Dreams are a window to our memories

Updated: 3 days ago

Photo credit: Andrew Ostrovsky

Summary: To understand dreams, we need to explore why we sleep. As building blocks for why we sleep, we need to start with a description of how we think. This article builds on context from neuroscience and psychology. Several examples and digging deeper resources are provided along the way.

Most simply, sleeping is an important input to the thinking maintenance process, with dreams providing an observation window into that process. But dreaming is not just a thinking process bystander. Dreaming is like the proverbial "canary in the coal mine." Not reaching a dream state suggests there is a thinking process problem.

In this article, we build an understanding of why our thinking maintenance process requires sleep, including disease and related impacts resulting from not having proper sleep hygiene. We then present and explore our dream theory. Dreams provide an observation window to our memory consolidation, long-term memory transition, and synaptic development process. Reaching the dream state is an indication our thinking maintenance process is healthy. Not to worry, just because we do not remember our dreams, does not mean the thinking maintenance process did not occur.

Table of Contents:

  1. Summary

  2. Thinking

  3. Sleeping

  4. Dream Theory

  5. Conclusion & Notes

Extras - Deeper dives, Examples, and Analogies:

  • Brain biology deeper dive

  • The Flow State paradox

  • A call to align higher education with how we think

  • Building memory by building many synaptic connections

  • Cramming for an exam - A counterproductive memory development example

  • A cognitive bias and reasoning error analogy

These are extra resources intended to help deepen your understanding and pique your curiosity.

2. Thinking

The brain is an extraordinary and complex organ. People are born with about 80 billion neurons. Think of neurons as microscopic buckets holding biochemically activated electric signals. As well, at birth and especially earlier in life, we generate approximately 100 trillion synapses. Think of synapses as the network lines that connect the buckets. Together, neurons and synapses are the building blocks for thought, like putting together a sentence or completing a math problem. The basis for learning is the creation of new synaptic connections. This is known as neuroplasticity. Also, while the rest of your body will weaken as we age, neuroplasticity enables a healthy brain to strengthen throughout one’s life, all the way until one dies. In fact, many philosophers and scientists relate learning to life itself.

“Wisdom is not a product of schooling but of the lifelong attempt to acquire it.”

- Albert Einstein

“No problem can withstand the assault of sustained thinking.”

- Voltaire

Throughout this article, we will discuss approaches to encourage neuroplasticity and healthy brain habits. This helps maintain a sharp brain that will serve us well to the end of our days.

Let us start with a simple exercise - try writing your name with the opposite hand from which you normally write. It will likely feel odd to hold the pen in your offhand. Then writing itself will take much concentration. Every letter will need to be considered. You will likely be disappointed with the quality of the penmanship. Next, write your name with your offhand 10 times in a column on a single sheet of paper. Notice that each time you write your name it gets a little quicker and the quality increases. This is neuroplasticity in action! You are quite literally enabling new synaptic connections by learning something new. Learning to write well with your offhand will take significantly more practice, over many weeks. In the next section of the article, we will describe the process by which your brain learns and improves via synaptic growth. By the way, many neuroscientists believe we may decrease the chances of being affected by brain disease by ongoing synaptic growth. [i] So, learning something new on a regular basis will help you outrun brain disease!

You spend much of your waking day processing information. Some of your daily processing includes new, synaptic growth-based information. However, most of your mental processing occurs along existing synaptic pathways. Take driving as an example. For an experienced driver, driving “feels” very easy. They have practiced this activity so much that the neural pathways have become deeply habituated. Thus, while "thinking about driving" is occurring, it feels easy, almost automatic. The act of driving is mentally executed on highly habituated mental pathways. Next, we describe our brain's biology and habituated v. new learning.

Brain biology deeper dive: Our brain’s primary logic-seeking analytical center is known as the cerebral cortex. Our cerebral cortex is physically and logically divided into the right hemisphere (RH) and the left hemisphere (LH). Think of our LH as a serial processor - it handles:

  • Past information, often from long-term memory;

  • Current information, often from short-term memory; and

  • Creates future forecast information, to help us make future judgments.

Our LH also tends to be slow, given the volume of information needing processing and especially the information needed from our long-term memory. Our RH is different. Our RH is connected solely to the present. Think of the RH as a parallel processor, able to input information from our 5 senses as processed via our short-term memory. Our RH relies on readily available habituated information. Our RH tends to be fast. In fact, very fast. From a long-term evolutionary standpoint, our brain evolved to be our protector from danger. For example, our RH evolved from the danger of wild animals or rival tribes of many millennia ago. It is our RH that responds quickly to brake when an animal darts in front of our car. You may thank our brain’s evolution for the fact that we are alive to read this article today! Based on the driving example, a well-trained and experienced (highly habituated) driver may enable fast memory access via the RH. Other than heavy traffic (a situation requiring the LH), the driver may rely upon their RH. [ii]

Much of our day-to-day mental processing is more automatic. Thus, few new synaptic connections are created with habituated activity. A minority of our daily mental activities will include learning something completely new. This newly learned behavior will likely “feel” difficult and challenging. This new activity will create many new synapses, including synaptic development while we sleep. We will discuss why we sleep in the next section.

The Flow State paradox: Reaching flow state is where our brain exists solely in the RH for an extended period of time. Some high-level athletes reach a flow state. They discuss “losing time” as the world around them slows. Even a driver may reach an almost hypnotic state where they seem to lose time in their journey. Some religions call this an enlightened state or “nirvana.” In the Pulitzer Prize-winning book, Goedel, Escher, Bach; the author, Douglas Hofstadter, makes the comment:

“Apparently the master wants to get across the idea that (Zen) enlightened state is one where the borderlines between self and the rest of the universe are dissolved."

One would think we may always wish to reach a flow state. It turns out that reaching a flow state is the outcome of highly disciplined LH-based training. The flow state is generally temporary, requiring both unique internal and environmental conditions. Most important, building synapses generally requires LH-based training. The flow state is the ultimate outcome of the habituated brain. Thus, even if we could remain in a flow state, we would be reducing our ability to develop new synapses. Thus, while the flow state may be “nirvana” it is our slow LH that enables a periodic visit to the flow state and provides for synaptic growth. For a deeper dive into reaching a flow state, please see our article: Soccer Brain - the making of the beautiful game.

All this mental processing sends electrical impulses between neurons and is transmitted via the synaptic connection between neurons. As transmitted via the synapses, much of the electric signals are associated with information signals generated from neurotransmitters. [iii] (the neurotransmitters are represented by the yellow dots in the diagram) As generated from our cognition, there is also a small amount of a waste peptide created. This peptide is named Amyloid Beta (“Aβ” is represented by the red dot in the diagram).

This Aβ byproduct occurs for a similar reason that your muscles create lactic acid after a big workout. By the way, lactic acid is what creates that sore feeling after exercising a muscle. Think of Aβ as the necessary byproduct after exercising the brain. In your synapses, you have microglia to sweep away the Aβ byproduct from your thinking process. The challenge is that sometimes your microglia may not be sufficient to sweep away all the Aβ.

3. Sleeping

A very important reason mammals sleep, particularly humans, is to help the brain cleanse the excess Aβ built up in the synapses. The microglia do not always completely cleanse all the Aβ. When you sleep, specifically in your deep, delta wave sleep, your brain releases spinal fluid to wash away excess Aβ and to help the microglia finish its job of cleaning away the Aβ. Human brains:

  1. Have significant neuron and synaptic capacity and

  2. Are very active in their daily usage of this mental capacity.

As such the need for daily brain maintenance to ensure the brain’s long-term well-being is increased compared to other mammals. Not cleansing the Aβ is the basis for many brain diseases, including Alzheimer’s disease. Excess Aβ may eventually bind together to form Amyloid Plaques (“AP”). AP may eventually overwhelm the synapse and cause synaptic death. [iv] A significant amount of synaptic death will likely cause Alzheimer’s disease symptoms. Thus, sleep is a critical activity to ensure long-term brain health.

In a healthy brain, some synaptic death is normal. No matter how good our sleep hygiene is, most people can expect some synaptic death throughout their lives. It has been shown that active learners are not as symptomatic of brain disease as compared to less active learners. Even though the active learner brains may present synaptic death. This enables the new synaptic activity to effectively “outrun” synaptic death by rewiring our neural network around the AP-impacted synapses. This means the volume of the new learning will generate enough new synapses to offset the effects of the synaptic death associated with AP. [v]

Confirming research suggests that sleep hygiene, along with learning, healthy exercise, and good eating behaviors, is a critical personal mental health activity. Broad-based neuroscience literature reviews suggest that lifestyle is the critical determinant of long-term brain health. [vi] As an analogy for why good sleep hygiene and other healthy habits are important to prevent brain disease, neuroscientist and author Lisa Genova said:

“Think of amyloid plaques as a lit match, at the tipping point [of brain disease], the match sets fire to the forest. Once the forest is a blaze, it doesn’t do any good to blow out the match.”

Another critical reason we sleep is to process our daily mental activity for the purpose of categorizing and strengthening the synaptic growth associated with daily learning. Generating synaptic connections is an associative activity. Like in the handwriting example, the muscles, balance, language, and visualization of training the offhand to write will generate new synapses. These new synapses will be associated with the existing handwriting-trained neurons. Plus, the new synapses will associate with other related existing neurons. (Like neurons associated with your offhand, with holding a pen, etc.) Thus, new synaptic connections were made to accommodate the newly learned behavior.

Also, the volume of training for the newly learned behavior matters. As we all experience in our daily lives, the more we practice something, the better we become. This is the process of strengthening synaptic development via repetition. This is a key aspect of the neuroplastic processes to create and strengthen new neural synaptic pathways. As an important nuance, the learning associated with creating new synaptic connections is challenging. It is far easier to do something we know versus something we do not. But only traveling down the same mental road repeatedly will not create new synapses.

A call to align higher education with how we think: The extraordinary amount of brain capacity, along with the significant differences in childhood educational opportunity, suggests all people are incredibly neurodiverse. Sure, we all share a genome with a similar starting point and we mostly share the neuron/synapse brain capacity necessary for college success. Traditionally, U.S. high school graduates are about 17 years old. It is also traditionally assumed the 17-year-old is ready for college. However, given our significant neurodiversity, the tradition of universal college readiness at 17 years old is flawed. Even those attending college, many have different learning needs. Only a minority of students in the U.S. who start college are able to graduate, get a good job, or not default on a student loan. This suggests the current U.S. higher education system is not working to support the majority of our citizens. The U.S. has a significant opportunity to use the latest technology and our understanding of neurodiversity to revamp higher education in a way that delivers the “3-D’s”:

  • De-bundles - academics from the other typical college amenities

  • De-leverages - by greatly reducing student debt burdens.

  • Delivers - high-quality, technology-enabled, and cost-effective academic services.

At its core, the concept uses an approach called “Flip the classroom” –

Traditional: In the traditional higher education approach, the professor spends time lecturing many students, with less time for individual assistance. Homework is assigned as problems to help the student learn the content. Homework may also include tutoring, self-learning, or one-on-one office hours with the professor. Time, generally the semester, is the operating standard for class completion.

New: This new approach will facilitate “flipping the classroom,” where the teacher assigns a lecture and reading for homework and spends class time tutoring or being available for individualized learning. Time ceases as the operating standard. Once subject mastery is reached, the class is considered complete - regardless if it takes 2 weeks or 2 years.

“Formal education must change. It needs to be brought into closer alignment with the world as it actually is, into closer harmony with the way human beings actually learn and thrive.”

- Sal Khan, Founder and CEO, Kahn Academy

To dig deeper, please see our article Higher Education Reimagined.

The essence of healthy brain life is learning the mastery associated with a strong synaptic connection, then pursuing something new and challenging. In other words, work on learning something new until it goes from being a challenge to being automatic. Then it is time for the next challenge. Plus, all the while, having good sleep hygiene, along with the other healthy behaviors mentioned earlier, supports synaptic development for new learning.

In the next section, I present my theory on an additional purpose of sleep and how dreams are associated with this sleep purpose. Like any theory, ongoing testing needs to be completed to validate this dream theory.

4. Dream Theory

Our days are spent performing various mental processing activities as discussed in the earlier “Thinking” section. Some of those processing activities are new and require categorization for long-term memory. It has been shown that we have different memory storage types, from the most short-term to the most long-term. [vii] Think of sleep as the means for converting our more temporary short-term memories into more permanent, recallable long-term memories. The quality of our sleep relates to 1) more efficiently processing daily experiences and 2) creating long-term memories. Importantly, sleep helps us to “connect the dots” between memories, both newer memories from the day and existing memories. The creation of additional synapses is like the connection of another network branch in our brain’s neural network. The strength of the memory and our ability to recall is based on the number of synaptic connections to like memories.

Building memory by building many synaptic connections: If all you have ever learned about Jeff Hulett is that he wrote the article "Dreams are a window to our memories," then you likely will not remember Jeff Hulett. However, if you have also learned that Jeff Hulett is married to Patti Hulett, Jeff has 4 children, Jeff is a behavioral economist, Jeff is a software and services company executive, Jeff leads a nonprofit, and Jeff declares himself a Christian but he left the Catholic Church; then you are much more likely to remember Jeff Hulett. Each of those disparate facts about Jeff, when committed to long-term memory, will create a separate synaptic connection. Additional synaptic connections increase the likelihood and speed of long-term memory recall.

On the flip side, lack of sleep will reduce your ability to commit those short-term memories to your synapse-connected long-term memory. Unconnected short-term memories will get crowded out by newer short-term memories and ultimately lost.

Cramming for an exam - A counterproductive memory development example: College students are famous for “pulling all-nighters.” This is characterized by an exam scheduled for tomorrow and the student staying up through the night to study. The ability to stay awake is likely enhanced by caffeine or related stimulants. The challenge is that without sleep, the learning associated with the ”all-nighter” will not be effectively processed, categorized, and made available for the synaptic processes associated with the subject studied and long-term learning. The student’s ability to perform well on the exam will be limited, perhaps only to uncategorized, unconnected learning that may still exist in their short-term memory. Think of a stack of 100 randomly ordered pieces of paper. Properly ordered, those pieces of paper may mean something, like a story. However, without order, they are mostly confusing noise. Most of the short-term memory will lack context. Without proper synaptic or neuroplastic-related processing, the information will be mostly lost and not committed to one’s synaptic neural network. Also, any existing learning, properly processed earlier in the semester, should be available. Although a tired brain does not typically perform memory recall as well as a rested brain. As such, my advice: Go to every class and study as you go. Study enough for your final ahead of the exam, enabling you to get a good night's sleep the night before the final. If faced with an “all-nighter” or “sleep” tradeoff choice, I would prioritize getting a good night's sleep.

Generally, your day’s experiences are a jumble of information that lacks the strength of connection to existing long-term memories. This is not to say that some of your day’s experiences were not connected to long-term memories. Very likely some were. But connecting an immediate experience to long-term memory requires a conscious process that is not available for most of your day’s experiences. This is because, given the extraordinary volume of sensory input received in our awake life, there are simply too many discrete experiences to process and commit to long-term memory. Therefore, your short-term memory is holding an inventory of your day’s experiences and waiting for long-term processing. Sleep is how synaptic processing and related neuroplastic processes occur. In sleep, the experiences found in your short-term memory, many of which were created that day, are available for synapse creation or synapse strengthening to associate with like neurons.

Dreams are like a window into that synaptic creation process. Imagine every memory is like a picture or a short video. The collection of memories is kept in an inventory, but in no particular order. As a matter of practice, a dream will randomly get seeded with a single piece of information stored in short-term memory. From there, the dream narrative is built by rearranging the remaining information available to short-term memory.

Even in sleep, our brains have the amazing ability to make sense of the world, even our discrete, jumbled experiences as found in our short-term memory inventory. The brain's instinct to seek causality is the same instinct enabling cognitive biases and related reasoning errors. In the Black Swan, N.N. Taleb notes:

"Our minds are wonderful explanation machines, capable of making sense out of almost anything, capable of mounting explanations for all manner of phenomena."

A cognitive bias and reasoning error analogy: Confirmation bias is a type of cognitive bias. This is where your brain naturally weighs information confirming an existing belief more than information contrary to an existing belief. Dream formation is also consistent with a reasoning error known as an error of omission. An error of omission is where one's reasoning utilizes only a subset of available information. In this kind of reasoning error, the remaining information needed to make an accurate reasoning conclusion is omitted. Whether our brain is in a dream state as part of the thinking maintenance process or is reasoning subject to cognitive biases or reasoning errors, all these brain situations work under the same set of rules as evolved from our genome. Thus, our evolutionary-based need for causal ordering is at play in key aspects of our brain functions. Also, the comparison of a "bias" or an "error" type to a dream is appropriate because a dream, like a bias or error, is not an accurate representation of reality.

Dreams are an assortment of those discrete memories rearranged to tell a story. The discrete memories found in the short-term memory inventory may have occurred that day and some may have happened in the more distant past. Even while asleep, your consciousness is observing this synaptic creation process and seeking to make sense of the jumble of pictures and short videos it perceives as your brain processes the jumbled memory inventory while sleeping. This is why dreams seem authentic and original. Dreams are a unique rendering of reordered memory snippets. Your consciousness is viewing a window into the long-term memory creation process by observing and seeking to tell a story with the jumbled inventory of discrete short-term memories. As presented in the cramming for an exam example, think of your consciousness as viewing those 100 random pieces of paper. The dream will reorder the stack of papers in a way that tells a fanciful story.

Conclusion & Notes

Dreams are part of our overall thinking process. Dreams are like a byproduct, that provide our consciousness a glimpse of the extraordinary synaptic development and strengthening process. This process converts our short-term memories to long-term memory storage. The strength of long-term memories and speed of conscious recall is a function of the synaptic volume and synaptic strength. Sleep is a critical part of learning and synaptic growth. A lack of sleep is related to insufficient synaptic development and brain diseases such as Alzheimer's disease. A lack of sleep is generally counterproductive to cognition.



[i] Nahum, Lee, Merzenich, Principles of Neuroplasticity-Based Rehabilitation, Progress in Brain Research, Volume 207, 2013, Pages 141-171

[ii] We provide more explanation for brain biology in the following articles:

Hulett, Brain Model, The Curiosity Vine, 2020 Hulett, Soccer Brain - the making of the beautiful game, The Curiosity Vine, 2020

[iii] We provide more explanation of neurotransmitters in the following articles:

Hulett, Brain Model, The Curiosity Vine, 2020

Hulett, Origins of our tribal nature, The Curiosity Vine, 2022

Hulett, Soccer Brain - the making of the beautiful game, The Curiosity Vine, 2020

[iv] Genova, What you can do to prevent Alzheimer's, TED Talk, 2017 [v] Snowdon, Aging and Alzheimer's disease: lessons from the Nun Study, National Library of Medicine, 1997

[vi] Toricelli, Pereira, Abrao, Malerba, Maia, Buck, Viel, Mechanisms of neuroplasticity and brain degeneration: strategies for protection during the aging process, Neural Regeneration Research, 16(1), 58-67, 2021

[vii] Camina, Guell, The Neuroanatomical, Neurophysiological and Psychological Basis of Memory: Current Models and Their Origins, Frontiers in Pharmacology, 2017

109 views0 comments