Overview – Maguire et al (2000)

Maguire, E. A., Gadian, D. G., Johnsrude, I. S., Good, C. D., Ashburner, J., Frackowiak, R. S. & Frith, C. D. (2000) Navigation-related structural change in the hippocampi of taxi drivers. Proceedings of the National Academy of Sciences of the United States of America. Volume 97, No. 8, pages 4398-4403.

Maguire et al (2000) is a key study in biological psychology. It investigated whether extensive experience with spatial navigation was associated with differences in the structure of the human brain. The researchers compared the brains of London taxi drivers with the brains of people who did not drive taxis. The study focused particularly on the hippocampus – a structure involved in forming memories and thought to play an important role in spatial memory and navigation. Maguire et al used magnetic resonance imaging (MRI) to examine the brains of the participants. They found taxi drivers had a greater volume of grey matter in the posterior hippocampus while controls had greater volume in some anterior regions. The researchers also found that the amount of time a person had worked as a taxi driver was related to the volume of particular parts of the hippocampus. The findings thus suggest that extensive experience with navigation can cause structural changes in the human brain.

You can jump to different sections of this article via the links below:

• Aim
• Method (sampling and procedure)
• Results
• Conclusions
• Evaluation and methodological issues
• Relation to biological psychology more broadly

Maguire et al (2000)

Aim

The aim of Maguire et al (2000) was to investigate whether extensive experience of spatial navigation was associated with structural differences in the hippocampus of the brain.

The hippocampus is a structure located inside each hemisphere of the brain. It’s involved in forming memories and is thought to play an important role in spatial memory (i.e. the ability to remember locations and navigate around a physical environment). The hippocampus is also part of the limbic system – a group of brain structures involved in processes such as memory, motivation, and emotion.

Previous research suggested that the size and structure of the hippocampus can differ between animals that rely heavily on spatial memory and those that do not. For example, some birds and small mammals that store food (e.g. squirrels) have relatively large hippocampuses because they need to remember the locations of food stores. In some species, hippocampal volume also increases during periods when spatial memory is particularly important.

Research had also shown that healthy human brains can differ in structure – such as differences between males and females and differences between musicians and non-musicians. However, previous research had not clearly established whether experience changed the human brain structure. It was also unclear whether the human hippocampus showed the same relationship with spatial experience that had been observed in some animals.

London taxi drivers provided a useful opportunity to investigate this question. To get the license, you must pass ‘The Knowledge’ – a rigorous test that involves memorising 25,000 streets and 20,000 landmarks across the city. By comparing these London taxi drivers with non-taxi drivers, researchers could investigate the neural impact of extensive spatial training.

So the main aim of the study was to investigate whether the amount of time someone had worked as a taxi driver was related to the volume of grey matter in the brain’s hippocampus.

Method

Maguire et al used a quasi-experiment (i.e. an experiment where the independent variable is naturally occurring and can’t be randomly assigned). The researchers compared groups of London taxi drivers and people who did not drive taxis:

• The key independent variable (IV) was whether the participant was a London taxi driver or a non-taxi driver. This was a naturally occurring variable because it’s not like the researchers could dictate who becomes a taxi driver with their life and who doesn’t.
• The main dependent variable (DV) was the structure and volume of the participants’ hippocampus. This was measured by analysing MRI brain scans. The researchers also examined the relationship between the length of time participants had worked as taxi drivers and the volume of different regions of their hippocampus.

The study used an matched pairs design. This means the researchers allocated control participants to the taxi drivers on the basis of similarities such as age, sex, right-handedness, and health.

Participants and sampling

The experimental group consisted of 16 licensed London taxi drivers. All 16 participants were:

• Male
• Right-handed
• Healthy
• London taxi drivers.

The mean age was 44 years, with an age range of 32-62 years. The taxi drivers had been licensed for an average of 14.3 years. However, there was considerable variation, with experience ranging from approximately 1.5 years to 42 years.

The control group consisted of people who did not drive taxis, which enabled the researchers to investigate whether the brains of taxi drivers differed from these non-taxi drivers.

These controls were selected from an existing database of structural MRI scans at University College London (UCL). For the voxel-based morphometry (VBM) analysis, the researchers compared the 16 taxi drivers with 50 non-taxi drivers. And for the more detailed pixel-counting analysis, the researchers used 16 controls selected from the larger control group. These controls were matched to the taxi drivers for characteristics including age, sex, and right-handedness.

Procedure

Maguire et al conducted MRI scans on the 16 licensed London taxi drivers. They then used an existing MRI database at UCL to collect control MRI scans from non-taxi drivers.

The two MRI scans – taxi drivers and matched controls – were then compared using two methods:

• VBM analysis: The researchers compared the MRI scans of the 16 taxi drivers with those of 50 controls using voxel-based morphometry (VBM). This technique standardises the scans and allows researchers to identify group-level differences in grey matter volume in different areas of the brain.
• Pixel-counting analysis: The researchers then carried out a more detailed analysis using the scans of the 16 taxi drivers and 16 individually matched controls. They compared the volume of different sections of the hippocampus, including the anterior, body and posterior regions.

The researchers calculated the volume of different regions of the left and right hippocampus. They then compared these measurements between taxi drivers and controls.

The researchers also examined whether the length of time someone had worked as a taxi driver was related to the volume of different hippocampal regions.

The person analysing the scans was blinded to whether each scan belonged to a taxi driver or a control participant. This means that they did not know which group each scan came from and so reduced the possibility that the researcher’s expectations could influence the measurements. 

In short, the researchers compared MRI scans from London taxi drivers with MRI scans of non-taxi drivers and focused particularly on the hippocampus.

Results

The MRI analysis showed differences in the distribution of grey matter within the hippocampus of taxi drivers and controls.

VBM results

The voxel-based morphometry analysis found no significant differences between the brains of the two groups in most areas. However, there were significant differences in the hippocampus specifically:

• Taxi drivers had significantly more grey matter volume in the right and left posterior hippocampus than controls.
• Controls had relatively more grey matter volume in the right and left anterior hippocampus than taxi drivers.

Pixel-counting results

The more detailed pixel-counting analysis found that there was no significant difference in the overall volume of the hippocampus between taxi drivers and controls but there were significant differences in particular regions:

• Taxi drivers had a significantly more volume in the posterior (left and right) hippocampus than controls.
• Controls had a significantly more volume in the right anterior hippocampus than taxi drivers.
• Both taxi drivers and controls had a significantly greater volume of the right hippocampal body compared with the left.

Correlation results

The researchers also examined the relationship between taxi-driving experience and hippocampus volume. They found a:

• Positive correlation between the number of years spent as a taxi driver and the volume of the right posterior hippocampus. In other words, the longer someone had been a taxi driver, the bigger their right posterior hippocampus.
• Negative correlation between the number of years spent as a taxi driver and the volume of the anterior hippocampus. In other words, the longer someone had been a taxi driver, the smaller their anterior hippocampus.

Summary of results

The table below summarises the regions of the hippocampus where the researchers found differences between taxi drivers and controls. Where ‘taxi drivers’ is shown, this means they had significantly more volume in that region and where ‘controls’ is shown this means they had larger volume in that region. ‘No effect’ obviously means there was no significant difference between the two groups.

In short, taxi drivers had greater posterior hippocampal volume and controls had greater anterior hippocampal volume.

Conclusions

Maguire et al concluded that London taxi drivers showed specific structural differences in the hippocampus compared with people who did not drive taxis. The findings also provide evidence that the normal human brain is capable of structural change in response to prolonged environmental experience.

The hippocampus is involved in spatial memory and navigation

The study found regional differences in hippocampal volume between taxi drivers and controls.

The important finding was not that taxi drivers had larger hippocampi overall. Instead, the results suggested a difference in the distribution of grey matter within the hippocampus. Taxi drivers had greater volume in the posterior hippocampus, while controls had greater volume in some anterior regions. The amount of time spent working as a taxi driver was also related to these differences.

The researchers suggested that these differences were related to the extensive spatial memory and navigation demands of being a London taxi driver. Taxi drivers need to learn a large number of streets, locations, and routes in order to pass ‘The Knowledge’ and perform their job.

The findings thus supported the idea that the hippocampus plays an important role in spatial memory and navigation.

Neuroplasticity and the effects of experience

The findings also provided strong evidence for neuroplasticity – the ability of the brain to change its structure or function in response to experience.

Taxi drivers had different patterns of hippocampal structure from the control group and the amount of time spent working as a taxi driver was linearly correlated with hippocampal volume. In other words, the more years participants had been taxi-driving, the more additional volume in the right posterior hippocampus. The researchers thus suggested that these differences in hippocampal structure were acquired over time through experience. In other words, prolonged environmental demands – such as extensive navigation – can cause changes in the adult brain.

Normally, quasi-experiments only show correlation and not causation. This is because of self-selection: people bring pre-existing traits with them (like habits, motivations, or in this case brain structures) and so you can’t tell whether the variable caused the outcome or if those pre-existing traits did. It could be, for example, that people who already had these brain structures were more likely to become taxi drivers in the first place rather than that being a taxi driver caused the brain to change. 

But Maguire et al directly countered this argument with their correlation analysis. Again, time spent working as a taxi driver positively correlated with right posterior hippocampal volume in a linear fashion. In other words, the longer someone had been working as a taxi driver, the larger that region was. If these brain differences were simply pre-existing, volume would not increase proportionally with years on the job like this because a pre-existing trait would remain static regardless of time spent in the profession. As such, this strongly supports the idea that the structural differences were acquired through years of navigation, i.e. that the experience of being a taxi driver caused the changes in the brain.

So, overall, the study provides strong evidence for neuroplasticity and supports the idea that experience can cause changes in the hippocampus.

Evaluation

Research methods and techniques

Maguire et al used a quasi-experiment with MRI brain scanning. The researchers compared naturally occurring groups of taxi drivers and non-taxi drivers.

• Objective brain measurement: MRI provided objective measurements of brain structure. This reduced the need for researchers to make subjective judgements about participants’ hippocampi.
• Matched participants: The control group was matched with the taxi drivers on important characteristics such as age, sex, handedness and health. This helped reduce the influence of some extraneous variables.
• Two analysis techniques: The researchers used both VBM and pixel counting. Using two methods to investigate hippocampal structure provided more than one source of evidence.
• Blinded analysis: The person carrying out the pixel-counting analysis did not know whether each scan belonged to a taxi driver or a control. This reduced researcher bias.
• No random allocation: Participants could not be randomly assigned to become taxi drivers. The groups were naturally occurring. Therefore, other differences between the groups may have contributed to the results.
• Correlation does not prove causation: The study found correlations between taxi-driving experience and hippocampal structure. However, it cannot prove that taxi driving caused these brain differences. For example, people with these kinds of brains might have been more likely to want to become taxi drivers in the first place, which could also explain the link.
  • Response: Maguire et al countered the self-selection argument by showing that posterior hippocampal volume increased linearly with time spent driving. Since pre-existing structural advantages would not progressively expand over a driver’s career, this correlation strongly indicates that brain structure changed as a result of driving experience (see above for more).

Population and sampling

The experimental group consisted of 16 licensed London taxi drivers. The VBM analysis also included 50 controls, while 16 matched controls were used for the more detailed pixel-counting analysis.

• Relevant sample: London taxi drivers were highly suitable for this study because their occupation requires extensive spatial memory and navigation.
• Wide range of experience: The taxi drivers had between approximately 1.5 and 42 years of experience. This allowed the researchers to investigate whether hippocampal volume was related to length of experience.
• Matched controls: The control participants were matched with the taxi drivers for characteristics such as age, sex and handedness. This reduced some potential confounding variables.
• Small experimental sample: Only 16 taxi drivers took part. This is a small sample and limits how confidently the findings can be generalised.
• Restricted sample: All of the taxi drivers were healthy, right-handed men from London. This means the sample did not represent women, left-handed people or people from other occupations and locations.
• Opportunity sampling of controls: The control scans came from an existing MRI database at the same research unit. This means the controls were people whose scans were already available rather than a randomly selected sample of the population.

Types of data

Maguire et al collected mainly quantitative data. This included measurements of grey matter volume and the volume of different hippocampal regions.

• Precise: MRI provided numerical measurements of brain structure. This allowed the researchers to compare hippocampal volume objectively.
• Objective biological data: The measurements were based on physical characteristics of the brain rather than participants’ self-reports. This reduces problems with inaccurate reporting.
• Detailed measurements: The researchers could examine different regions of the hippocampus rather than simply measuring the brain as a whole.
• Correlational data: The researchers could examine whether the number of years spent as a taxi driver was related to hippocampal volume.
• Limited information about experience: The numerical brain measurements do not explain exactly how participants learned to navigate London or what strategies they used to remember routes and locations.
• Structural data do not directly measure behaviour: MRI showed differences in brain structure, but it did not directly measure participants’ navigation ability or prove that the structural differences caused better navigation.

Representativeness and generalisability

The study has several limitations when applying its findings to the wider population.

• Relevant occupational group: Taxi drivers provided a particularly useful sample for studying spatial memory because their occupation requires extensive navigation.
• Matched controls: Matching the control participants on important characteristics made the comparison between groups more meaningful.
• Small sample: Only 16 taxi drivers were included in the experimental group. A small sample may not represent the wider population.
• Male-only sample: All taxi drivers were male. Therefore, the findings cannot confidently be generalised to female taxi drivers.
• London-specific sample: The taxi drivers were all from London. The findings may not apply to taxi drivers in cities with very different geographical layouts or licensing systems.
• Occupational specificity: Taxi drivers have unusually high levels of navigation experience. The findings may therefore not apply to people whose everyday lives involve less demanding spatial navigation.

Ethical issues

• Informed consent: Participants gave informed consent before undergoing MRI scanning.
• Protection from physical harm: MRI is a non-invasive technique and does not involve e.g. exposure to ionising radiation like CT scans or X-rays. However, participants would still need to be screened for safety because the scanner contains a very strong magnetic field.
• Right to withdraw: Participants could choose whether to take part in the research and could withdraw from the study.
• Privacy: Brain scans contain sensitive personal information. Researchers therefore needed to ensure that participants’ scans and personal information were kept confidential. The control scans came from an existing MRI database. Appropriate consent and confidentiality procedures would therefore be important when using previously collected scans for new research.

Overall, the study involved relatively low physical risk and was carried out according to the British Psychological Society’s ethical guidelines.

Validity

• Internal Validity: Do the findings accurately show a relationship between taxi-driving experience and hippocampal structure?
  • Matched participants: The controls were matched with the taxi drivers on characteristics such as age, sex, handedness and health. This reduced the influence of some extraneous variables.
  • Objective measurements: MRI provided objective measurements of hippocampal structure. This reduces the risk of researcher judgement affecting the results.
  • Blinded analysis: The person carrying out the pixel-counting analysis did not know which scans belonged to taxi drivers. This reduced researcher bias.
  • Two measurement techniques: Both VBM and pixel counting identified differences in hippocampal structure. This provides converging evidence for the findings.
  • Confounding variables: Taxi drivers and controls may have differed in other ways besides their occupation. For example, people who naturally have better spatial memory may be more likely to become successful taxi drivers. This could explain some of the differences in hippocampal structure.
    • Reply: However, the linear correlation between years driven and posterior hippocampal volume suggests these changes were acquired over time rather than pre-existing (see above).
      • Counter: However, this linear correlation still does not rule out other explanations. For example, chronic stress can also affect the hippocampus through increased cortisol. If driving taxis is particularly stressful, the linear relationship between time spent driving taxis and increased hippocampal volume could be related to the effects of stress rather than navigational experience.
• Ecological Validity: Do the conditions of the study reflect real-life navigational ability?
  • Real-world experience: The study investigated an occupation that genuinely requires extensive spatial memory and navigation. This gives the research strong relevance to real-life behaviour.
  • Natural variation in experience: The taxi drivers had different amounts of professional experience, ranging from 1.5 to 42 years. This allowed the researchers to examine how real-world experience was related to hippocampal structure.

Reliability

• Internal Reliability: Are the measurements consistent within the study?
  • Standardised MRI procedures: Participants’ brains were scanned using the same general imaging method. This provided a consistent way of measuring brain structure.
  • Objective measurements: Hippocampal volume was measured numerically rather than being judged subjectively.
  • Blinded analysis: The person carrying out the pixel-counting analysis did not know which group the scans belonged to. This reduced the possibility of subjective bias.
  • Two techniques: The researchers used VBM and pixel counting to investigate hippocampal structure. Similar findings across methods increase confidence in those measurements.
• External Reliability: Can the study be repeated with another sample?
  • Clear procedures: MRI scanning and the analysis techniques were clearly defined. This means that other researchers could use similar procedures.
  • Objective measurements: MRI measurements can be collected and analysed using standardised procedures.
  • Small sample: Only 16 taxi drivers were studied. A larger sample might produce different results.
  • Changes in technology: MRI technology and methods of analysing brain structure have developed since 2000. A modern replication may use different equipment and analysis techniques.

Ethnocentrism

• UK sample: All of the taxi drivers were from London. This means the findings may reflect the particular cultural and environmental conditions of London rather than being universally applicable.
• Specific occupation: London taxi drivers have to pass The Knowledge, which requires unusually extensive knowledge of London’s streets and locations. Other cultures and countries may have different taxi-licensing systems and different demands on drivers.
• Cultural and environmental differences: Navigation depends partly on the environment in which people live. People in cities with different layouts, transport systems or geographical features may develop different navigational demands. Therefore, the relationship between navigation experience and hippocampal structure may not be identical across cultures.
• Limited cultural diversity: The study did not compare people from different countries or cultural backgrounds. Therefore, it cannot establish whether the findings apply universally to human brain development.

Evaluation summary table

Relation to biological psychology more broadly

Maguire et al (2000) is important to biological psychology demonstrated a clear link between spatial memory and the hippocampus: London taxi drivers showed differences in the distribution of hippocampal grey matter compared with non-taxi drivers. The study also provided empirical support for experience-dependent neuroplasticity – the brain’s ability to alter its physical structure in response to environmental demands. Because posterior hippocampal volume positively correlated with time spent driving a taxi, Maguire et al challenged the long-held belief that the adult brain is biologically fixed, proving instead that intensive learning can actively reshape neural architecture over time.

The table below summarises how Maguire et al compares with other core biological psychology studies and evaluates its contribution to this area: