One particularly fascinating aspect of her research deals with dreaming as a mechanism for regulating negative emotion and the relationship between REM sleep and depression:

The more severe the depression, the earlier the first REM begins. Sometimes it starts as early as 45 minutes into sleep. That means these sleepers’ first cycle of NREM sleep amounts to about half the usual length of time. This early REM displaces the initial deep sleep, which is not fully recovered later in the night. This displacement of the first deep sleep is accompanied by an absence of the usual large outflow of growth hormone. The timing of the greatest release of human growth hormone (HGH) is in the first deep sleep cycle. The depressed have very little SWS [slow-wave sleep, Stages 3 and 4 of the sleep cycle] and no big pulse of HGH; and in addition to growth, HGH is related to physical repair. If we do not get enough deep sleep, our bodies take longer to heal and grow. The absence of the large spurt of HGH during the first deep sleep continues in many depressed patients even when they are no longer depressed (in remission).

The first REM sleep period not only begins too early in the night in people who are clinically depressed, it is also often abnormally long. Instead of the usual 10 minutes or so, this REM may last twice that. The eye movements too are abnormal — either too sparse or too dense. In fact, they are sometimes so frequent that they are called eye movement storms.

love dreams

Well, there are some clearly ascertainable facts, but then some inferior and reductive inferences eat them for lunch.  First some of the interesting facts:

"But memory is never a precise duplicate of the original; instead, it is a continuing act of creation. Dream images are the product of that creation. " and,

"Dreams are not without sense, nor are they best understood to be expressions of infantile wishes."

And then the reductive inferences:

"They [dreams] are formed by pattern recognition between some current emotionally valued experience matching the condensed representation of similarly toned memories. Networks of these become our familiar style of thinking, which gives our behavior continuity and us a coherent sense of who we are." 

Really? So somehow I pattern tone match all my memories and relive those experiences every night in some boring, appropriately condensed after school special rendition of who I think I am or should be?

"Thus, dream dimensions are elements of the schemas, and both represent accumulated experience and serve to filter and evaluate the new day’s input."

I don't know how to tone regulate any concluding "thus" but such blather is not well-represented in my dream dimensions.  

I can share my dreams are not experienced nor recalled as either tone or valued experience matching at either a conceptual or emotive level.  I recall many different and divergent experiences and felt senses in dreams as compared to my waking states.  

Perhaps we can only imagine what others could share in recalling their dreams--especially those with the accomplished skill of lucid dreaming.

But maybe I'm missing the point here... as I'm not sure that REM regulates emotions as much as it is simply implicated as an observable phenomena in an otherwise organic, creative and holistic experience that can be either emotionally rich or poor and many standard deviations away from our current illusions of Life.