Cognitive psychology: practice questions with explanations

1What cognitive psychology studies

Cognitive psychology studies the mental processes behind behaviour: perception, attention, memory, language, problem solving and decision making.

Its main idea is information processing. The mind is treated like a system that takes in information, stores it, changes it and uses it. Models are drawn as boxes (stores) and arrows (processes).

You cannot see a thought, so cognitive psychologists measure it indirectly:

• Experiments that measure accuracy and reaction time, often in milliseconds.
• Case studies of patients with brain damage, which show what happens when one process fails. Patient H.M. and Clive Wearing are famous examples.
• Brain imaging. Functional MRI (fMRI) detects changes in blood oxygenation, the BOLD signal. It shows where activity happens well, but it is slow. EEG is fast but shows less precisely where.

2Memory: stores and systems

The multi-store model (Atkinson and Shiffrin, 1968) has three stores:

1. The sensory register holds raw input for a moment. Visual (iconic) memory lasts under a second, as Sperling showed with briefly flashed letter grids.
2. Short-term memory (STM) holds a few items for about 18–30 seconds without rehearsal. Miller estimated its capacity at 7 ± 2 items; grouping items into chunks helps.
3. Long-term memory (LTM) has an almost unlimited capacity. Rehearsal moves information from STM into LTM.

Baddeley and Hitch (1974) replaced the single STM with working memory, a system that holds and works with information:

• The central executive directs attention and controls the other parts.
• The phonological loop holds sounds and words. Repeating a word aloud (articulatory suppression) blocks it.
• The visuospatial sketchpad holds images and locations.
• The episodic buffer, added in 2000, links working memory with LTM and combines information into episodes.

Long-term memory has several kinds. Explicit (declarative) memory is conscious: episodic memory for personal events and semantic memory for facts. Implicit memory works without awareness: procedural skills such as riding a bike, and priming.

Memory is reconstructive. Bartlett showed people change stories to fit their expectations. Loftus and Palmer (1974) found that asking how fast cars “smashed” into each other gave higher speed estimates than “hit”.

We forget through interference and retrieval failure. In proactive interference old learning disrupts new learning; in retroactive interference new learning disrupts old memories.

3Attention, perception and thinking

Attention. Selective attention means focusing on one input and ignoring others, as at a noisy party. In Cherry’s dichotic listening studies, people repeated a message played to one ear and noticed little of the other ear apart from physical changes, such as a voice switching from male to female.

Broadbent proposed an early filter that blocks unattended input; Treisman proposed that it is only weakened, which explains why people still hear their own name. The Stroop effect shows that reading is automatic: naming the ink colour of “RED” printed in blue is slow.

People also miss things in plain view. In inattentional blindness an unexpected object is not noticed; in change blindness a change between two views is not noticed.

Perception. Bottom-up processing builds a percept from the sensory input; top-down processing uses knowledge and expectations. Gestalt principles describe how we group elements, for example by proximity (close together) or similarity.

Thinking. Tversky and Kahneman described heuristics, mental shortcuts that lead to biases:

• Availability: judging how likely something is by how easily examples come to mind.
• Representativeness: judging by how typical something looks, which leads to errors such as the Linda problem.
• Anchoring: estimates stay close to a first number, even an irrelevant one.
• Confirmation bias: looking for information that supports what you already believe.
• Functional fixedness (Duncker’s candle problem): seeing an object only in its usual use.

Language. A phoneme is the smallest unit of sound; a morpheme is the smallest unit of meaning. Damage to Broca’s area mainly affects speech production; damage to Wernicke’s area mainly affects understanding.

4How to answer a cognitive psychology question

Many questions describe a task, a finding or a patient. Work through it like this:

1. Name the process: perception, attention, memory, language or thinking.
2. Choose the model that fits, such as the multi-store model or working memory.
3. Find the part of the model or the effect involved, such as the phonological loop or the recency effect.
4. Link it to the classic study that showed it.
5. Rule out the look-alike: proactive or retroactive, episodic or semantic, availability or representativeness.

5A worked example

Work through the study. Try to predict the next step before you open it.

6Common mistakes

• Swapping proactive and retroactive interference. Proactive: old disrupts new. Retroactive: new disrupts old.
• Treating working memory as one store. It has several parts and actively processes information.
• Mixing up episodic memory (events you experienced) and semantic memory (facts).
• Swapping the availability and representativeness heuristics.
• Mixing up Broca’s area (producing speech) and Wernicke’s area (understanding speech).
• Mixing up inattentional blindness (an unexpected object is missed) and change blindness (a change is missed).

7How exam questions look

• Easy (key terms and facts): definitions and classic findings. Miller’s 7 ± 2, the types of long-term memory, what selective attention or confirmation bias means.
• Medium (applying the concepts): a short everyday example or a task. You name the effect, the heuristic or the working memory component involved.
• Hard (cases and analysis): a classic experiment or a patient. You explain what a finding shows, such as Sperling’s partial report, the Linda problem or the dissociations in amnesia.

For hard questions, ask which model the result supports or challenges, and which alternative explanation it rules out.