Long before the applications of EEGs to the analysis of the nervous system, people were studying the electrical activity of the brain. In 1875, Richard Caton of the United Kingdom first displayed this electrical activity by applying what is called a galvanometer to the exposed brains of rabbits and monkeys. With this, he determined that the outer gray matter of the brain was more positive than the inner layers, and that the electric current of the gray matter exhibited negative variation with the onset of activity. Caton was also the first to witness and report the visually-evoked fluctuating potentials we now know exist (Bronzino et al., 1995).
With discoveries such as those by Caton came about the invention and later improvements of the electroencephalograph to further understand the potential of the brain. Around 1929, Hans Berger, a psychiatrist from Jena, Germany, became interested in the previous studies of electroencephalography in animals and began to conduct his own experiments using dogs. Two devices were originally available for recording brain activity. Berger found the first device, the capillary electrometer, developed by Etienne Jules Marey in 1876, to be unsatisfactory for his experiments. This led him to concentrate on the second device, the string galvanometer, developed by a Dutch physiologist named Willem Einthoven in 1903. The string galvanometer was a modified version of an instrument used to detect, measure, and determine the direction of small electric currents by means of mechanical effects produced by a coil in a magnetic field (Bronzino et al., 1995).
Berger saw problems with the galvanometer and immediately sought to improve it. He first attempted to increase the sensitivity of the device to current by reducing the tension in the string, but with this there was a reduction in the speed of the response. Other problems involved the electrodes that were used. The cortical signals had low amplitudes so the electrodes had to be very stable and produce no voltages that could interfere with electrocortical readings. He used zinc electrodes for his first dog studies just after the turn of the century. For humans, he used zinc-plated needles sterilized in 10% formaline solution. These were high-resistance electrode pairs that reduced the amplitude of the recorded activity. He tried to solve this problem with the use of thin lead-foil electrodes wrapped in flannel and soaked in 20% NaCl solution. This reduced the resistance from the previous value of 1600 ohms to a range from 500-7600 ohms, depending on the size of the electrodes. Berger's final improvement involved placing a capacitor in series with the electrodes to block the steady potential difference that previously arose due to slight electrochemical differences in the two lead electrodes (Thatcher et al, 1994).
At this time, he began to use the more modified two-string galvanometer. With this, ECGs and EEGs of his subjects could be recorded. This was important because few people believed that his recordings actually originated in the brain. He also measured activities such as the friction of blood in the cerebral arteries, the pulsations of the brain and scalp, respiration, and glandular activity and showed that the time course and frequency of the EEGs did not match these measurements (Bronzino et al., 1995).
The study of EEGs first came to the United States in 1935 when two men named Jasper and Carmichael published an English version verifying Berger's observations. They also confirmed his findings and went on to show that a two-channel system could be used to record the EEG of a person with a compulsive disorder. They found that the alpha wave frequency on the left side of the brain was 10/second compared to 6-8/second on the right side. This was one of the first applications of an EEG to study brain pathology. Jasper went on to create the Electrophysical Laboratory at the Montreal Neurological Institute. Around the same time, three men (Gibbs, Davis, and Lennox) at Harvard University were pursuing their interest in epilepsy by using EEGs for its diagnosis. For this study, they used an apparatus built by Lovett Garceau consisting of a four-stage, single-sided, resistor-capacitor-coupled amplifier made with high-grain, screen-grid tubes driving a direct-inking telegraphic recorder called an Undulator. This Undulator was quickly replaced in 1935 by the d'Arsonval-type inkwriter. This was also the time when the Grass Instrument Company was founded to produce electrophysical equipment. This equipment led to the use of folding chart paper that would provide three-, four-, and six-channel EEGs. This new apparatus included pens situated at the right of a knee-hole console and ample viewing space for the evolving record. The standard chart speed of 30 mm/s came about from this machine (Bronzino et al, 1995).
Franklin Offner of Quincy, Massachusetts also contributed to the improvements of EEGs through his company in Chicago by using a new piezoelectric inkwriter, called the Crystograph. This inkwriter used two slabs of rochelle salt crystals. One pair of these slabs was clamped on each of three of its four corners. The fourth corner's pair was free to move with the application of voltage to electrodes on the crystals. This movement created motion of a connected inkwriting stylus. The chart speed on this machine was a little less than standard at 25 mm/s (Thatcher et al., 1994).
The last great improvement of the EEGs was in the late 1940s when instruments began to focus on the use of two rotary switches which could connect each side of a differential amplifier input to any of 21 electrodes of a 10-20 system. This allowed the addition of a low-current ohmmeter which measured the resistance of any pair of electrodes for studying more specific locations in the brain (Bronzino et al., 1995).