Mazes have been a part of human culture for thousands of years, although their nature has changed significantly over time. They were originally designed to guide the visitor on a long, circuitous but crucially not confusing route. As mazes became larger and more complicated, visitors frequently became lost in them and realised that becoming lost could be an enjoyable aspect of the experience. Thus the idea of the maze as a puzzle was born.
Navigating a maze makes use of a range of key cognitive skills including spatial learning and memory, working memory, decision making, and general reasoning ability. Human beings differ in their level of cognitive skill, and hence also in their ability to solve mazes. The psychologist Stanley Porteus realised that these differences in maze-solving performance could be measured, and used as an index for certain kinds of intelligence.
In this article, we will discuss in detail the Porteus Maze, the cognitive test created by Stanley Porteus in the wake of this realisation. We will discuss what the maze is, how robust it is as a test, how it can serve as a diagnostic tool for mental illness, and how performance on the test is associated with genetic attributes of those who take it.
Mazes and the Brain
Navigating a maze is a cognitively demanding task, and so we would expect it to be associated with activity in a range of human brain regions. More specifically, we might expect to see activity in parts of the brain central to planning and decision making such as the prefrontal cortex, as well as intensive recruitment of the visual cortex, the hippocampus and motor regions.
A 2006 study conducted by a group of researchers in Germany used fMRI to ascertain which parts of the human brain are specifically involved with solving mazes. A group of healthy volunteers solved a series of mazes and pseudo-mazes (the latter being much easier to navigate on account of having no bifurcations in the route). Activity across their brain was measured and compared between states when solving mazes, pseudo-mazes and when at rest.
Increased blood flow was seen in various brain regions when subjects were solving either a maze or pseudo-maze compared to when at rest. Both kinds of navigation problem recruited the dorsal visual stream, the thalamus and major motor areas including the basal ganglia. However, only the maze produced a significant activation of the prefrontal cortex, an observation the researchers explain by the increased need for executive decision making in solving mazes.
As the researchers argue, this fMRI evidence supports the belief that mazes can serve as a specific tool to compare the executive function and decision-making ability of different people. It is interesting that no significant activity was seen in the hippocampus, despite this brain region playing a key part in spatial navigation under most circumstances.
What is the Porteus Maze?
The Porteus Maze emerged as a psychometric tool in the early twentieth century, at a time when psychometrics—and intelligence testing in particular—were in vogue. The famous Stanford-Binet scale (originally the Binet-Simon scale) was released in 1905 and served as the most popular cognitive ability test at the time. It was frequently used to determine whether schoolchildren qualified as mentally handicapped, and fed into a general culture of eugenics.
Stanley Porteus developed his Porteus Maze Test as a supplement to the Stanford-Binet scale, claiming that it would measure cognitive attributes that other psychometricians had ignored. The test is designed to focus on the capacity to plan ahead, what Porteus refers to as “prudence, foresight, planning capacity” as well as the abilities to deal with new situations and get better with practice.
Unlike an IQ test, the Porteus maze test is not timed and does not rely upon the subject’s verbal or mathematical intelligence. Those who take the test are simply required to solve a series of mazes of increasing difficulty. Each maze can only be attempted twice (or four times for the more difficult ones), and it is not permitted to backtrack once a wrong turn has been taken.
The results are tabulated into a quantitative and qualitative score. The quantitative score is a “test age” which reflects how many attempts were required for each maze as well as the difficulty level of the hardest maze solved. The qualitative score reflects one’s ability to follow instructions, waviness of lines drawn, corner-cutting, and other drawing errors. Higher figures for both scores indicate worse overall cognitive performance.
The Porteus Maze and Disease
In addition to its historical role as a form of intelligence test, the Porteus Maze has also proven popular as a diagnostic tool for various mental illnesses and neurological conditions. The test has been used to assess patients with lesions in the basal ganglia, traumatic brain injury, Alzheimer’s disease, anxiety disorders, mental retardation, ADHD and schizophrenia.
The maze is designed to assess executive function, and as such lends itself to the diagnosis of disorders associated with poorer planning, decision making, and behavioral inhibition/disinhibition. A 1998 study describes the test as “under-utilized” (a fair assessment, as most studies using the test, seem to date from the 1960s or earlier) and advocates for its employment with neuropsychological patients.
The use of the Porteus Maze with patients suffering from neurosis is longstanding. For example, a 1963 study from researchers at Columbia University in the US reports on associations between subjects’ qualitative scores on the Porteus Maze Test and their levels of stress and anxiety. 30 adolescent boys, divided into control and “neurotic” groups (boys who had displayed “acting out” and “delinquency”), took the test, and their scores were corrected for age, socioeconomic background and intelligence.
The researchers observed that there was a significant difference in the distribution of qualitative scores between the two groups, with the neurotic group scoring much higher on average, and with little overlap between the distributions. In particular, neurotic boys spent less time on each maze than boys in the control group and showed greater variability in error rate.
A 1960 experiment with schizophrenic patients (which would most likely not have passed the approval of an ethics committee in modern times) involved using the Porteus Maze to assess the effects of surgical intervention. Schizophrenics underwent a “topectomy” in which parts of their frontal cortex were removed, and they were then allowed eight years to recover. The patients’ scores on a Porteus Maze Test significantly decreased over the eight years, contradicting prior claims that this operation did not cause long term cognitive deficits.
More recent research has linked Porteus Maze performance with dementia via driving ability. A 2012 study involved administering the test to a number of elderly patients, some with Alzheimer’s disease and some healthy. Out of a range of metrics tested, the Porteus Maze was the only one that significantly predicted differences in driving ability between the diseased and healthy groups. The authors suggest that the test could be used to assess whether dementia patients are safe to continue driving.
The Robustness of the Porteus Maze
In order to be useful as a diagnostic tool, it is essential that cognitive ability tests show a range of robust statistical qualities. In this respect, various data suggest that the Porteus Maze test holds up well, exhibiting both validity and reliability.
The test displays a high internal consistency i.e. it is rare for aspects of a subject’s performance to contradict one another, and scores on different parts of the test tend to correlate with one another. High internal consistency strengthens the claim that a test is measuring a single particular underlying phenomenon, rather than tabulating a meaningless average for a range of different measurements.
When subjects take the test multiple times, a slight regression to the mean is observed i.e. those who score above average tend to score less in retakes and those who score below average tend to score more in retakes. A small, self-attenuating regression to the mean is normal for all tests and does not undermine their validity.
The Porteus Maze is less discriminating at the high end (known as a “ceiling effect”), with just over a quarter of subjects getting the maximum possible score. There is also a low correlation between qualitative score errors, as well as a very high weighing of this score towards “wavy lines” and “pencil lifting errors” (these errors accounting for 70% of the variance in the qualitative score).
There is controversy over the extent to which the Porteus Maze Test actually measures something not captured by IQ tests, with some researchers finding low correlations and others finding high correlations between the two test types.
Overall, the Porteus Maze appears to be a highly robust test which is most useful for discriminating between cognitive abilities at the lower to middle range and shows promise as a metric of a non-IQ-related cognitive ability. It also does not seem to be biased by socioeconomic status.
Genetics and the Porteus Maze
One key question that always arises in the context of psychometric testing is the extent to which a test is measuring phenomena arising from genetic factors as opposed to environmental ones. This is important because, if we see that certain subjects perform badly on the Porteus Maze, we would like to know if it is possible to change these people’s circumstances in order to improve their performance.
The interaction between genetics and environment is complex, which makes it difficult to isolate genetic causes as such. This is why psychologists tend to rely on the concept of heritability i.e. the extent to which the variance in some measured attribute of a person is explained by a certain factor under certain circumstances. With IQ tests, for example, it is widely observed that performance is around 50% heritable, although this heritability varies with age and environment.
A 2017 study examined the heritability of the Porteus Maze Test according to the standard paradigm, comparing the performance of identical twins (who share 100% of their DNA) to non-identical twins (who share only 50% of their DNA). Comparing twins allows for a kind of genetic control, allowing for non-genetic contributions (and hence also genetic ones) to be isolated.
The authors report that the test’s qualitative score was 33% heritable in a cohort of age 9-10 years and 52% heritable in a cohort at ages 11-13 years. This increase is heritability with age is reminiscent of a similar increase seen with IQ and may reflect reductions in brain plasticity during development. No significant gender differences were seen in heritability for the qualitative score.
In contrast, the test age did show a notable influence from gender: at 11 to 13 years genetic factors were seen to be more important for girls, whereas non-shared environmental factors were more important for boys. The heritability of the test age in the cohort with ages 9-10 years, however, was 53% and did not differ significantly between genders.
Mazes have served a role in human entertainment for millennia, but their relationship to the human brain and cognitive abilities also allow them to serve a role in characterising cognitive function and diagnosing mental illness. Solving mazes relies upon a number of key brain regions, especially those involved with executive decision making.
The Porteus Maze, a test designed to measure cognitive abilities not captured by IQ tests, has shown robustness, significant association with genetics, and usefulness in a range of medical contexts. Nevertheless, the test appears under-utilized in research today. Perhaps, armed with the knowledge of its strengths, researchers of the future can raise the Porteus Maze Test to a more prominent status in modern psychometrics.
- National Building Museum. 2019. A Brief History of Mazes – National Building Museum. [ONLINE] Available at: https://www.nbm.org/brief-history-mazes/. [Accessed 27 January 2019].
- Peter Kirsch, Stefanie Lis, Christine Esslinger, Harald Gruppe, Peter Danos, Jochen Broll, Jörg Wiltink, Bernd Gallhofer. Brain Activation during Mental Maze Solving. Neuropsychobiology. 2006;54:51–58.
- Serge Nicolas, Bernard Andrieu, Jean-Claude Croizet, Rasyid B. Sanitioso, Jeremy Trevelyan Burman. Sick? Or slow? On the origins of intelligence as a psychological object. Intelligence. 41. 2013. 699–711.
- Carlozzi N.E. 2011. Porteus Maze. In: Kreutzer J.S., DeLuca J., Caplan B. (eds) Encyclopedia of Clinical Neuropsychology. Springer, New York, NY.
- Robert Krikorian & John A. Bartok. Developmental Data for the Porteus Maze Test. The Clinical Neuropsychologist. Volume 12. 1998. Issue 3.
- Anthony, A. A. 1963. The relationship between neuroticism, stress, and qualitiative Porteus Maze performance. Journal of Consulting Psychology. 27(6), 513-519.
- Smith, A. 1960. Changes in Porteus Maze Scores of Brain-Operated Schizophrenics After an Eight-Year Interval. Journal of Mental Science, 106(444), 967-978.
- Ott, B. R., Heindel, W. C., Whelihan, W. M., Caron, M. D., Piatt, A. L., & DiCarlo, M. A. 2003. Maze test performance and reported driving ability in early dementia. Journal of geriatric psychiatry and neurology. 16(3), 151-5.
- Catherine Tuvblad, Marcella May, Nicholas Jackson, Adrian Raine, Laura A. Baker. Heritability and Longitudinal Stability of Planning
and Behavioral Disinhibition Based on the Porteus Maze Test. Behav Genet. 2017. 47:164–174.