition in a novel or high-stakes problem. Switching between them is a control problem, not a moral choice between good and bad thinking.

Recognising a familiar face is fast and usually useful. Checking whether a sale price is really lower than the original requires deliberate calculation because the display can invite a misleading shortcut.

System 1 is not always irrational and System 2 is not always correct. State the trigger, knowledge and opportunity for checking before explaining an error.

A schema is an organised framework built from prior experience that guides attention, interpretation and recall. It helps a person fill gaps and act quickly, but can also introduce consistent distortions.

Ambiguous or incomplete information is especially vulnerable to schema-driven reconstruction. New details may be assimilated to the existing framework, while surprising details can force revision or be remembered less reliably.

After visiting a hospital, a person may recall expected medical equipment that was not present. The insertion reveals a reconstruction mechanism; it does not show that every memory detail is invented.

A schema is not always a conscious stereotype and schema influence is not proof of lying. Specify the prior framework, task and evidence for its effect.

Social learning theory proposes that people learn by observing models in social contexts. Attention, retention, reproduction and motivation mediate whether an observed act is learned and whether it is later performed.

Vicarious reinforcement changes expectations when a model is rewarded or punished. Status, similarity, culture and perceived competence affect which models are noticed and imitated.

A child watches an older sibling praised for helping, remembers the sequence and later reproduces it when able. Learning can precede performance, and a different setting can alter the opportunity to act.

Observed behaviour is not guaranteed to be copied and immediate performance is not the only evidence of learning. Identify the mediational step that links model to action.

Biological factors include genes, brain structures, neural networks, hormones and neurotransmitters that can influence cognition. They provide mechanisms and constraints, not a complete explanation of every thought or behaviour.

Evidence from drugs, lesions, imaging and genetic comparisons answers different causal questions and has different confounds. Biology interacts with learning, language, culture and environment across development.

A medication that changes receptor activity may improve attention for some people, while sleep, stress and task design also change performance. The biological pathway is part of the explanation, not a single switch.

Biological does not mean fixed or inevitable. Avoid one-molecule stories and state what the evidence actually isolates.

For memory, the hippocampus helps bind elements of an experience into a retrievable episodic representation and supports consolidation with wider cortical networks. Damage can impair formation of new episodic memories while leaving some skills or older memories less affected, showing specialization without treating memory as one location. Lesion evidence supports necessity for particular functions but must be interpreted with damage extent, compensation and task demands.

A cognitive model represents stages or components involved in attention, memory, reasoning or decision-making. It is valuable when it predicts an observable pattern, such as interference, capacity limits or selective recall.

The working-memory model separates a control system from specialised verbal and visuospatial processes, but those components are theoretical functions rather than literal boxes. Task design determines which prediction is tested.

Repeating a phone number while reading another number creates verbal interference, whereas a visual route may compete less with spoken digits. The result supports a processing distinction, not a complete map of the brain.

A model is not a diagnosis and a laboratory task is not the whole mind. Link each prediction to the measured behaviour and its boundary.

Cultural schemas are shared frameworks for interpreting people, events and meanings. They can guide attention, encoding and reconstruction, making the same material easier to remember in one cultural context than another.

Schema effects can support efficient understanding within a familiar setting but create bias when a researcher assumes their own categories are universal. Language, experience and task instructions can moderate the effect.

A story organised around a culturally familiar social role may be recalled with more detail by insiders. Comparing groups fairly requires equivalent instructions and attention to the meaning of the material.

A cultural difference in recall is not automatically a capacity difference. Check familiarity, language, motivation and the construct being measured before inferring cognition.

Environmental factors include education, stress, sleep, nutrition, technology, social resources and physical surroundings. They can change cognitive performance by altering practice, attention, physiological load or available strategies.

The pathway needs a defined exposure, timing and outcome. A developmental theory may describe how thinking changes, but the exposure and learning conditions still need separate evidence.

A quiet study space can improve working-memory performance by reducing interruptions. That does not prove a permanent cognitive change; the effect may disappear when the environment changes.

An environmental influence is not a developmental stage or a permanent trait. Specify the exposure, timing, cognitive process and comparison, then distinguish a temporary performance effect from lasting cognitive change.

Cognitive improvement targets a process such as attention, memory strategy, appraisal or problem representation. A useful intervention states the mechanism, practice conditions and outcome that should change.

Worked examples, retrieval practice, feedback and reducing extraneous load can support learning, but gains may remain task-specific. Transfer requires testing a new context, not only repeating the trained exercise.

A learner who practises retrieving ideas from memory may improve delayed recall; a second task checks whether the strategy transfers beyond the original list. Motivation, sleep and prior knowledge also moderate the result.

Better performance on the trained task is not proof of general intelligence improvement. Specify what changed, how it was measured and where the effect should generalise.

A laboratory experiment manipulates an independent variable, measures a dependent variable and controls or randomises other influences. Random assignment helps balance participant differences so causal claims are more defensible.

Standardisation improves reliability, but artificial tasks, demand characteristics and narrow samples can reduce ecological or external validity. Ethics, debriefing and informed consent remain part of the design.

If participants are randomly assigned to two memory instructions and tested with the same list, a difference in scores supports an effect of the instruction under those conditions. It does not automatically generalise to real classrooms.

Control is not the same as realism and a significant result is not a universal law. State the manipulation, comparison, sample and validity boundary.

To conduct the practical: state a directional or non-directional hypothesis; operationalise the independent and dependent variables; decide whether random assignment makes it a true experiment or a pre-existing group makes it quasi-experimental; control or record plausible confounds; standardise instructions; obtain teacher permission, consent and debriefing; pilot the task; collect the required sample ethically; analyse the planned comparison; then report effect, uncertainty, demand characteristics and limits on causality and generalization.