Brain activity frequently resembles awake behaviour during during rapid eye movement (REM) sleep. The brain can actually be more active during REM sleep than it is when you are awake.
This activity is why REM sleep is sometimes referred to as “paradoxical sleep,” according to Virginia Tech neuroscientist Sujith Vijayan. And for those suffering from post-traumatic stress disorder, this very active sleep stage is rife with emotionally charged dreams over and over.
The sleeping brain can typically bring up emotional memories, process them, and remove their emotional charge. It’s a process that could be explained by an evolutionary drive to assess important memories, including those associated with fear.
However, in patients suffering from post-traumatic stress disorder (PTSD), the brain appears to bring bad dreams back night after night, driven to evaluate fear memories but unable to ever remove their emotional charge.
So, the question is, what keeps the brain in that loop?
Norepinephrine and Serotonin
Vijayan led a team in developing biophysically based models of the sleeping brain in order to investigate at a deeper, mechanistic level how the brain may — or may not, in the case of PTSD — effectively process and extinguish fear memories.
The neurotransmitters norepinephrine and serotonin typically decrease during REM sleep. The researchers used their models to link low neurotransmitter levels to the brain’s ability to inhibit fear expression cells via rhythms sent between the prefrontal cortex and the amygdala.
Vijayan’s team then investigated how the atypical neurotransmitter levels seen in the sleeping PTSD patient’s brain could disrupt the fear-reducing work of those brain rhythms.
Neurotransmitter levels remain elevated during REM sleep in people suffering from PTSD. The models developed by the team show that under these altered conditions, brain rhythms that are normally effective in healthy people can no longer inhibit fear memories.
Instead, the brain of a PTSD patient appears to require higher-frequency rhythms to extinguish fear memories. The researchers believe that unlocking those higher frequencies could inform therapies to restore the restorative quality of sleep to people suffering from PTSD.
REM Sleep Rhythmic Interactions
Vijayan, who has long studied how sleep affects learning and memory, describes REM sleep as a kind of “Wild West” in terms of what we know about its relationship with memory.
Experiments have shown that rhythmic interactions between the amygdala and the prefrontal cortex during REM sleep reduce the expression of fear-related memories. However, it is unclear how this works.
With the majority of neuroscience research focusing on non-REM sleep, REM sleep is much more difficult to grasp.
“There are really good models out there for how non-REM sleep might consolidate memories and what role it might play in learning and memory. But when we talk about REM, there are no real, good models on how that stuff is happening,”
said Vijayan, an assistant professor in the School of Neuroscience, part of the Virginia Tech College of Science.
Inhibiting Fear Expression
Vijayan sought to change that by developing biophysically based models of REM sleep, allowing his team to learn more about what allows brain rhythms to aid in emotional memory processing and how PTSD interferes with it. The models enabled the scientists to manipulate the REM sleep conditions that they thought were important in answering this question: neurotransmitter levels.
Beginning with models of sleep conditions in a healthy individual’s brain, Vijayan’s team reduced norepinephrine and serotonin levels to represent REM sleep.
As a result, rhythmic interactions between prefrontal cortex neurons and amygdala neurons strengthened connections between those two areas. Brain rhythms from the prefrontal cortex effectively inhibited the activity of amygdala fear memory cells.
The researchers also discovered that a particular frequency of brain rhythms was especially effective at inhibiting fear expression cells. They discovered that lower-frequency theta rhythms of around four hertz most effectively strengthened connections between frontal areas and the amygdala when inputting frequencies in the theta range of rhythms typical for humans — four to eight hertz.
The researchers then simulated REM sleep in PTSD patients. It is well understood that PTSD causes norepinephrine levels to remain elevated during REM sleep.
When Vijayan’s team mimicked those conditions, they discovered that when brain rhythms at four to eight hertz were introduced, those rhythms could no longer suppress fear expression cells.
“I’m a little surprised that the four hertz didn’t work. I thought maybe it would still be effective, but it really wasn’t at all,”
Vijayan said.
Covert Auditory Stimulation
The team found that there is still hope for ending the cycle of nightmares fueled by fear.
Although the typical frequency range of brain rhythms, four to eight hertz, was ineffective at inhibiting fear expression cells, researchers tried other frequencies. In their PTSD model of the sleeping brain, they discovered that at a higher frequency of 10 hertz, brain rhythms could effectively inhibit those cells.
Vijayan believes that by identifying the atypical frequencies of brain rhythms that inhibit fear memories in PTSD patients, his team can develop therapies.
The next step is to determine how to adjust the brain rhythm frequencies of sleeping PTSD patients in order to achieve optimal rhythms. Using what Vijayan calls covert auditory stimulation, this is possible.
“That means I’m playing those sounds and you’re not aware of it while you’re sleeping,”
Vijayan said.
It might be helpful for any disorder where sleep is disturbed, including PTSD, traumatic brain injury, and Parkinson’s disease. The theory is that the restorative properties of sleep can be activated by inducing the desired neural dynamics.
Reference:
- Young-Ah Rho, Jason Sherfey, Sujith Vijayan. Emotional Memory Processing during REM Sleep with Implications for Post-Traumatic Stress Disorder. Journal of Neuroscience 18 January 2023, 43 (3) 433-446; doi:10.1523/JNEUROSCI.1020-22.2022