In his blog, Doug Rett, OD, FAAO, digs deeper into questions every eye doctor has had at one point in his or her career. He sees patients and works with students and residents at the optometry clinic for the VA Boston Healthcare System and will use his blog to bring the teaching points he sees in clinic to a larger audience. Contact him at

Disclosure: Rett has no relevant financial disclosures.

BLOG: In dreams begin – ocular – responsibilities

I recently read a hypothesis from Fitt and colleagues that claimed, “contrary to previous opinion, the purpose of REM during sleep is to ensure corneal respiration in the absence of the buoyant mixing that routinely takes place ... during waking conditions.”

This didn’t seem right to me, but admittedly sounded intriguing. Then I realized that as an eye doctor, I knew very little about the rapid eye movements (REMs) that occur during sleep.

Why do they happen? Are they random? I came into REM literature with a lot of questions. I left humbled and with a lot more questions. The literature involving sleep and its stages is fascinating, but very complex and dense. It’s also – as you might imagine – hard to test many of these hypotheses, what with the patient being unconscious and all. But we have responsibilities as eye doctors, so let’s try to learn about REM.

There are four stages of sleep: the first three stages are non-REM (NREM) and cleverly named stages 1, 2 and 3, with the fourth stage being REM sleep. Stage 1 is the lightest sleep; we are easily awakened, but muscle tone relaxes, and EEG-measured brain activity slows. Stage 2 is a deeper sleep, and awakenings are fewer. Brain waves continue to slow, the heart rate drops, and body temperature falls. In stage 3, brain waves are traveling very slowly, and it can be difficult to arouse someone from this stage. This is the most restorative sleep we have, but adults spend only about 5% to 15% of our sleep in this stage (Ermis et al.) as opposed to 40% to 60% of our sleep time in stage 2 (children and adolescents spend longer in stage 3 than do adults). After stage 3, we go into REM stage, which is where we do most of our dreaming.

REM sleep is also known as paradoxical sleep, deriving that name because our brain activity rises to near wakened levels of activity, but our muscles undergo atonia, rendering us close to paralyzed. Our bodies accomplish this by hyperpolarizing motor neurons, thus requiring a very large neuronal stimulus to overcome this threshold to depolarize the nerve. It’s thought that muscle atonia occurs to prevent us from acting out our dreams and possibly injuring ourselves (Hobson et al.). But why do the extraocular movements not undergo atonia? Why do they fire so aggressively in this phase?

Rapid eye movements are immediately preceded by something known as Ponto-geniculo-occipital (PGO) waves, so it stands to reason that these waves of neuronal activity in the pons, geniculate bodies and occipital cortex are generating REMs (Peigneux et al.). However everything in sleep science seems to be controversial (even the definition of dreaming is argued), and it’s unclear exactly how PGOs and REMs are related.

I’ll present two leading theories vis-à-vis how they relate to REM sleep saccades: the activation synthesis model and the activation-only model. The activation synthesis model argues that PGOs are started in the brainstem (Antrobus), and during REM sleep, enough chaotic brainstem activity eventually triggers a depolarization, and a PGO wave occurs, and a saccade is generated essentially from randomness.