Despite the great potential of motor neuroprosthetics, the technology performs poorly in real world situations (Makeig, Kothe, Mullen, et al., 2012).The primary obstacle to high-functioning performance, I argue, is the mind. For one, the psychological mechanisms which underlie motor intention are poorly understood (O’Shea & Moran, 2017). Even more fundamentally, the mind as a whole — consciousness and unconscious processes — remains a mystery. The brain science community (through no fault of their own) is in the dark when it comes to a clear understanding.  No one can answer the most basic questions. What exactly IS the mind? What are its main components? How do these vary — across people, time, situations, life circumstances, and measurement conditions? How are the mind’s components — i.e. mental processes — connected, functionally and structurally? What physical form does the mind take, inside the brain?

In short, the exact nature of the mind’s relationship to the brain is poorly understood (Bassett & Gazzaniga, 2011). There are a number of very difficult problems standing in the way of defining the mind and mapping it to the brain (Poldrack & Yarkoni, 2016).

In other words, there is no (agreed-upon) neural correlate of the mind. This is mainly caused by a poorly-defined mind. How can you accurately identify the neural correlate of an entity (a state of mind or mental process) that is poorly-defined?

This makes logical sense. How could the neural networks that represent a specific intention (ex: “pick up THAT glass, quickly”) be defined, without listing the components involved? These components include the perception, and intention, of “arm & hand, their motion toward a glass, move hand to the right, fingers in grasp position, and grasp.” Without these components you MAY inadvertently hone-in on a particular set of neural networks; but without understanding, or being able to define them. You are left with a set of undefined, vaguely defined, or inaccurately defined networks.

The reasons for the lack of mind understanding are many. One is the blind spot of the brain science community. The topic of the mind is almost totally ignored. The answer to the question “how does the mind or human consciousness manifest inside the brain?” is seen as a remote goal to be achieved in the distant future. Therefore neuroscience has put the issue on the back burner, and turned its attention to the brain. The decision to forego a serious investigation of the mind is entirely rational. Why pursue a problem where little if any progress is being made? Also, the current efforts to understand the brain ARE making (extremely slow but steady) progress. The hope is the continuing collection of brain data will some day (in the distant future) add up to a brain theory.

But, as I argued above, this is a vain hope. The only way to understand the brain is to first understand the mind.

References

Bassett, D.S., & Gazzaniga, M.S. (2011). Understanding complexity in the human brain. Trends in Cognitive Sciences, 15, 200-209.

Makeig, S., Kothe, C., Mullen, T., Bigdely-Shamlo, N., Zhang, Z., Kreutz-Delgado, K. (2012). Evolving signal processing for brain-computer interfaces. Proceedings of the IEEE, 100, 1567.

O’Shea, H., & Moran, A. (2017). Does motor simulation theory explain the cognitive mechanisms underlying motor imagery? Frontiers in Human Neuroscience, 17, 1.

Poldrack, R.A., Yarkoni, T. (2016). From brain maps to cognitive ontologies: informatics and the search for mental structure. Annual Review of Psychology, 67, 587.

Because consciousness and the mind are not well-understood, its connection to the brain is poorly-understood. Though it may sound implausible, Mind Brain Insights claims the solution to both problems already exists — in the form of the MA Method. The core idea is mind = a set of active general memories = a set of active functional neural network ranges.

The method rests on an expanded view of the human mind. Cognitive neuroscience funnels or compresses the mind into a brain-based view. The brain is said to “compute” or “process” (external and internal) events, to create the mind. The MA Method broadens this narrow view of mind to include an obvious, yet mostly ignored, phenomenon — external experience. The conscious mind, as it is expressed within the brain, encompasses all of experience.

Most people clearly understand a person’s state of mind includes the most salient, and attended-to, events in one’s conscious experience. These occur within his visual field, including his body at its center. This is how most people throughout the world (who are not brain scientists) view the conscious mind.

The brain continually makes copies of this field (ex: that apple), which initially form and subsequently shape general memories (ex: “an apple,” “apple taste,” “fruit” etc.). The idea human memories — both episodic and general — are formed by copies of our “field of experience” is also common knowledge.

In short, the mind is filled with general memory, built from everyday experience. The idea mind = general memory is the foundation for a more accurate understanding of neural networks. For example, whenever an apple is seen, a particular set of memories is triggered: “apple, apple taste, apple smell, food, grasp an apple, take a bite” etc. These can not only be listed, but connected, weighted, and labeled excitatory/inhibitory. At the same time the apple’s recognition, meaning, related thoughts, emotions, goals, attention, intention, etc. can also be defined — again as a general memory set. The mind during common daily activity is filled with memory (with an assist from the afferent signal & low-level perception).

The MA Method also extends beyond the cognitive neuroscience conception of the mind. The mind during everyday tasks is not limited to cognitive neuroscience categories such as executive control, reward system, and working memory. In fact the mind encompasses ALL human experience: all aspects of perception, recognition, meaning, thought, emotion, executive function, goals, intention, language, motor control, etc. and combinations thereof.

An obvious candidate for the neural correlate of an (active) general memory is a functional neural network. A FNN is a coordinated, local or global, population-level neural firing event. Active state general memories and FNNs share the same basic characteristics. Both are highly associative. As a memory activates, component and associated memories spring to mind. Trigger a FNN and many associated FNNs come to life. Also, both memories and FNNs act, simultaneously, as a receiver of similar signals and as a transmitter of its own signal. And, each is arguably the dominant feature of their respective (mind/brain) systems.

Similar FNN activity (ex: “the taste of an apple”), repeated over time, will form a range of expression. As a FNN range forms, so does a corresponding structural neural network (SNN). The latter stores the former when dormant, and works (along with a matching FNN) to recall & support it when active.

In short, once mental processes are defined accurately — mostly as a set of general memories — they can be mapped to the brain’s functional and structural neural networks. This is mind/brain mapping based on the MIND first, as it really exists during daily living, inside and outside the lab.

Let’s consider the mind during voluntary movement. An intention to perform a particular movement is a central feature of it. Intention is also a common mind/brain signal used to control a device. But what exactly IS a movement intention? Can it be defined with precision? As it fluctuates across people, situations, time, and other variables? And, what about the dozens of other signals (noise, signal, and both) active at the same time?

I argue it’s possible to optimize the definition of any movement intention — if this is done within the context of overall state of mind. A mind (and corresponding brain signal) signature can be defined that is less noisy — i.e. more representative of the target state. A signature can be defined that is easier to repeat across conditions, more desirable (for user and BCI designer), and overall more useful than existing (task-based) signatures.

State of mind, and corresponding brain signal, signatures, or “targets,” can be defined with accuracy and precision. For instance, a movement intention is always accompanied by some state of emotion & arousal (ex: “calm and focused”). Somatosensation and visual perception are also part of any movement intention. To characterize these states of mind as signal, or noise, you first need to define them.

In an ideal world, the user’s state of mind (ex: “I want to pick up my phone — now”) will control the device with skill, ease, and reliability. The key is to look beyond task-based definitions of mind, to the mind as a whole DURING a task.  The idea is to define the mind — its contents and function — across people, movement disorders, device types, situations (lab, home, work, devise/object spatial relationships…) and other variables.

The good news is this is do-able! The main components of a movement intention — and the mind overall — can be listed, weighted, and connected. And, its strongest associations can be accounted for (listed, weighted, and connected) as well.

It’s true that as technical equipment and other capabilities to gather information from the brain improve, signal classification will be enhanced. So will gathering more brain data and its ability to add to our store of brain knowledge.

However the biggest potential improvement by far involves the mind. There is great potential here. The human mind – our conscious and unconscious awareness — is currently neither well-understood nor well-defined. And its relationship or connection to the brain is equally murky. But what if the mind could be clearly and accurately defined, and mapped to the brain?

Consider if someone working on an applied neuroscience project, such as a BCI and motor neuroprosthetic research group were to classifying a brain signal of the user’s intention “reach forward to grasp a cup.” To begin, there is no consensus on what aspects of mind are best to try to classify. Should it be the intention “(me) reach forward? or “my arm reach forward?” Should associated motor imagery be the main target of classification? Should a prediction (I will soon reach forward?) be included? What about the accompanying perceptual experience and recognition (of “my arm,” its motion, and what it is reaching toward)?

Or, maybe an overall brain signal that (typically) occurs during this task should be the target of classification. In other words, maybe the best thing to do is to ignore the mind altogether and classify the signal based purely on the type of task being performed (task-based signal classification).

The reason definition of mind is critical to signal classification extends beyond choosing the right mind targets to classify. Mind definitions are also crucial to defining a given signal accurately. After all, no two states of mind (during movement or any other task, or resting state) are alike. Each will have perceptual, emotional, thought, intention, attention content that will differ across time.

In short, the mind is currently poorly-defined; which is a major shortcoming giving its tremendous variability. The good news: it can be defined accurately, given a correct mind/brain theory.

The user of a BCI device controls it via his or her brain signal. The (EEG or other neuroimaging) signal is analyzed, processed, decoded, and/or classified. The end result is an intended movement.

But, what controls the user’s brain signal? Some might say task determines the signal: its location, frequency, amplitude and other characteristics.

I argue task is a secondary variable. The primary variable affecting the brain is the mind. Brain is influenced heavily by task, for sure. But it’s also affected by the environment, measurement situation, recent life events (good and bad) and other life context. All of these variables affect the mind.

The user’s mind — perception, recognition, meaning, thought, emotion, language, goals, imagination, intention etc. — is affected by many things, including the task. The brain signal thus reflects the mind primarily, task secondarily.

If a user’s mind, or mental processes, are activated optimally, and the brain signal processed and classified accordingly, the desired action will occur. Mind Target Hit → Functional Neural Network Activity → Brain Signal → Processing & Classification → (Desired) External Action.

For example, imagine a user’s mind dominated by a “move the cursor to the left” intention. In this case the corresponding functional neural network activity, and signal, will be classified as “move cursor left,” triggering that action.

What if the mind could be defined more accurately? This would allow more accurate brain signal labeling i.e. classification. Better mind definitions would enable more accurate, precise, and comprehensive classification of these signals.

More importantly, a group could use more optimally-defined mind/brain signatures as “targets.” The entire team — BCI designer, tester, trainer, and user — could use these to improve their performance. For example for the user, the targets would be more clear, precise, and personalized; desirable for the user to strive toward, easier to hit (strongly and consistently), and adjusted for specific movements, situations, and other context.                                                                                                        

The contents of the conscious mind, contrary to what many brain scientists might say, are not a mystery. They are our experience of life. All of our awareness, moment by moment, is encapsulated by our mind, as it changes (or stays the same) inside the brain. The good news is these contents can be defined — or at least summarized – quite accurately!

The mind can also be seen, more precisely, as a set of active general memories. For example, when a user tries to moves a cursor to the left, a “me moving a cursor left” memory will activate. This memory is comprised predominately of (visual & somatosensory) perception: of “my arm & hand moving to the left,” a computer screen, a cursor (a range of cursor size & shapes), and cursor moving left (a range of cursor speeds and directions). These mental content ranges will activate as part of the larger memory “me moving a cursor left” (itself a general memory).

A memory range is the same thing as a general memory. For example the general memory “an apple” includes a range of apple characteristics: shapes, sizes, colors, tastes, etc.

The neural correlate of a general memory, I argue, = a functional neural network (FNN) range. Both are highly associative, function — simultaneously — as signal transmitters and receivers, and dominate their respective (mind/brain) systems.

Overall, a state of mind = (mostly) an active general memory set = (mostly) an active set of FNN ranges = (most of) the corresponding brain signal.

The current cognitive neuroscience framework, despite the best efforts of professionals more skilled and knowledgeable than myself, is very limiting when applied to BCI development and use. Working within it yields vague and inaccurate brain signal targets. There is no one’s fault. The brain sciences are comprised of smart, talented people. The problem is they are laboring under the wrong (cognitive neuroscience i.e. mind/brain) paradigm.

Current mind brain targets for BCI control are either ill-defined, or too task-focused. A target might be “move the cursor left” intention (not a bad start, but very incomplete), or “attend to and do whatever you did before to make that action happen” (vague and ill-defined). Or, the target might be 100% task-based, such as “whatever the brain does when the user performs movement X, under conditions Y and Z.”

Task-based targets are sub-optimal because they are much less strongly correlated with the brain signal than the mind. A given state of mind includes not only task influences, but those of environment, situation, recent life history, stage of learning, and other context.

Contrast brain signal labeling based on task with labeling based on mind, defined as a set of general memories. For instance, the intention “move the cursor left” will include the memories “cursor,” “a computer screen,” and “move left.” This intention — a general memory in its own right – is itself comprised of general memories. These are mainly visual and somatosensory. Among these are “seeing an object move left,” and “the feeling of my right hand & mouse as it moves left.”

The above memories also have strong associations. These include attention (“visuosomatosensory focus on the cursor”), visual imaging (“imagine it will move left”), and emotion (“feeling calm and relaxed”).

Why label the brain signal based on task when it can be more accurately labeled based on the mind? In fact, I argue it’s possible to define a memory set — its components and associations — for any common movement & its context. Movement context would include the person, occupation, lifestyle, movement situation (environmental, social, work/home/lab…), and stage of learning (beginner/intermediate/advanced).

Overall, more optimal brain signal targets — based on the mind not just the task — have the potential to greatly enhance BCI design, testing, learning, and use.