Biological Psychology: Key Question & Practical Investigation

The Edexcel specification asks you to do more than learn biological psychology — it asks you to use it. Two of the topic's assessed elements do exactly this. The first is a key question: a real-world issue of relevance to society that you analyse using the topic's concepts and theories. The second is a practical investigation: an ethical study you could actually design and run, in which you apply the research methods of the topic to a biological question. This lesson does both. Part A takes the key question "To what extent should biological explanations of aggression reduce a person's criminal responsibility?" and analyses it with the neural, hormonal and genetic concepts you have met. Part B designs an ethical correlational study relating a biological/trait measure of aggression to a behavioural measure, setting out the hypothesis, co-variables, design, sampling, ethics and analysis — including a named descriptive statistic and a named inferential test (Spearman's rho). Together these show how the topic connects to society and to real research practice.

Key Definition: A key question in Edexcel Psychology is a question of relevance to today's society that can be explained and evaluated using the concepts, theories and research of the topic — bridging academic content and real-world application.

---

Spec Mapping

This lesson addresses two Edexcel 9PS0 — Paper 1, Topic 3: Biological Psychology requirements. First, the key question: an issue of relevance to society analysed using biological-psychology concepts (here, whether biological explanations of aggression should reduce criminal responsibility). Second, the practical investigation: a study applying the topic's research methods, using correlational methodology with an ethical biological/trait-versus-behaviour design, and appropriate descriptive (mean, scattergram, correlation coefficient) and inferential (Spearman's rho) analysis. It draws together the topic's neural, hormonal and genetic explanations of aggression, and the research-methods content (correlations, sampling, ethics, statistics). In assessment-objective terms, you should be able to describe the key question and the practical design (AO1), apply biological concepts to the question and methods concepts to the study (AO2), and evaluate the strength of the biological case and the merits and limitations of the practical design (AO3).

Connects to…

• Classic and contemporary studies (this topic): Raine et al. (1997) and Brendgen et al. (2005) are the primary evidence base for the key question — the neural correlates of violence and the heritability of aggression are exactly what a "reduced responsibility" argument would cite, so this lesson applies both.
• Hormones, genetics and evolution of aggression (this topic): the concepts used to analyse the key question (prefrontal control, serotonin, testosterone, the MAOA "warrior gene", heritability) come directly from these lessons, and the practical's trait/biological co-variable is drawn from the same territory.
• Research methods (across the course): the practical investigation applies correlational design, sampling, operationalisation, ethics and inferential testing, so this lesson is the natural place to consolidate those methods in a biological context.

---

Part A — Key Question

The Question

To what extent should biological explanations of aggression reduce a person's criminal responsibility?

Why this is a question of relevance to society

Courts routinely confront defendants whose violence is linked to a biological factor — a brain injury, a tumour pressing on the frontal lobes, a genetic profile, or abnormal neurochemistry. As brain imaging and behavioural genetics advance, defence lawyers increasingly present neuroscientific and genetic evidence to argue for diminished responsibility or mitigation in sentencing, a practice at the heart of the emerging field of neurolaw. If aggression is partly written into a person's biology, is it fair to hold them fully responsible? Yet if biology excuses violence, does that endanger the public and erode the principle that people are accountable for their actions? This tension — between compassion and explanation on one side and protection and accountability on the other — makes the question socially urgent, with direct consequences for justice, public safety and how society treats offenders.

Analysing the Question with Biological-Psychology Concepts

The case that biology should reduce responsibility

Several strands of the topic can be marshalled in support.

Neural evidence — the under-regulated brain. Raine, Buchsbaum and LaCasse (1997) found that NGRI murderers had reduced prefrontal glucose metabolism together with abnormal amygdala and hippocampal activity. Because the prefrontal cortex governs impulse control and the regulation of limbic emotion, reduced prefrontal function may genuinely impair a person's capacity to inhibit an aggressive impulse. The Phineas Gage case and the well-known instances of tumour-driven violence (where removal of a frontal tumour ends the aggression) reinforce the point: if the neural machinery of self-control is damaged, the voluntariness on which criminal responsibility depends may be compromised.

Hormonal and neurochemical evidence. Low serotonin activity is associated with impulsive aggression, and elevated testosterone correlates with dominance and aggressive behaviour. To the extent that these are not chosen, a defendant might argue their aggression was biologically "pushed" rather than freely willed.

Genetic and evolutionary evidence. Twin studies (including Brendgen et al., 2005, for physical aggression) show a substantial heritable component, and the MAOA gene variant — dubbed the "warrior gene" — has been linked to antisocial behaviour, famously in the Brunner (1993) family and in Caspi et al. (2002). A person does not choose their genotype, which appears to strengthen the diminished-responsibility argument. Evolutionary accounts add that aggression may be an adaptive disposition selected because it once secured resources, status and mates — arguably "natural" rather than freely chosen.

Taken together, these strands support the position that biological factors can reduce culpability, because criminal responsibility traditionally rests on the assumption that the person could have acted otherwise — an assumption that biological impairment of self-control appears to weaken.

The case that biology should not (fully) reduce responsibility

The topic supplies equally strong counter-arguments.

Correlation is not causation, and biology is not destiny. Raine's study is correlational: it identifies brain correlates of extreme violence, not a cause, and Raine himself cautioned that the findings do not show murderers cannot control their behaviour. The brain differences might even be a consequence of a violent, traumatic life rather than its cause. A biological predisposition is not a biological compulsion.

Gene–environment interaction undermines pure determinism. Caspi et al. (2002) showed that the MAOA variant predicted antisocial behaviour only in men who had been maltreated — biology expressed itself through environment. Brendgen showed that social aggression is largely learned and therefore modifiable. If aggression is shaped by experience and can be changed, then people retain meaningful agency, and full exculpation is unwarranted.

Reductionism and the free-will/determinism debate. Reducing a complex, situated act of violence to "a low-serotonin, small-prefrontal brain" is biologically reductionist: it ignores the psychological (anger, intent, provocation) and social (culture, upbringing) levels at which the behaviour also operates. The legal system presupposes a degree of free will; abandoning it entirely for biological determinism would make the whole notion of responsibility incoherent.

Consequences for society (socially sensitive implications). If biology excused violence, dangerous individuals might avoid accountability, undermining deterrence and public safety. There is also a labelling risk: telling people they are "born aggressive" could become a self-fulfilling prophecy, and using scans or genotypes to judge individuals raises the spectre of "biological pre-crime". Because such conclusions are socially sensitive, the bar for using them to diminish responsibility must be high.

A Reasoned Position

The most defensible answer is "to a limited and case-by-case extent, in mitigation rather than as a blanket excuse." Biological factors can legitimately inform judgements of degree — a person with documented frontal damage and demonstrably impaired impulse control may warrant reduced culpability or treatment-oriented sentencing — but the correlational, probabilistic and interactionist nature of the evidence means biology rarely, if ever, removes responsibility entirely. Because aggression is multiply determined (biological and psychological and social) and because predisposition is not compulsion, the fairest stance treats biological explanations as one factor among several, weighed individually, rather than as an automatic exemption. This mirrors current legal practice, in which neuroscientific evidence is admitted in mitigation but the defendant is generally still held accountable.

---

Part B — Practical Investigation

The specification requires a practical investigation applying the topic's research methods. Aggression cannot be manipulated ethically, so a correlational design is appropriate: we measure a biological/trait co-variable and a behavioural co-variable in each person and test whether they are related. The design below is deliberately ethical — no aggression is provoked, and only validated questionnaire and behavioural-record measures are used.

Research Question and Rationale

Does a trait/biological predisposition to aggression relate to actual aggressive behaviour? Drawing on the neural and hormonal explanations (which predict that a biological predisposition should manifest behaviourally), we relate a validated trait-aggression questionnaire score (a trait/biological-disposition measure) to a behavioural measure of aggression derived from objective records. Both are measured as they naturally occur — nothing is manipulated — so the study is correlational.

Aim

To investigate whether there is a relationship between individuals' trait-aggression scores (measured by the Buss–Perry Aggression Questionnaire) and their frequency of recorded aggressive incidents (a behavioural measure).

Hypotheses

• Alternative (experimental) hypothesis (H₁): There is a positive correlation between participants' trait-aggression questionnaire scores and their frequency of recorded aggressive incidents. (One-tailed/directional, because the neural and hormonal theories predict the direction: a higher predisposition should be associated with more aggressive behaviour.)
• Null hypothesis (H₀): There is no correlation between participants' trait-aggression questionnaire scores and their frequency of recorded aggressive incidents; any relationship observed is due to chance.

Exam Tip: A correlational hypothesis must state a relationship ("positive/negative correlation between X and Y"), not a difference ("X is higher than Y"). Naming a direction (positive) makes it one-tailed; predicting merely "a correlation" makes it two-tailed.

Co-variables (operationalised)

In a correlation there is no IV or DV; there are two measured co-variables. Careful operationalisation is essential.

Co-variable	How it is operationalised	Level of measurement
Co-variable 1 — trait/biological predisposition to aggression	Total score (29 items, each rated 1–5) on the Buss–Perry Aggression Questionnaire (AQ), a validated self-report trait measure; higher = greater dispositional aggression	Ordinal (treated as such for analysis)
Co-variable 2 — aggressive behaviour	Frequency of recorded aggressive incidents over a fixed 6-month period, taken from anonymised, existing behavioural records (e.g. logged conduct incidents), counted by an independent rater	Ordinal / ranked count

Because both co-variables are best treated as ordinal (questionnaire scores are not a true interval scale, and incident counts are ranked frequencies), the appropriate inferential test is a non-parametric correlation — Spearman's rho (see Analysis).

Design and Procedure

1. Design: a correlational study (no manipulation; both co-variables measured as they occur).
2. Obtain ethical approval and appropriate permissions for access to anonymised behavioural records.
3. Recruit the sample (see Sampling) and obtain informed consent.
4. Measure co-variable 1: each participant completes the Buss–Perry AQ under standardised conditions; compute each participant's total score.
5. Measure co-variable 2: an independent researcher, blind to the questionnaire scores, counts each participant's recorded aggressive incidents over the fixed 6-month window from the anonymised records, using a clear operational definition of "aggressive incident" to ensure inter-rater reliability.
6. Pair the data: each participant now has two scores (trait score, incident frequency).
7. Analyse: plot a scattergram, compute descriptive statistics, and run Spearman's rho (see Analysis).

Keeping the person counting incidents blind to the questionnaire scores controls researcher/expectancy bias, and the fixed 6-month window and operational definition of an "incident" support standardisation and reliability.

Sampling

• Target population: adults (e.g. 18+) within a defined setting from which anonymised behavioural records are ethically available.
• Sampling method: opportunity or volunteer sampling for practicality, with the well-known limitation that neither is fully representative (opportunity samples reflect who is available; volunteers may be atypically cooperative). A stratified approach would improve representativeness if resources allowed.
• Sample size: at least N = 20–30 paired scores, so that a correlation coefficient is reasonably stable and the critical-value table for Spearman's rho can be used.

Ethics

Because aggression is a sensitive topic, ethics are central to the design.

Ethical issue	How it is managed
No harm / no provocation	Aggression is never induced; only existing behaviour (records) and a self-report questionnaire are used, so no participant is put at risk or distressed by being provoked.
Informed consent	Participants are told the study concerns "attitudes and behaviour", consent to completing the AQ, and consent to the use of their anonymised incident records.
Confidentiality & data protection	Records are anonymised and scores stored against a code, not a name; results are reported only at group level.
Right to withdraw	Participants may withdraw themselves and their data at any point.
Social sensitivity	Findings are framed carefully to avoid labelling individuals as "aggressive"; the study reports a correlation, not a diagnosis, and cannot be used to judge any single person.
Debrief	Participants are fully debriefed, told the true aim, and offered the opportunity to ask questions.

Analysis

Descriptive statistics

• Measure of central tendency: report the mean (and, given the ordinal, potentially skewed data, the median) trait-aggression score and the mean/median number of incidents, to summarise the sample.
• Measure of dispersion: report the range or standard deviation to show the spread of scores.
• Graph: draw a scattergram, plotting each participant as one point (trait score on the x-axis, incident frequency on the y-axis). The pattern of points shows the direction (positive/negative) and strength of any correlation; a clear upward trend would support H₁. (A scattergram is the correct graph for correlational data — never a bar chart.)

Inferential statistic

The appropriate inferential test is Spearman's rank-order correlation coefficient (Spearman's rho, $\rho$ρ ). It is chosen because:

• the study tests a correlation (a relationship between two co-variables), not a difference;
• the data are ordinal (questionnaire scores and ranked incident frequencies), so a non-parametric test is required rather than the parametric Pearson's r;
• the data are related/paired (two scores from each participant).

Spearman's rho ranks each co-variable and computes the correlation between the ranks. It yields a coefficient between $-1$−1 and $+1$+1:

• $\rho \approx +1$ρ≈+1 = strong positive correlation (as trait aggression rises, incidents rise);
• $\rho \approx -1$ρ≈−1 = strong negative correlation;
• $\rho \approx 0$ρ≈0 = no correlation.

The observed coefficient is found using:

$\rho = 1 - \frac{6 \sum d^{2}}{n(n^{2} - 1)}$ρ=1−n(n2−1)6∑d2​

where $d$d is the difference between the two ranks for each participant and $n$n is the number of participants.

Significance