The Brain That Changes Itself Summary and key ideas

The Brain That Changes Itself, by Norman Doidge.

Doidge begins with a reversal of an older picture of the brain. For centuries, mainstream science treated the adult brain as something that could decline but not meaningfully rebuild itself. Damaged or poorly developed cells were assumed to remain damaged, and lost abilities were expected to stay lost. The brain was imagined as a machine with permanent circuits, each assigned one function. This view encouraged what Doidge calls neurological nihilism: if the brain could not change, many problems seemed untreatable.

The book does not deny that brain regions can specialize. Lesions often correspond to lost abilities, and Broca’s and Wernicke’s language areas supplied real evidence. The overreach was turning such correlations into a rigid rule: every mental or sensory function had one immutable location. Doidge describes findings from the late 1960s and early 1970s that challenged this rule. Activity could alter brain structure, refine circuits, and sometimes let another region help when one failed. He uses neuroplasticity for this capacity to change through activity, learning, injury, and new sensory input.

Cheryl Schiltz makes the reversal physical. Her vestibular system—the inner-ear system that detects motion and gravity and helps stabilize the eyes—was reported as almost entirely lost. Another test found roughly two percent remaining function, so the figures were not identical measures. Cheryl experienced a constant sense of falling, visual instability, exhaustion, anxiety, and difficulty working. Ordinary words such as balanced or grounded described a bodily emergency rather than a mood.

Bach-y-Rita gave her a device with a motion sensor called an accelerometer and electrodes touching the tongue. It translated head movement into patterns of stimulation. With it operating, Cheryl could stand without falling, and the persistent falling sensation disappeared. Her vestibular organs had not been repaired; her brain had learned to use another channel for information normally supplied by them.

Training changed the significance of the device. When it was removed, Cheryl could initially remain upright, and later she could stand with her eyes closed and balance on one foot. The residual effect lasted far longer than simple exposure would predict: after brief use it persisted for about an hour, then for hours, days, and eventually months as practice continued. Bach-y-Rita proposed that damaged tissue was producing confusing signals while repeated exposure strengthened alternate pathways. Cheryl eventually stopped using the device and no longer considered herself a “Wobbler.” The early results varied, and the mechanism remained an inference, not a guarantee.

The same logic appeared in sensory substitution for vision. In Bach-y-Rita’s tactile-vision system, a camera sent images through a computer to vibrating stimulators on a blind person’s back. With practice, users learned to recognize objects, faces, shadows, perspective, and depth. They described the scene as located in the external world, even when stimulation came through the abdomen. The apparatus could be bulky and its resolution poor, but the brain learned to interpret patterned touch as spatial information.

Bach-y-Rita therefore argued that vision is constructed by the brain. The eyes register changes in light; the brain interprets electrical signals. A cane offers a simpler version: pressure arrives in the hand, yet the user perceives objects and layout at the cane’s tip. For substitution to work, the brain must learn a new code and reorganize the system processing it.

These cases challenged localizationism without making the brain a blank slate. Children had sometimes developed normal speech despite severe loss of the left hemisphere, and young animals retained movement after motor-cortex removal. In one cat experiment, visual cortex responded to images, touch, and sound. Bach-y-Rita inferred that sensory cortices might be more polysensory than their names suggested. Different receptors convert forms of energy into electrical patterns, and cortical areas share a similar layered organization. The adapting brain, not the machine alone, was the central discovery. The idea met resistance: a paper using the word plasticity was reportedly rejected six times, and a mentor dismissed the device as an adult toy.

Pedro Bach-y-Rita supplied a different proof. After a 1959 stroke, he was paralyzed, unable to speak, and considered likely to need an institution. A conventional four-week rehabilitation program produced little improvement. His son George began with crawling, then led him through standing and walking. The family turned ordinary activities into practice: picking up coins, washing pots, and playing games that required controlled movement. Pedro’s speech returned. He relearned typing through progressively smaller movements, resumed full-time teaching within a year, and later remained active in work, hiking, and travel.

An autopsy found a massive lesion, including brain-stem damage and destruction of ninety-seven percent of the nerves running from cortex to spine. The case did not depend on a small original injury. It showed that late recovery could follow devastating damage when practice was motivated and resembled life. Bach-y-Rita treated plateaus as possible consolidation periods rather than automatic proof that learning had ended. Still, Pedro’s recovery was an exceptional case, not a promise for every severe stroke.

Together, these cases define the book’s foundational claim. The brain is not a set of sealed modules. Activity can make a new input meaningful, recruit alternate circuits, and restore function after severe injury. Yet plasticity is double-edged: it can make useful change possible, but it can also stabilize harmful patterns and remains incomplete. The fixed-brain model gives way to a more demanding idea of recovery—one that depends on training, time, and the brain’s capacity to reorganize.