Have we reached the limit of our intelligence?


Given the limits of physiology, it seems we can count on little evolutionary improvement, writes WILLIAM REVILLE

HUMAN BEINGS are a product of biological evolution and, so it is interesting to ponder where evolution may take us in the future. For example, self-conscious intelligence is a primary distinguishing characteristic of humans in the animal world. Will human intelligence improve as we evolve further? Apparently, we can expect little if any further improvement because basic physical laws will probably prevent the brain from evolving into a significantly more powerful thinking machine. This whole area was reviewed this July by Douglas Fox in Scientific American.

Could we not evolve bigger brains and get smarter that way?

Well, for a start, brain size alone does not correlate simply with intelligence. Brain size increases with animal body size, but, as we all know, a cow is not smarter than a mouse, a cat or a terrier. In the animal kingdom, on average, as body size increases, brain size increases not as a fixed percentage, but as the ¾ power of body mass, eg. a muskrat’s body is 16 times larger than that of a mouse, but its brain size is only eight times larger. Unusually smart animals deviate from this law by having larger brains than would be expected from the ¾ power law. Humans are the top species in this regard, beating the ¾ law by a factor of 7.5. However, increasing brain size beyond a certain limit brings diminishing returns.

First, it would be very expensive energetically. Although the human brain represents only 2 per cent of body weight, it consumes 20 per cent of the calories we burn up while at rest. The thinking work of the brain is done by a fantastically complex system of interconnected nerve cells, called neurons, all communicating with each other through electrical signals. A neuron is composed of a cell body, containing the cell nucleus, where the computing is done, and a long thin cell extension called an axon, which extends out and connects with other neurons, carrying electrical signals to them. Extent of interconn- ectivity and speed of signal transmission are critical to the efficient functioning of the brain.

As brains get bigger, signals have to be transmitted over longer distances, which slows things down. In order to compensate, signals must travel faster, which can be partly achieved by making axons thicker but this means that neurons get bigger and pack together less efficiently, further increasing brain size. Obviously, increasing brain size to further improve intelligence quickly runs into limits imposed by energy consumption and efficiency of electrical signalling.

So, what about leaving brain size alone but improving intelligence by making the neurons smaller, particularly by making the axons thinner, but increasing neuron number? Here, you run into another constraint. The basis for the electrical signal that runs along the axon is the movement of ions (charged atoms) across the axon’s membrane wall. These ions move through protein channels that open and close. When they open, ions flow across the membrane, producing the electrical signal.

Individual channels are very sensitive and can open or close spontaneously – even a thermal vibration can cause this. When a neuron decides to fire, individual channels are likely, but not constrained, to obey. The overall system works because a very large number of channels are involved and a reliable majority always votes the right way. Furthermore, current axon size ensures that spontaneous opening hiccups do not trigger anomalous firing of electrical signals along the neuronal axon. However, if the axons were made significantly thinner, they would accommodate significantly fewer channels and spontaneous openings of channels would have a much greater chance of initiating anomalous electrical signals causing the signal to noise ratio to decrease to an unacceptable level.

If we have limited capacity to become individually smarter, Fox speculates that we may become collectively smarter by pooling our individual intelligences and acting in concert, somewhat like honeybees and other social insects do. Shades of the global thinking sphere (“noosphere”) here, as foreseen by Teilhard de Chardin.

And now that we also have the technology to store limitless information outside our own bodies on the Internet, we are fast approaching the day when individual memories will be used more to remember how to access various data bases than to remember data itself.

William Reville is Public Awareness of Science Officer and a Professor in the Biochemistry Dept at UCC. See understandingscience.ucc.ie