You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
The biological approach argues that behaviour has physical, measurable roots in the brain — but nowhere is that claim more provocative than when it is applied to violent crime. If the brains of murderers differ from the brains of non-murderers, then aggression may be, at least in part, a matter of neural structure and function rather than free choice. The prescribed classic study for Edexcel's Biological Psychology topic tests exactly this idea. Raine, Buchsbaum & LaCasse (1997), in their paper "Brain abnormalities in murderers indicated by positron emission tomography", used PET scanning to compare glucose metabolism in the brains of 41 people accused of murder with a matched control group. Their findings — reduced activity in the prefrontal cortex and abnormal activity in a set of deeper limbic structures — became one of the most cited pieces of evidence in the biological explanation of aggression. This lesson sets out the aim, method, results and conclusions in accurate detail, then evaluates the study methodologically and ethically and locates it within the wider debate about the neural basis of violence.
Key Definition: Positron emission tomography (PET) is a functional brain-scanning technique that measures metabolic activity by tracking a radioactive tracer (here, radioactively labelled glucose), producing colour-coded images of how active different regions are while a task is performed.
This lesson addresses the Edexcel 9PS0 — Paper 1, Topic 3: Biological Psychology requirement to study one classic study in depth: Raine, Buchsbaum & LaCasse (1997), "Brain abnormalities in murderers indicated by positron emission tomography". You must know its aim, method/procedure (the sample of NGRI murderers and matched controls, the PET technique, and the continuous performance task), results/findings (reduced prefrontal glucose metabolism and abnormal asymmetries in the amygdala, hippocampus and thalamus), and conclusions, and you must be able to evaluate it methodologically and ethically. It links the topic's earlier content on localisation of function, scanning techniques and the biological explanation of aggression to a single, assessable investigation. In assessment-objective terms, you should be able to describe the study accurately (AO1), apply it to novel scenarios and to the biological explanation of aggression (AO2), and evaluate its design, sampling, controls, generalisability, causation and ethics (AO3).
Connects to…
By the mid-1990s a number of separate lines of evidence hinted that violent behaviour might have a neural basis. Case studies of frontal-lobe damage (from Phineas Gage onward) linked the prefrontal cortex to impulse control and social judgement; animal studies implicated the amygdala in aggression and fear; and early imaging work suggested that antisocial individuals sometimes showed reduced frontal activity. What was missing was a direct, controlled comparison of the living brains of seriously violent offenders with those of ordinary people, using a technique that could measure function rather than merely structure.
Raine, Buchsbaum and LaCasse set out to provide exactly that. Their central aim was to use PET scanning to identify whether murderers pleading not guilty by reason of insanity (NGRI) showed differences in brain activity — specifically glucose metabolism — compared with a matched group of non-murderers, focusing on regions previously linked to violence. They were particularly interested in the prefrontal cortex (implicated in impulse control and the regulation of aggression) and a set of subcortical and limbic structures — the amygdala, hippocampus, thalamus and corpus callosum — thought to be involved in emotion, learning and the transfer of information between the hemispheres.
It is important to be precise about what the study set out to establish. Raine did not claim that brain dysfunction causes murder, nor that all violent offenders share the same abnormalities. The stated aim was more cautious: to test whether a specific, extreme group of violent offenders showed a distinctive pattern of brain activity relative to matched controls. Framing the aim accurately is important for the evaluation, because much of the criticism levelled at the study is really criticism of over-interpretations that Raine himself was careful to avoid.
Key Definition: NGRI (not guilty by reason of insanity) is a US legal plea in which a defendant admits the act but argues they should not be held criminally responsible because of a mental-health condition. Raine's participants had all entered this plea, which is why they were referred for brain assessment in the first place.
The study was a quasi-experiment using a matched-pairs design. It was quasi-experimental because the independent variable — whether or not a person was a murderer — was naturally occurring and could not, obviously, be manipulated or randomly allocated by the researchers. The independent variable was group membership (NGRI murderer versus control) and the dependent variables were the measures of glucose metabolism in the various brain regions, expressed as relative activity so that overall differences in brain size or metabolism did not distort the comparison.
The experimental group comprised 41 participants (39 men and 2 women), with a mean age of about 34, who had been charged with murder or manslaughter in the United States and had pleaded not guilty by reason of insanity. They had been referred for brain imaging to obtain evidence relating to their plea. Their reasons for pleading NGRI varied and included schizophrenia, head injury or organic brain damage, history of psychoactive drug abuse, epilepsy, and affective (mood) disorders — an important detail, because it means the sample was diverse in its underlying conditions rather than a single, homogeneous "type" of offender.
The control group comprised 41 participants matched to the murderers on sex and age. For the six murderers with a diagnosis of schizophrenia, the matched controls were also people with schizophrenia (drawn from a mental-health setting), so that the effects of schizophrenia would not be confounded with the effects of being a murderer. Crucially, no controls were taking medication, none had a history of serious psychiatric illness (other than the schizophrenia-matched pairs), and they were screened to be physically and psychiatrically healthy. This matching was central to the study's logic: by holding sex, age and (where relevant) schizophrenia constant across the two groups, Raine aimed to isolate the effect of the "murderer" variable itself.
The core measurement technique was PET scanning combined with a continuous performance task (CPT). The procedure ran as follows:
| Stage | What happened |
|---|---|
| 1. Drug-free period | Participants were kept medication-free for two weeks before scanning, so that psychoactive drugs would not distort brain activity. |
| 2. Practice | Each participant practised the continuous performance task before the scan so they understood what was required. |
| 3. Tracer injection | A radioactively labelled glucose tracer (fluorodeoxyglucose) was injected into the participant. Active brain regions take up more glucose, so they accumulate more tracer. |
| 4. Task performance | Participants performed the continuous performance task for 32 minutes — a target-detection task requiring sustained attention, chosen because it is known to activate the frontal lobes. |
| 5. PET scan | After the tracer had been taken up, a PET scan was performed, imaging ten "slices" (horizontal levels) of the brain and measuring glucose metabolism in cortical and subcortical regions. |
The logic of pairing PET with the CPT is worth spelling out, because it is a common exam discriminator. PET measures metabolic activity, but activity only means something relative to a task. By having every participant perform the same frontal-lobe task while the tracer was taken up, Raine ensured that any group differences in frontal activity reflected differences in how the participants' brains handled that task — not differences in what they happened to be thinking about. The CPT therefore functioned as a standardised cognitive challenge that made the two groups' brain activity directly comparable.
Exam Tip: A frequent mistake is to describe the CPT as an aggression task. It is not — it is a neutral attention task chosen because it reliably engages the frontal lobes. The point is to compare frontal engagement, not to provoke aggression in the scanner.
The scans revealed a consistent pattern of differences between the murderers and the controls. The findings fall into three groups: reduced prefrontal activity, abnormal activity in limbic and subcortical structures, and a set of regions where no difference was found (which matters just as much for interpretation).
The murderers showed reduced glucose metabolism in the prefrontal cortex — both in the lateral and medial prefrontal regions — compared with controls. This is the study's headline finding. Because the prefrontal cortex is associated with impulse control, planning, regulation of the limbic system and social judgement, reduced activity there is consistent with the idea that these individuals may be less able to inhibit aggressive impulses generated by deeper emotional structures.
The murderers also showed abnormal asymmetries and altered activity in several deeper structures:
| Region | Finding in murderers | Why it may matter |
|---|---|---|
| Amygdala | Reduced activity on the left, increased on the right (abnormal asymmetry) | The amygdala governs fear and emotional responses; abnormal function may reduce fear of consequences and impair emotional regulation. |
| Hippocampus | Abnormal asymmetry (reduced left relative to right) | Involved in learning and memory, including learning from punishment; dysfunction may impair conditioned fear and behavioural inhibition. |
| Thalamus | Increased activity on the right | Relays sensory and emotional information to the cortex; abnormal activity may distort emotional processing. |
| Corpus callosum | Reduced activity | Connects the two hemispheres; reduced function may impair the left hemisphere's ability to regulate right-hemisphere-generated negative emotion. |
Just as importantly, the murderers and controls did not differ in several regions, including areas such as the cerebellum and much of the temporal, parietal and occipital cortex not implicated in emotion. This selectivity strengthens the interpretation: the differences were concentrated in exactly the structures a neural theory of aggression would predict (prefrontal cortex and limbic system) rather than being spread randomly across the whole brain. A finding of global differences would have suggested some general, non-specific problem; the localised pattern points to a specific emotional-regulation circuit.
The overall pattern, then, is one of a dysregulated emotion circuit: over- or abnormally-active deep limbic structures generating emotional and aggressive impulses, coupled with an under-active prefrontal cortex less able to inhibit them, and a poorly functioning corpus callosum limiting inter-hemispheric regulation.
Raine and colleagues drew a set of deliberately cautious conclusions from these findings.
This last point is subtle but important. Raine's own later theorising distinguished reactive/impulsive aggression (associated with the prefrontal–limbic dysregulation seen here) from proactive/premeditated aggression, which may not show the same pattern. The NGRI sample is likely to be weighted towards impulsive, disordered violence, so the study speaks most directly to that subtype.
A major strength of the study is its use of a matched-pairs design with rigorous controls, which increases internal validity. The 41 controls were matched to the murderers on sex and age, and the six schizophrenic murderers were matched with schizophrenic controls, so schizophrenia was not confounded with murderer status; all participants were medication-free for two weeks and performed the same continuous performance task during tracer uptake. Because so many potential confounds were held constant, differences in glucose metabolism can be attributed more confidently to the "murderer" variable itself. The implication is that the observed prefrontal and limbic differences are unlikely to be artefacts of medication, age, sex or a specific psychiatric diagnosis, which strengthens the study's core claim.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.