In Our Messy, Reptilian Brains

By Sharon Begley
Newsweek
April 9, 2007

Let others rhapsodize about the elegant design and astounding complexity of the human brain—the most complicated, most sophisticated entity in the known universe, as they say. David Linden, a professor of neuroscience at Johns Hopkins University, doesn't see it that way. To him, the brain is a "cobbled-together mess." Impressive in function, sure. But in its design the brain is "quirky, inefficient and bizarre ... a weird agglomeration of ad hoc solutions that have accumulated throughout millions of years of evolutionary history," he argues in his new book, "The Accidental Mind," from Harvard University Press. More than another salvo in the battle over whether biological structures are the products of supernatural design or biological evolution (though Linden has no doubt it's the latter), research on our brain's primitive foundation is cracking such puzzles as why we cannot tickle ourselves, why we are driven to spin narratives even in our dreams and why reptilian traits persist in our gray matter.

Just as the mouse brain is a lizard brain "with some extra stuff thrown on top," Linden writes, the human brain is essentially a mouse brain with extra toppings. That's how we wound up with two vision systems. In amphibians, signals from the eye are processed in a region called the midbrain, which, for instance, guides a frog's tongue to insects in midair and enables us to duck as an errant fastball bears down on us. Our kludgy brain retains this primitive visual structure even though most signals from the eye are processed in the visual cortex, a newer addition. If the latter is damaged, patients typically say they cannot see a thing. Yet if asked to reach for an object, many of them can grab it on the first try. And if asked to judge the emotional expression on a face, they get it right more often than chance would predict—especially if that expression is anger.

They're not lying about being unable to see. In such "blindsight," people who have lost what most of us think of as vision are seeing with the amphibian visual system. But because the midbrain is not connected to higher cognitive regions, they have no conscious awareness of an object's location or a face's expression. Consciously, the world looks inky black. But unconsciously, signals from the midbrain are merrily zipping along to the amygdala (which assesses emotion) and the motor cortex (which makes the arm reach out).


Primitive brains control movement with the cerebellum. Tucked in the back of the brain, this structure also predicts what a movement will feel like, and sends inhibitory signals to the somatosensory cortex, which processes the sense of touch, telling it not to pay attention to expected sensations (such as the feeling of clothes against your skin or the earth beneath your soles). This is why you can't tickle yourself: the reptilian cerebellum has kept the sensation from registering in the feeling part of the brain. Failing to register feelings caused by your own movements claims another victim: your sense of how hard you are hitting someone. Hence, "but Mom, he hit me harder!"

Neurons have hardly changed from those of prehistoric jellyfish. "Slow, leaky, unreliable," as Linden calls them, they tend to drop the ball: at connections between neurons, signals have a 70 percent chance of sputtering out. To make sure enough signals do get through, the brain needs to be massively interconnected, its 100 billion neurons forming an estimated 500 trillion synapses. This interconnectedness is far too great for our paltry 23,000 or so genes to specify. The developing brain therefore finishes its wiring out in the world (if they didn't, a baby's head wouldn't fit through the birth canal). Sensory feedback and experiences choreograph the dance of neurons during our long childhood, which is just another name for the period when the brain matures.

With modern parts atop old ones, the brain is like an iPod built around an eight-track cassette player. One reptilian legacy is that as our eyes sweep across the field of view, they make tiny jumps. At the points between where the eyes alight, what reaches the brain is blurry, so the visual cortex sees the neural equivalent of jump cuts. The brain nevertheless creates a coherent perception out of them, filling in the gaps of the jerky feed. What you see is continuous, smooth. But as often happens with kludges, the old components make their presence felt in newer systems, in this case taking a system that worked well in vision and enlisting it higher-order cognition. Determined to construct a seamless story from jumpy input, for instance, patients with amnesia will, when asked what they did yesterday, construct a story out of memory scraps.

It isn't only amnesiacs whose brains confabulate. There is no good reason why dreams, which consolidate memories, should take a narrative form. If they're filing away memories, we should just experience memory fragments as each is processed. The cortex's narrative drive, however, doesn't turn off during sleep. Like an iPod turning on that cassette player, the fill-in-the-gaps that works so well for jumpy eye movements takes the raw material of memory and weaves it into a coherent, if bizarre, story. The reptilian brain lives on.

[Acknowledgements to Julian.]