Key questions and practical investigation (cognitive psychology)

Chapter Overview: Key Questions and Practical Investigation in Cognitive Psychology

Welcome to this essential guide for Pearson Edexcel AS Psychology (8PS0)! In Topic 2: Cognitive Psychology, you explore how the mind processes, stores, and retrieves information like a complex computer. However, psychology is not just about memorising models in a textbook; it is about putting those concepts into action to tackle real-world problems and conducting your own scientific experiments.

This chapter covers two vital components of your AS course:
1. The Key Question (Section 2.4): How cognitive theories and research help us understand and resolve an issue of contemporary relevance to society.
2. The Practical Investigation (Section 2.5): How to design, conduct, analyse, and evaluate your own laboratory experiment investigating human memory using quantitative data and inferential statistics.

Note: For foundational details on memory architectures (such as the Multi-Store Model and Working Memory Model) or inferential testing mechanics, refer to the companion chapters in this topic. Here, our focus is strictly on their practical application and evaluation.

Part 1: The Key Question in Cognitive Psychology (Specification 2.4)

What is a "Key Question"?

In the Edexcel specification, a Key Question is a contemporary issue of genuine relevance to society today. It is not framed as an abstract academic debate (such as "Is memory reconstructive or multi-store?"). Instead, it addresses an urgent, everyday real-world problem that affects individuals, families, workplaces, or public policy, asking: "How can cognitive psychology help us explain, understand, or manage this issue?"

In your exam, questions on the Key Question assess two core assessment objectives:
• AO1 (Knowledge and Understanding): Describing the issue itself clearly, detailing why it matters to society, who is affected, and what challenges it creates.
• AO2 (Application): Applying cognitive concepts, models, and empirical research to explain the causes of the issue or to propose practical, evidence-based solutions.

Illustrative Key Question 1: How can cognitive psychology help people living with dementia?

Describing the Issue (AO1)

Dementia, including Alzheimer's disease, is a progressive neurodegenerative condition that impacts millions of older adults globally. As societies age, the prevalence of dementia rises, placing immense emotional strain on families and significant financial strain on healthcare systems and social care services. Individuals living with dementia experience severe memory loss, disorientation, difficulty recognising familiar faces or places, and an inability to complete everyday tasks like cooking, managing medication, or dressing. Understanding the specific nature of this memory loss is vital so that carers, clinicians, and families can design supportive environments that preserve independence and dignity for as long as possible.

Applying Cognitive Concepts and Theories (AO2)

Cognitive psychology provides direct insights into what is happening in the brain of someone with dementia and how to help:

1. Multi-Store Model of Memory (Atkinson and Shiffrin, 1968):
Dementia typically attacks the ability to transfer new information from the short-term memory (STM) into the long-term memory (LTM). In early-to-mid stages of Alzheimer's, older autobiographical memories stored decades ago remain intact, while recent events (such as what was eaten for breakfast ten minutes ago) are rapidly lost because encoding and consolidation fail. Application: Carers can use written memory prompts, digital clocks with day-and-night displays, and clear daily visual timetables to bypass faulty STM encoding.

2. Working Memory Model (Baddeley and Hitch, 1974):
The central executive coordinates attention and manages processing across the phonological loop and visuospatial sketchpad. In dementia, the central executive suffers pronounced impairment. Patients struggle with divided attention, complex multi-step instructions, and dual-task activities. Application: Caregivers should communicate using short, direct, single-clause sentences, minimizing background distractions (like a television playing) to avoid overloading the limited capacity of the patient's working memory.

3. Episodic vs Semantic Memory (Tulving, 1972):
Tulving distinguished between episodic memory (personal, time-stamped events) and semantic memory (general knowledge and facts). In Alzheimer's disease, episodic memory is often damaged first—patients forget recent family visits or personal experiences. Semantic memory often deteriorates later, though naming objects (anomia) eventually becomes difficult. Application: "Validation therapy" and reminiscence work use surviving episodic memories from childhood or youth (sparked by old photographs or songs) to reduce anxiety and promote emotional well-being.

4. Reconstructive Memory and Schema Theory (Bartlett, 1932):
As memory traces fade, individuals rely heavily on existing schemas (mental packages of expectations based on past experience) to reconstruct their reality. A person with dementia may believe they must get up to go to work or collect their children from school because their older, deeply ingrained schemas fill in the blanks of their missing present memories. Application: Rather than aggressively correcting or distressing the patient, carers are trained to understand the schema driving the behaviour, ensuring safety while validating the patient's feelings.

Illustrative Key Question 2: How can understanding working memory help in supporting students with dyslexia?

Describing the Issue (AO1)

Dyslexia is a specific learning difference that primarily affects the skills involved in accurate and fluent reading and spelling. In school environments, children with dyslexia often struggle to follow long verbal instructions, retain arithmetic steps, take notes from a whiteboard, or read continuous text at the same pace as their peers. This can lead to academic underachievement, low academic self-esteem, and school-related anxiety. Understanding the cognitive architecture behind dyslexia allows educators to move away from blaming a child's motivation and instead implement targeted educational interventions.

Applying Cognitive Concepts and Theories (AO2)

1. Working Memory Model — The Phonological Loop (Baddeley and Hitch, 1974):
Cognitive research reveals that individuals with dyslexia frequently have reduced capacity or processing speed in the phonological loop, which comprises the phonological store ("inner ear") and the articulatory process ("inner voice"). Because verbal information decays quickly, dyslexic learners struggle to hold multiple spoken words long enough to translate them into written symbols or process lengthy instructions. Application: Teachers should break spoken instructions into small, manageable chunks, repeat key phrases, and avoid asking learners to listen and write simultaneously.

2. Dual-Coding and the Visuospatial Sketchpad:
While the phonological loop may have deficits, the visuospatial sketchpad is often intact and strong. Application: Teachers can pair verbal explanations with visual diagrams, colour coding, mind maps, and pictorial cues. By distributing information across both slave systems, cognitive overload is reduced.

3. Processing Speed and Memory Span Development:
Research into developmental memory shows that memory span increases across childhood alongside articulation rate. In learners with dyslexia, slower phonological processing creates a bottleneck. Providing assistive technology, such as text-to-speech software and extra time in examinations, compensates directly for this cognitive processing constraint.

How to Structure an Exam Response for a Key Question

When an Edexcel AS exam question asks you to discuss or evaluate a key question, structure your answer carefully:
• Step 1 (Identify & Describe): State the key question clearly. Explain what the contemporary issue is and why it represents a serious concern for society (AO1).
• Step 2 (Connect to Cognitive Concepts): Select 2 or 3 distinct cognitive ideas, models, or research findings (e.g., MSM, WMM, Tulving's episodic/semantic memory, or Bartlett's schemas) (AO2).
• Step 3 (Explain the Mechanism): Show precisely how each cognitive concept explains why the behaviour happens or how it guides a practical, workable intervention (AO2).
• Step 4 (Evaluate / Balance): Where required by command words such as assess or discuss, consider the strengths and limitations of relying solely on cognitive explanations (e.g., cognitive models provide clear practical strategies, but neurodegenerative conditions like Alzheimer's also have underlying biological causes that cognitive strategies cannot fully cure) (AO3).

Quick Summary of Key Questions: A Key Question is always a contemporary societal issue. You earn marks by describing the societal problem clearly (AO1) and applying cognitive concepts and research to explain or solve it (AO2).

Part 2: The Practical Investigation in Cognitive Psychology (Specification 2.5)

Overview of the Practical Requirement

As part of your AS Cognitive Psychology course, you are required to design and conduct a laboratory experiment gathering quantitative data, which you then analyse using descriptive statistics and a non-parametric inferential test of difference (either the Mann-Whitney U test or the Wilcoxon signed-ranks test).
Important Assessment Rule: You do not submit your practical report for coursework marks (there is no coursework component in 8PS0). However, your design choices, procedural decisions, data handling, and evaluative reflections are directly tested across the written examination papers!

1. Formulating Aims and Hypotheses

Every cognitive experiment begins with a clear aim and two operationalised hypotheses:

• The Aim: A broad statement stating what the study intends to investigate (e.g., "To investigate whether acoustic similarity affects immediate word recall from short-term memory" or "To investigate whether contextual cues improve recall from long-term memory").

• The Alternative / Experimental Hypothesis (\(H_1\)): A testable prediction stating that the independent variable (IV) will have a significant effect on the dependent variable (DV).
- Directional (One-Tailed) Hypothesis: Predicts the specific direction of the outcome (e.g., "Participants who learn a word list with contextual category headings will recall significantly more words out of 20 than participants who learn the list without headings."). Use this when prior research or theory suggests a clear direction.
- Non-Directional (Two-Tailed) Hypothesis: Predicts a difference will occur, but does not state which condition will perform better (e.g., "There will be a significant difference in the number of words recalled out of 20 between participants in the acoustically similar condition and participants in the acoustically dissimilar condition."). Use this when previous research is contradictory or exploratory.

• The Null Hypothesis (\(H_0\)): Predicts no effect, stating that any observed difference is due to chance alone (e.g., "There will be no significant difference in the number of words recalled out of 20 between the two conditions; any observed difference is due to chance factors.").

2. Experimental Design Decisions

When designing your cognitive lab experiment, you must select an appropriate experimental design to compare your conditions:

• Independent Groups Design:
Different participants take part in each condition of the IV (e.g., Group A learns acoustically similar words; Group B learns acoustically dissimilar words).
- Advantage: No order effects (fatigue or practice cannot influence the second condition), and demand characteristics are reduced because participants only see one condition.
- Limitation: Participant variables (individual differences in baseline memory ability, verbal IQ, or age) between the two groups can act as confounding variables.
- Control: Use random allocation (e.g., drawing names out of a hat or using a random number generator) to ensure participants have an equal chance of being in either condition, spreading individual differences evenly.

• Repeated Measures Design:
The same participants complete both conditions of the IV (e.g., participants recall words under silent conditions, and then recall a different equivalent list with background noise).
- Advantage: Perfectly controls for participant variables since individual memory capacity remains constant across both conditions; requires fewer total participants.
- Limitation: Prone to order effects (participants may improve on the second trial due to practice, or perform worse due to boredom or fatigue). Also higher risk of demand characteristics.
- Control: Use counterbalancing (the AB/BA technique). Split the sample in half: Half does Condition A first then Condition B; the other half does Condition B first then Condition A. This balances out order effects across the conditions.

• Matched Pairs Design:
Different participants are used in each condition, but they are pre-tested and matched in pairs on key cognitive variables (such as baseline digit span or age). One member of each pair is assigned to Condition A, the other to Condition B.
- Advantage: Combines the benefits of both designs—no order effects, while greatly reducing participant variables.
- Limitation: Time-consuming, difficult to match people perfectly, and if one participant drops out, data for the entire pair is lost.

3. Operationalisation and Control of Variables

To establish a rigorous cause-and-effect relationship in a cognitive laboratory experiment, variables must be strictly controlled:

• Operationalisation: Defining variables in precise, measurable terms. Rather than saying "memory was tested," you state: "The number of words correctly recalled in any order from a 20-word list presented for 2 seconds per word on a computer screen."

• Extraneous Variables: Any variable other than the IV that might affect the DV if not controlled (e.g., background room noise, room lighting, word familiarity, or time of day).

• Confounding Variables: An uncontrolled extraneous variable that has varied systematically with the IV, meaning you cannot tell whether the IV or the confound caused the change in the DV.

• Situational Variables: Features of the experimental environment that could affect performance. Controlled through standardisation (keeping all instructions, exposure times, delays, and testing environments identical for every participant).

• Demand Characteristics and Experimenter Effects:
- Demand characteristics: Cues that allow participants to guess the research aim, leading them to alter their natural behaviour. Controlled via standardized written instructions and single-blind procedures.
- Experimenter effects: Unconscious cues from the researcher that influence outcomes. Controlled via standardized scripts and automated computer presentation.

4. Ethical Considerations (BPS Code of Ethics and Conduct, 2009)

In designing your practical, you must demonstrate adherence to core ethical principles:
• Informed Consent: Participants must be told the general nature of the memory task before agreeing to take part.
• Right to Withdraw: Participants must be clearly informed that they can stop the experiment at any point without penalty, and can withdraw their data afterwards.
• Protection from Harm: Memory tasks should not cause excessive cognitive stress, embarrassment, or anxiety.
• Deception and Debriefing: If slight deception was required to avoid demand characteristics, a full debrief must be provided immediately after the experiment, explaining the true aim and ensuring the participant leaves in the same state in which they arrived.

5. Descriptive Data Analysis

Once you collect raw quantitative memory scores, you must summarise them using descriptive statistics:

• Measures of Central Tendency:
- Mean: The arithmetic average (sum of all scores divided by the total number of scores). Highly sensitive because it uses every raw value, but easily distorted by extreme outliers.
- Median: The middle score when all data points are ranked in numerical order. Unaffected by extreme anomalous scores, making it ideal for skewed cognitive data.
- Mode: The most frequently occurring score. Useful for categorical data, but uninformative if there are multiple modes or no repeating scores.

• Measures of Dispersion:
- Range: The difference between the highest and lowest score (often \(+ 1\)). Simple to calculate, but heavily distorted by a single extreme outlier.
- Standard Deviation (SD): A measure of the spread of scores around the arithmetic mean. A low standard deviation shows that participants performed very consistently close to the mean; a high standard deviation shows wide individual variation in memory performance.

• Graphical Displays:
- Bar Charts: Used when the IV represents discrete categories (e.g., Condition A vs Condition B on the x-axis) and the mean or median score is plotted on the y-axis. The bars must have spaces between them.
- Histograms: Used for continuous quantitative data (e.g., recall test scores divided into equal intervals). The bars touch to represent continuous data.
- Distributions: Data may form a symmetrical bell curve (normal distribution where mean, median, and mode coincide) or a skewed distribution (positive skew where scores bunch at the lower end, or negative skew where scores bunch near the maximum possible score due to a ceiling effect).

6. Inferential Statistical Analysis

Descriptive statistics show patterns, but only an inferential statistical test tells you whether the difference between your conditions is statistically significant or merely a result of random chance.

In Topic 2, you must know how to select, interpret, and justify the appropriate non-parametric test of difference:

Criteria for Choosing Your Test:
• Test 1: Mann-Whitney U Test
Used when:
1. The study tests for a difference between two conditions.
2. The experimental design is independent groups (unrelated data).
3. The data is at least ordinal level (ranks, rating scales, or non-normally distributed score data like test recall counts).

• Test 2: Wilcoxon Signed-Ranks Test
Used when:
1. The study tests for a difference between two conditions.
2. The experimental design is repeated measures or matched pairs (related data).
3. The data is at least ordinal level.

Understanding Significance and Critical Value Tables

To decide whether to reject the null hypothesis, follow this universal decision procedure:
1. Identify the Significance Level (\(p\)): In psychology, the standard threshold for accepting an experimental hypothesis is \(p \le .05\). This means there is a \(5\%\) or lower probability that the observed difference occurred by chance.
- For stricter research (e.g., clinical or pharmacological settings), \(p \le .01\) (\(1\%\) chance) is used.
- For exploratory pilot studies, \(p \le .10\) (\(10\%\) chance) may occasionally be inspected.
2. Determine the Calculated (Observed) Value: This is the test statistic generated by running your data through the test formula (e.g., the calculated \(U\) value or calculated \(T\) value).
3. Locate the Critical Value from the Statistical Table: Use the specification's formula and statistical table. You need:
- The chosen significance level (usually \(p \le .05\)).
- Whether your hypothesis was one-tailed (directional) or two-tailed (non-directional).
- The sample sizes: \(n_1\) and \(n_2\) for Mann-Whitney U, or \(N\) (number of pairs excluding zero differences) for Wilcoxon.
4. Apply the Decision Rule:
Crucial Exam Convention: For both the Mann-Whitney U test and the Wilcoxon signed-ranks test, the calculated value must be LESS THAN OR EQUAL TO the critical value to be statistically significant at the chosen level!
- If Calculated \(\le\) Critical: Reject the null hypothesis (\(H_0\)) and accept the experimental hypothesis (\(H_1\)). The difference is significant.
- If Calculated \(>\) Critical: Retain (accept) the null hypothesis (\(H_0\)). There is no significant difference.

Memory Trick: Tests with a letter "o" in their name (like Wilcoxon) or tests of difference using ranks (like Mann-Whitney U) require the calculated value to be low—equal to or smaller than the table critical value!

Type I and Type II Errors

When drawing conclusions from inferential testing, two potential errors can occur:
• Type I Error ("False Positive"): Rejecting the null hypothesis when it was actually true. You claim you found a significant memory difference between conditions, but in reality, the difference was just down to chance. This happens if your significance level is set too leniently (e.g., \(p \le .10\)).
• Type II Error ("False Negative"): Retaining the null hypothesis when it was actually false. You claim there was no significant difference, when in reality your IV did have a genuine effect. This happens when the significance level is set too stringently (e.g., \(p \le .001\)) or the sample size is too small.

7. Evaluating and Refining Your Practical Investigation

In exam questions assessing AO3 (develop, refine, and evaluate practical designs), you may be asked to critique your practical or suggest methodological improvements. Always evaluate using the core psychological concepts:

• Reliability:
- Internal reliability: The consistency of the procedure. Did all participants see words for the exact same duration (e.g., 2 seconds)? Was the delay before recall identical? Standardised lab procedures ensure high replicability.
- How to improve: Use automated software to control word presentation and record responses to eliminate human timing errors.

• Validity:
- Internal validity: Did the IV genuinely cause the change in the DV, or did confounding variables interfere? For example, were word lists matched for syllable length and frequency in everyday language? If not, word difficulty was an extraneous variable.
- Ecological validity: Can these findings be generalised to real-world memory? Recalling artificial lists of unrelated monosyllabic nouns in a silent room lacks mundane realism, meaning ecological validity may be low.
- Predictive validity: Can performance on this laboratory recall task accurately predict how well a student remembers study material for an exam?

• Objectivity vs Subjectivity:
Quantitative memory scores (e.g., number of correctly spelled words recalled) are highly objective and free from personal interpretation. However, deciding whether an incorrect spelling or a close synonym counts as a correct recall can introduce subjectivity unless clear, operationalised scoring criteria are fixed in advance.

• Generalisability:
Consider your sample. Most student practicals rely on an opportunity sample of sixth-form peers. This creates a sample biased by age, educational level, and cognitive background, limiting generalisability to the wider population.

Summary and Quick Review Checklist

Before sitting your Paper 1 examination, make sure you can confidently answer each of the following practical and key question prompts:

1. Key Question: Can you state a clear societal key question for cognitive psychology, explain the problem (AO1), and apply at least two distinct cognitive models/theories to explain or solve it (AO2)?
2. Variables & Hypotheses: Can you fully operationalise an IV and DV, and write out both a directional/non-directional hypothesis and a null hypothesis?
3. Design Choices: Can you justify using an independent groups design (with random allocation) versus a repeated measures design (with counterbalancing)?
4. Descriptive Stats: Can you explain when to use the mean versus median, and describe what standard deviation tells you about memory performance?
5. Statistical Decision: Can you identify whether to use Mann-Whitney U or Wilcoxon, look up the critical value, and apply the rule that calculated must be \(\le\) critical to reject the null hypothesis at \(p \le .05\)?
6. Refinements: Can you identify two specific weaknesses in your practical investigation (e.g., confounding word familiarity, order effects, ceiling effects) and explain precisely how to modify the procedure to fix them?