Phrenology and the parliament of homunculi.
- the brain is the organ of the mind
- the brain is a composite of parts, each of which performs a distinctive task
- the size of the different parts of the brain is an index of the relative strengths of the different facilities
The first principle is broadly accepted in modern culture while the third is ridiculed. But the second principle, that the mind/brain is a composite of parts, is interesting.
The problem Gall sought to solve was reintroduced by John Locke in his Essay Concerning Human Understanding. Humans, according to Locke, are born without innate mental content as a tabula rasa (blank slate) and all knowledge is subsequently formed by our sensory experiences. Opponents argued Locke's theory was incorrect because the mind clearly organizes our experiences and knowledge into a coherent narrative giving us a sense of personal identify sustained over time. If such a facility did not exist, then the mind should be a chaotic mixture of experiences and confused memories leading humans to behave as the insane do. So our minds must have innate content for organizing knowledge and experiences.
Gall tried to solve this problem by breaking the brain into twenty-seven different facilities, each solely dedicated to a single function such as language, metaphysics or justice. But phrenology still failed to explain how a mind can integrate experiences and knowledge into a coherent whole. As the famous quip of 19th century philosopher Friedrich Lange put it, Gall merely introduced:
a parliament of little men together, each one of whom, as happens also in a real parliament, posses but a single idea which he ceaselessly tries to make prevail .. Instead of one soul, phrenology gives us forty.Neo-phrenology
Neo-phrenology is a pejorative, but reasonable, term for describing the theory of modularity, that a brain has specialized regions for different cognitive processes. While Gall divided the brain into separate functions for vanity, justice, metaphysics etc, a neo-phrenologist speaks of the 'fusiform face' module responsible for face recognition or of the 'frontal lobes' responsible for reasoning and planning.
The main driver for modularity is Functional Magnetic Resonance Imaging (fMRI). Bodily tissues are mainly water which contains two hydrogen nuclei. When tissues are placed in a magnetic field the hydrogen nuclei are aligned along the lines of force in the field. If a radio frequency electromagnetic field is then switched on and off the protons release energy which can be detected by a scanner. Protons in different tissues resonate at different frequencies. As a result, brain grey matter such as the cerebral cortex, white matter such as the nerve tracts that connect the cortex with other structures appear in distinctive shades. The contrast can be enhanced by administering agents that influence the resonance of the protons in various tissues.
Furthermore neural activity in a particular region is accompanied by changes in blood flow in that region. These changes are associated with alterations in blood oxygen level. Blood-Oxygen-Level-Dependent (BOLD) MRI is an indirect way of tracking brain activity over time when someone is responding to stimuli or engaging in an activity. This according to neo-phrenology allows us to map the modular human brain.*
MRI scans are invaluable for researching and diagnosing conditions like epilepsy and Alzheimer's. They do however have limitations.
Firstly there are questions over the reliability of the results. A report by Bennett & Miller found "quantifying the reliability of fMRI data is notoriously difficult ... The body of emerging evidence indicates that fMRI reliability is a highly variable construct, influenced by factors both within and outside the control of the investigator." Others such as Harold Pashler found the methodology used in some experiments all but guaranteed a link between a brain region and the emotion or behaviors under investigation: "A disturbingly large and quite prominent segment of fMRI scan research on emotion, personality and social cognition is using seriously defective research methods and producing a profusion of numbers that should not be believed." The full report is [here]. I am not remotely qualified to evaluate these findings; I simply note their existence.
Secondly, the very nature of fMRI emphasizes localized neural activity over activity distributed throughout the brain. Further refinement of the technology could reveal a much wider disbursement of neural activity than hard line neo-phrenologists would like us to believe. As Karl Friston put it, "the brain acts more as if the arrival of inputs provokes a wide spread disturbance in some already existing state."
Thirdly, there is the ever present epistemological problems between causation or correlation, necessary or sufficient, and identity theory. Suspicion of causation isn't particularly impressive and I've covered the identity theory of the brain in Are you your brain?. Debates around necessary or sufficient conditions typically arise over damaged areas of the brain. Suppose a person suffers a brain injury and loses the ability to speak. A neo-phrenologist will then claim this brain area is responsible for speech. But the injury only establishes the area is necessary for speech; it does not establish the area is sufficient for speech.
Does neo-phrenology remove the parliament of homunculi?
We have seen phrenology was an attempt to solve a problem with Locke's theory: if the mind lacked any innate rules for organizing sensory experiences, then we should expect sensory experience to become churned together into an incoherent whole. Gall's attempted to resolve this difficulty by dividing the mind into separate facilities but failed to explain how the separate facilities could operate as an unified whole.
Neo-phrenologists face similar challenges. The standard explanation is that bundles of neurons from specialized areas bind together to create the illusion of unity and self-hood. However there is no existing model of binding activity between different brain regions that does not result in an incoherent whole. Consider an analogy of 2 + 2. When aggregated we get 4 but the individual elements are lost. Likewise with bundles of neurons: the individual elements become lost during aggregation**. To see why this is unsatisfactory, consider the room about you. You are probably simultaneously aware of an unified visual field and of it's individual components. So what must be explained is both a sense of unification and of individual elements making up the whole.
To solve this problem, some neo-phrenologists propose multiple hierarchies of consciousness all leading to an unified consciousness. We are simultaneously conscious of several things, which, somehow, bind into an illusion of unified consciousness. Frankly, this offers us nothing but a parliament of homunculi with a speaker trying to keep order.
*This explanation is stolen from Raymond Tallis.
**Apologizes for the poor explanation. This post is already too long .