Break it down, before breaking down!?

Hello good people!

I am back again. I spent last week with my family in the Australian country side. Hence, nothing from my in your inbox last week. It was cold, I can tell you that!

Anyway, today I want to look into a walk you through some of my research on the back of guidance I have received especially early in my career.

“If too complex, break it down!”

But is that really a solution? And if so, why? Let’s look at working memory, long-term memory and schemas, the brains’ energy distribution and how to use all of this for tackling complex tasks.

Let’s break it down.

Working memory

Working memory is a system in the prefrontal cortex, that allows for short-term storage and processing of incoming information.

It seems like the brain can hold a limited amount of information in the working memory. Depending on the individual, it can vary between 7-9 pieces of information for around 120 seconds. Maybe this is why most authentication codes for 2-factor-authentication have only 6 digits?

Within the working memory, information is stored through neurons that fire together in a specific way. In a fMRI, it looks like they light up together. Like Christmas tree lights. Over the time of the 120 seconds though, the signal fades more and more until the information is lost, unless the information is provided again, e.g. you look at the same number.

The left prefrontal cortex is more evolved in verbal working memory. The right prefrontal cortex is more involved in spatial working memory.

To summarize: working memory is finite. We can hold 7-9 pieces of unique information for about 120 seconds at a time.

After enough repetitions, information could get stored in long-term memory.

Activation of experiences

Information gets processed into long-term memories as schemas or a set of actions. These schemas are associated with specific triggers and act as energy-efficient mental frameworks. They are easily recallable - part of your mental toolkit for life.

Knowledge gets stored in long-term memory mostly during sleep, deep sleep in specific, which occurs predominantly during the first part of the night.

The brain seems to store neural location of these schemas depending on what is required: movement, knowledge or both together. It “queries” brain regions required to “pull” the necessary schemas, when triggered through external stimuli.

It is believed, that retrieving schemas for working memory is considered similar to one piece of new information. The only difference is that a “schema piece” of information can include more information in one “package”.

If schemas do not exist for a specific problem statement, the brain has to do a bit more.

Cognitive load: high performers vs. low performers

During an experiment, individuals were asked to perform mental memory-recall tasks while fMRI scans were done. The complexity of tasks, that is, numbers increased while the intervals between new numbers decreased over time. Interestingly, the more complex the task became, the more brain regions lit up during the fMRI scan for most of the people.

For low-performing individuals, more brain regions lit up the more complex the task became. Well-performing individuals remained with fewer engaged brain regions.

Simply put: the more brain regions are activated, the more performance dropped.

It seems that “high-performers” were able to recall useful schemas for the execution of the task, requiring less brain regions to solve the problem statement.

It is important to note that “high-performers” does not mean this group was smarter than the other group. They were mentally better equipped to do the exercise, probably because of prior similar experience compared to the “low-performers”.

Energy distribution

The last piece to look at is the energy distribution of the brain.

It seems like the brain continuously uses the same amount of energy all day long. There are very few spikes, even when we sleep. There the brain is busy with memory consolidation and other things. It is constantly working, consciously and subconsciously.

When we start to utilize our working memory and short-term memory, the brain diverts energy from ongoing activities to our memory recall functions.

There is no 10 watts vs 35 watts energy consumption. It is almost always around 20 watts. This is our 100%. The question is how the 100% is being distributed across our brain regions. The brain needs to analyze, how to keep you alive while solving that one issue at hand.

At one point, the brain sends a “tired” signal through your body if it believes the energy levels are out of whack to make you stop the mental load (e.g. writing a project plan) and get back into a more “balanced” state.

Let me summarize the points briefly.

1. Long-term memories and schemas are stored in the brain in various regions and can be recalled into working memory when needed. Schemas are a set of actions - learnt processes.

2. The working memory is limited. It can combine short-term memory and long-term memories and hold approximately 7-9 pieces of information at one go.

3. Better performance at one task most likely shows you have prior experience with that specific task.

4. Engaging in new tasks you do not have prior experience, the brain activates more brain regions. As the brains’ energy consumption is almost always at the same level, activating new brain regions means the “muting” others to save energy.

All of this basically describes the process around the “cognitive load” theory.

Now that we understood some basics, let’s go back to the initial question: why do we need to break down big chunks of work?

Breaking it down

Let’s consider an example, like developing a business plan across sales, marketing, and HR. My focus is sales and marketing, I enjoy Excel, I am not good at HR-related items.

1. I need to identify parts of the work I can solve through experiences, i.e. apply schemas. I need to be smarter about breaking tasks up and ask myself “what have I done in the past” vs. “where do I need help through research or other people”. I should tackle these separately

2. I should identify “similar” work and put them into 60 to 90 min work sprints. For example, instead of building the Excel design and formulas while entering sales numbers, one sprint could be only to think about sales numbers, disregarding everything else. Why? I want to avoid combining horizontal thinking (sales, hr, marketing and excel design) and vertical thinking (only sales). I want to engage similar schemas into my working memory. This is more efficient, ideally requires less brain regions in parallel.

3. I want to define work sprints that are achievable within 60-90 minutes. We have a natural rhythm, that e.g. reflects in our sleep as well; our typical sleep cycle is 90 minutes for all phases. We go through similar waves in our awake time. If we can optimize the 90-minute cycle, we can play with our dopamine system. A positive outcome (”I got something done!” after a phase of focused work will release dopamine and hence improve my motivation levels. Even if the next work is tackling an entirely new topic, I can carry the dopamine level into the new task. Remember the research about stacking enjoyable and not so enjoyable work?

4. I need to pay attention to my focus. If I continue to drift out, I might need to break down further. Continuously drifting attention after the first 20-25 minutes of work might show that I haven’t found the right combination of schemas and hence I compensate by throwing on more brain regions. Typically, we are more distractable, when this happens.

Sometimes though, there is no way around digging deep into an unknown topic. But even there we can use systems to organize the learning process based on cognitive load limits and existing schemas.

Next week, we will discuss the best ways to do that, backed by neuroscience! And straight from India, as I will be in Chennai and Bangalore all of next week!

Very excited!

Alex