The string galvanometer was not suitable for making recordings over extended time periods. Oscillographs could also be used with bromide paper but long-duration recordings presented a technical problem. However, another device invented decades earlier could be adapted. A German physiologist named Carl Ludwig invented a “wave writer” or kymograph [3] that utilized a cylinder or drum around which a piece of paper could be attached (see Fig. 13.2). A stylus marked the paper as the drum rotated. Early versions used the stylus to scrape off soot from pre-smoked sheet but eventually ink became the preferred writing medium. In this manner, scientists could make recordings over longer time periods, especially when the stylus traversed a worm screw preventing overwriting after the drum completed a full rotation. An alternative approach involved moving a continuous strip of paper (from a roll or fanfolded) under ink pens at a constant speed. The drive mechanism typically incorporated several gear ratios that could move the paper ribbon or strip with rotating sprockets, pinch rollers, or some other mechanism.

Gibbs and Garceau were able to construct a one-channel EEG machine by adapting the Weston Union Morse Code inkwriting undulator while Albert Grass who was working on earthquake seismographs created a three-channel EEG machine. After 1938, EEG rapidly gained popularity as both a research and clinical tool. Polygraphic recording devices (e.g., the grass model 1) became commercially available (see Fig. 13.3). Groups in both Europe and America applied this new approach to psychiatric and neurologic disorders, describing, documenting, and naming various waveforms. Some recordings were made during sleep by sampling EEG and researchers described periods of low-amplitude mixed-frequency activity alternating with high-amplitude slow waves. Gibbs and Gibbs also remarked that EEG was mostly symmetrical during sleep, except during the low-voltage, mixed-frequency episodes [4].

Alfred Loomis was an eccentric, and very wealthy, lawyer-turned Wall Street tycoon who established a research center in Tuxedo Park in the 1930s sometimes called the Loomis Laboratories [5]. Loomis excelled at mathematics and had a passion for science. He spent the fortune he made through investments to fund scientific research by the greatest minds in his time (including Heisenberg, Bohr, Fermi, and Einstein). In 1937, Loomis, along with Harvey and Hobart, published the first report from continuous, all-night studies of human sleep [6]. They had constructed an 8-ft.-long, 44-in.-circumference drum recorder (see Fig. 13.4) and monitored up to three channels at a time of signals switched through some sort of selector panel. Recording sites included (midline vertex, midline occiput, behind left and right ears, and just above and to the left of the left eye). They described a fourth channel that could be used to record signal markers, heartbeats , or respiration. They could manage 3.5 h of recording before paper needed changing. They wrote the following “…we have been able to establish very definite states of sleep which change suddenly from time to time, and to correlate these with movements, with dreams, and with external stimuli applied to the sleeping subject.” They described alpha, low voltage, spindles, “random,” and spindles plus random sleep state sequences. Interestingly, they recorded and described eye movement activity but did not establish an eye movement dreaming state of sleep. Their graphical depiction of sleep’s progression across the night looks remarkably similar to our contemporary hypnograms (see Fig. 13.5). Loomis invited Harvard professor Hallowell Davis and his wife Pauline to TuxedoPark and offered to fund his clinical EEG research. Davis accepted the offer. Loomis had a hopeful but ultimately misguided belief that EEG could be applied to psychological and psychoanalysis and he invited many distinguished psychiatrists to his laboratory. When World War II broke out, the Loomis Laboratories changed direction to help in the war effort. He helped by bankrolling the development of high-powered radio detection and ranging (RADAR) and Loomis radio navigation (LRN), later renamed long range navigation (LORAN); technologies that ultimately helped win the war (Fig. 13.6).

In 1953, Aserinsky and Kleitman published data showing dreaming occurred during the sleep stage associated with low-voltage, mixed-frequency activity [7]. Furthermore, rapid eye movement (REM) accompanied these events. In his book Sleep and Wakefulness, Kleitman describes the discovery in the following manner: “…we literally stumbled on an objective method of studying dreaming while exploring eye motility in adults…” [8]. He further noted that Ladd (in1892), based only on introspection, posited eye movements during dreaming and Jacobson (in 1937) had written “When a person dreams…most often his eyes are active.” Nonetheless, the electrophysiological evidence in their landmark paper officially ushered in a key step toward more fully describing sleep. Meanwhile, Jouvet found his cats lost muscle tone during “paradoxical sleep” episodes which completed our picture of stage REM sleep [9]. These discoveries also marshaled the beginning of multichannel electrophysiological recording of EEG, electrooculogram (EOG), and sometimes electromyogram (EMG) for studying sleep.

Exploring REM sleep and comparing its mental and physiological correlates to other stages of sleep became a standard paradigm. Dream content, both spontaneous and provoked, during REM represented a very hot topic. After all, Freud had called dreaming the royal road to the unconscious mind [10]. Sleep studies targeting stage REM now provided a laboratory tool to explore the unconscious processes. Concomitant recordings of respiration, heart rhythm, electrodermal activity, paved the way toward what we now call PSG .

As more investigators began researching sleep phenomena, scientists realized that a standardization of techniques could immensely facilitate communication and possibly progress. After REM sleep’s discovery, summarizing sleep integrity, continuity, and architecture required more sophistication that the Loomis system offered. Thus, research groups created their own terminology and scoring rules. Dement and Kleitman evolved a system to classify sleep according to EEG criteria [11]. Essentially, a time domain of the set duration was established and the dominant activity within that time-frame-dictated classification . This time domain, called an epoch, could be classified as sleep stage 1, 2, 3, or 4 or awake according to EEG characteristics. When REMs occurred during stage 1 sleep, the stage was designated REM sleep. However, recording technique and terminology differed between sleep laboratories. REM sleep was also called paradoxical sleep, desynchronized sleep, active sleep, D sleep, and even unorthodox sleep.

Ultimately, an ad hoc committee was formed to develop “A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects.” [12] Chaired jointly by Drs. Allan Rechtschaffen and Anthony Kales the committee was a who’s who of sleep experts, including William C. Dement, Michel Jouvet, Bedrich Roth, Laverne C. Johnson, Howard P. Roffwarg, Ralph J. Berger, Allan Jacobson, Lawrence J. Monroe, Ian Oswald, and Richard D. Walter. The group standardized procedures, scoring rules, and terminology. More important, they reached consensus and thereafter used the procedures (for the most part) to conduct human sleep research. In essence, this was the real key to R&K (as it was called after the chairmen’s initials) successes. My understanding from stories I have been told by various participants indicated there were very strong (verging on violent) disagreements, shouting arguments, and name calling. In the heat of one less than civil debate, reportedly one of the chairmen barred the doors, decreeing that no one could leave until consensus was reached. Ultimately, consensus was reached and this is why R&K succeeded.

PSG, at least for sleep staging was established!

This new field, PSG , was ripe for discovery. In 1961, a research society emerged called the Association for the Psychophysiological Study of Sleep (APSS) and annual meetings were held to discuss research findings. The meeting proceedings were originally published in the Psychophysiology (the journal for the Society for Psychophysiological Research) which sported the image of a PSG tracing on its cover. Research laboratories at major universities, at US Naval and Army research centers, in European research institutes, and large medical centers began recording human sleep to describe its biology and to try to understand its underlying process. Whether a particular line of inquiry aimed to elucidate the function of sleep or to derive the intricacies of its process, progress, and some breakthroughs were being made. Sleep deprivation was a popular paradigm in the attempt to infer function. The examination of sleep stage rebound in response to selective stage deprivation provided a model to explore sleep’s homeostatic regulation. The military was interested in the adverse effects sleep deprivation had on behavior, attention, cognition , and physical ability.

Many studies focused on REM sleep phenomena. Obviously, dreaming represented a very seductive and intriguing topic; however, no unified theory of dreaming grew out of PSG research. Nonetheless, scientists found the eye movements generally corresponded to direction of gaze in the dream sensorium, middle ear muscle activity concorded with dreamt sounds, and nocturnal erections accompanied REM sleep but did not correspond to sexual dream content. Physiologic events associated with PSG stages other than REM sleep included timing and release of growth hormone during slow-wave sleep , slow-wave activity as a marker for homeostatic drive, other endocrine secretions in sleep, and physiological stability of cardiopulmonary functions during non-REM (NREM) sleep. Note the intuitive appeal of referring to sleep as REM and NREM. However, this terminology highlights how enamored polysomnographers were with REM sleep. REM sleep is the minor partner in sleep, yet we label the majority of sleep as not REM. Perhaps we should refer to wakefulness as non-sleep.