A Simple Multiplicative Model of Adult “Brain Value”: Integrating Processing Capacity, General Knowledge, and Domain Mastery

Abstract

Research on adult cognition has long distinguished between fluid abilities, such as processing speed and reasoning, and crystallized abilities, such as vocabulary and accumulated knowledge. Fluid abilities typically peak in early adulthood and decline thereafter, while crystallized abilities often grow into midlife and later. More recent work emphasizes the importance of domain specific knowledge and expertise in understanding adult intellectual performance.

This article proposes a deliberately simple conceptual model that combines these strands. The model separates an individual’s age dependent “compute” component, representing broad processing capacity, from two knowledge components, general background knowledge and domain specific mastery. Overall “brain value” is represented as the product of compute and the sum of general knowledge and stacked mastery terms.

This multiplicative structure is intended as a communicative tool, not an empirical measurement model, but it captures several important intuitions. It explains why midlife adults often outperform their younger selves despite modest declines in processing speed, how sustained mastery efforts can compensate for lower baseline compute, and why polymaths who achieve mastery across multiple domains may have unusually high aggregate cognitive value.

The model is situated within existing theories of fluid and crystallized intelligence, investment theories, and Ackerman’s PPIK framework, and its limitations and possible extensions are discussed.

I’ve published a much simpler and probably more interesting version of this topic here should you want the human readable version.

1. Introduction

Debates about “when people get dumber” often treat intelligence as a single quantity that simply peaks and declines with age. Empirical research paints a more complex picture. Fluid abilities, such as reasoning with novel information and processing speed, tend to peak in late adolescence or early adulthood, then gradually decline. Crystallized abilities, including vocabulary and accumulated knowledge, generally increase into midlife and often remain stable until later life.

At the same time, adult cognition is strongly shaped by domain specific knowledge and expertise. Ackerman has argued that such knowledge is the “dark matter” of adult intelligence, especially in real world contexts, and embedded it in his PPIK model of intelligence as Process, Personality, Interests, and Knowledge.

The present paper introduces a simple multiplicative model that attempts to integrate these ideas in a form that is easy to visualize and communicate beyond specialist audiences. The goal is not to replace formal psychometric models, but to provide a conceptual scaffold for thinking about how age, baseline processing capacity, general knowledge, and domain mastery jointly contribute to what we might call “brain value” in everyday performance.

2. Background

2.1 Fluid and crystallized intelligence across adulthood

Cattell and Horn’s distinction between fluid intelligence (gf) and crystallized intelligence (gc) has become a central framework in cognitive aging. Fluid intelligence refers to the capacity to solve novel problems and to reason in situations that do not depend heavily on prior learning. Crystallized intelligence reflects the breadth and depth of acquired knowledge and skills that are valued in a given culture.

A broad literature indicates that fluid abilities increase through childhood and adolescence, then typically peak in late adolescence or early adulthood and begin a gradual decline. Crystallized abilities, in contrast, tend to increase into the 40s or 50s, and often remain relatively stable until later in life. Hartshorne and Germine’s large scale work with over 48,000 participants illustrates that different cognitive abilities peak at different ages, with some measures peaking near high school graduation, others in the 30s, and still others not until the 40s or later.

These patterns already suggest that no single “age of peak intelligence” exists, and they invite models that explicitly separate processing capacity from accumulated knowledge.

2.2 Adult intelligence, investment, and domain specific knowledge

Beyond the gf / gc distinction, several theorists have emphasized the role of investment and domain specific knowledge in adult cognition. Cattell’s investment theory proposed that individuals invest their fluid intelligence in learning, which over time produces crystallized intelligence. Ackerman’s PPIK framework extends this by including personality and interests as drivers that channel investment into particular domains, where knowledge accumulates and becomes a major determinant of performance.

In his “dark matter” paper, Ackerman argues that middle aged adults may appear less intelligent than young adults on narrow tests of fluid abilities, yet have greater knowledge in many content domains that are relevant for work and daily life. This aligns with expertise research showing that deep, domain specific knowledge and pattern recognition can dramatically improve performance, even when basic speed or working memory is unexceptional.

However, these models are complex and are usually expressed in factor analytic or structural equation frameworks that are not easily translated into an intuitive scalar metric for lay audiences. The next section proposes a deliberately simplified formulation.

3. A simple multiplicative model of “brain value”

For an individual at a given age, we define three main components:

1. Compute, C
   A normalized index of broad processing capacity at that age. This is meant to capture the shared contribution of fluid abilities such as reasoning, processing speed, and working memory.
   • C is scaled so that 1.0 represents the individual’s peak level (typically in early adulthood), with values below 1.0 representing lower capacity relative to that personal peak.
   • In reality C is a function of age, but for notational simplicity we write C rather than C(a), with the understanding that all quantities are age dependent.
2. General knowledge, G
   A normalized index of broad, non specialized knowledge: vocabulary, general facts, and everyday skills that accumulate with time.
   • G increases from childhood into adulthood, typically reaching higher levels in midlife before plateauing and eventually declining.
   • Again, G depends on age, but we suppress the explicit (a) in the notation.
3. Domain mastery, mᵢ
   The individual’s mastery level in domain i (for example welding, medicine, accounting, or software engineering) is represented by a value mᵢ between 0 and 1:
   • mᵢ = 0 indicates negligible competence.
   • mᵢ = 1 indicates substantial expertise acquired through sustained practice and learning.

The total mastery term for the individual at that age is defined as:$$M = \alpha \sum_i m_i$$M=αi∑​mi​

where:

• The sum runs over all domains in which the person has developed some level of mastery.
• α is a scaling constant that sets the relative weight of domain specific mastery compared with general knowledge.

For clarity here is the formula broken down in words to explain step by step it’s implementation:

“Total mastery M equals alpha times the sum of mastery levels mᵢ across all domains i.”

Here is what every piece means in concrete terms.

a. What is mᵢ?

• Think of each domain i as one specific area of mastery:
  • i = 1: welding
  • i = 2: plumbing
  • i = 3: finance
  • i = 4: psychology
  • etc.
• For each domain, mᵢ is that person’s mastery level in that domain, on a scale from 0 to 1:
  • mᵢ = 0.0 → no real skill there.
  • mᵢ = 0.5 → ok, competent, maybe intermediate level.
  • mᵢ = 1.0 → serious expert, real mastery.

So if we list them out:

• m₁ = mastery in welding
• m₂ = mastery in plumbing
• m₃ = mastery in finance
• m₄ = mastery in psychology
• … and so on

b. What does Σᵢ mᵢ mean?

The sigma Σ means “sum over all i”.

So:

Σᵢ mᵢ = m₁ + m₂ + m₃ + m₄ + …

In normal language:
“Add up all the mastery levels across every domain where the person has any skill.”

Example:

• Welding: m₁ = 0.9
• Plumbing: m₂ = 0.2
• Finance: m₃ = 0.0
• Psychology: m₄ = 0.7

Then:

• Σᵢ mᵢ = 0.9 + 0.2 + 0.0 + 0.7 = 1.8

That 1.8 is a raw “stacked mastery” score: higher if you are good in more domains, or extremely good in a few.

c. What is α?

α (alpha) is just a constant that is chosen to set the weight of mastery relative to general knowledge.

• If α is small (say 0.1), each mastery contributes a modest boost.
• If α is larger (say 0.5), each full mastery level has a bigger effect.

We’re basically saying:
“How powerful is being a true expert in a domain, compared to just having general knowledge?”

We introduce α so the model can be tuned.

SIDE NOTE: In reality, multiple domains may interact synergistically, so a more realistic model might make M grow faster than linearly with the number of mastered domains, for example $M = \alpha (\sum_i m_i)^{\beta}$M=α(∑i​mi​)β with $\beta > 1$β>1. For simplicity, I stick with the linear form in this article.

d. Putting it together: what is M?

Now:

M = α Σᵢ mᵢ

means:

1. For each domain, decide the mastery level mᵢ (0 to 1).
2. Add them all up to get Σᵢ mᵢ.
3. Multiply that sum by α to convert “raw stacked mastery” into a single mastery bonus M that goes into the main formula.

So step by step with numbers:

Example

Let us say a person has:

• Welding: m₁ = 1.0 (true master)
• Plumbing: m₂ = 0.7
• Accounting: m₃ = 0.0
• Software: m₄ = 0.8

And let α = 0.3.

Step 1: sum the mᵢ

Σᵢ mᵢ = 1.0 + 0.7 + 0.0 + 0.8 = 2.5

Step 2: multiply by α

M = α Σᵢ mᵢ
M = 0.3 × 2.5 = 0.75

So for this person, the total mastery term that goes into:

V = C × (G + M)

is M = 0.75.

You can interpret that as:
“Across everything they are good at, they get a mastery bonus of 0.75 added on top of their general knowledge G, then all of that is scaled by their compute C.”

3.2 Brain value as a scalar

At a given point in a person’s life, we define their overall brain value as:$$V = C \times (G + M)$$V=C×(G+M)

Substituting the definition of $M$M:$$V = C \times \left(G + \alpha \sum_i m_i\right)$$V=C×(G+αi∑​mi​)

Where, at that age:

• C is their compute (normalized processing capacity).
• G is their general, broad knowledge.
• mᵢ is their mastery level in domain i (0 to 1).
• Σ mᵢ is the sum of their mastery levels across all domains.
• α sets how powerful domain mastery is relative to general knowledge.
• M = α Σ mᵢ is the stacked mastery bonus.

Conceptually:

• Compute C scales how effectively the person can use what they know.
• Knowledge is the combination of:
  • General background G, plus
  • Stacked domain mastery M.

All of these quantities do change over time, but instead of writing C(a), G(a), mᵢ(a), M(a), V(a) everywhere, we treat them as “whatever their values are at this age” and keep the notation light.

The goal is to provide a transparent way to reason about:

• Why polymaths, who accumulate mastery in multiple areas, may achieve unusually high aggregate brain value.
• Why overall cognitive value can increase through midlife despite modest declines in processing capacity.
• How sustained mastery in one or more domains can compensate for lower baseline compute.
• Why individuals with lower or atypical baseline cognition, including those with intellectual disabilities, can still attain high brain value within specific domains through long-term practice and mastery.

4. Illustrative implications

4.1 Age related trajectories

Suppose, for illustration, that C(a) rises from childhood to a peak of 1.0 in early adulthood, then declines gradually to 0.85 by late midlife and to 0.7 in very old age. Suppose G(a) rises from near zero in childhood to 0.7 in midlife and then plateaus before slowly declining. These shapes are consistent with the broad empirical picture for fluid and crystallized abilities, though the numerical values here are hypothetical.

If we consider an individual with modest domain mastery (for example Σ mᵢ ≈ 1.0 by midlife, α = 0.3), the model predicts that V(a) will continue to rise into midlife even as C(a) declines modestly, because G(a) and M(a) are still increasing. Only when C(a) declines substantially, and when G(a) and M(a) flatten or fall, will V(a) show a marked downturn.

This captures the intuitive idea that many people are most effective in complex real world tasks in their 40s through 60s, even though their reaction time or working memory span may have already declined from their 20s peak. Empirical work on the age of peak achievement in various fields often finds maxima in midlife rather than early adulthood, which is consistent with this pattern.

4.2 High compute without mastery

Consider an individual at age 35 with very high compute, moderate general knowledge, and little sustained investment in any domain. For example:

• C = 0.95
• G = 0.5
• Σ mᵢ ≈ 0

Then M = 0, and:$$V = 0.95 \times (0.5 + 0) = 0.475$$V=0.95×(0.5+0)=0.475

In everyday language, this is the “bright but lazy” profile. The individual has high baseline capacity, yet their overall brain value in practical contexts is limited by the absence of deep, domain specific knowledge.

4.3 Moderate compute with strong mastery

Now consider an individual of the same age with somewhat lower compute, similar general knowledge, but sustained investment leading to mastery in one domain:

• C = 0.80
• G = 0.50
• Σ mᵢ = 1.0 (one fully mastered domain)
• α = 0.5 ⇒ M = 0.5 × 1.0 = 0.5

$$V = 0.8 \times (0.5 + 0.5) = 0.8 \times 1.0 = 0.8$$V=0.8×(0.5+0.5)=0.8×1.0=0.8

Despite lower compute, the individual’s overall brain value is higher than in the previous case, because domain mastery significantly increases the knowledge term. This expresses, in a simplified way, the investment idea that sustained learning can more than compensate for modest differences in baseline cognitive capacity, at least within a domain.

4.4 Polymaths and stacked mastery

Finally, consider a polymath at age 50 with moderately high compute, substantial general knowledge, and deep mastery in several domains:

• C = 0.85
• G = 0.70
• Σ mᵢ = 4.0
• α = 0.5 ⇒ M = 0.5 × 4.0 = 2.0

$$G + M = 0.7 + 2.0 = 2.7$$G+M=0.7+2.0=2.7 $$V = 0.85 \times 2.7$$V=0.85×2.7

Compute 0.85 × 2.7:

• 85 × 27 = 2295
• 0.85 × 2.7 = 2295 / 1000 = 2.295

In this toy example, brain value is nearly five times that of the bright but uncommitted individual in Section 4.2. Each additional domain of mastery adds an incremental contribution to the term in parentheses, and the multiplicative role of compute means that even modestly reduced processing capacity can still support high overall cognitive value when knowledge is extensive.

This provides a simple way to visualize why individuals who achieve high competence in several domains can appear disproportionately capable in integrated, real world tasks, even if they are not at their fluid cognitive peak.

4.4 Comparing the archetypes

If we were to examine the brain value trajectories of our three examples, the graph may appear somewhat like this.

Profile A – “Bright but lazy”

• Highest compute curve (peak C ≈ 1.0),
• No meaningful mastery, M(a) ≈ 0 for life,
• V(a) rises in youth as compute and general knowledge grow,
• Then flattens and slowly declines, because there’s no mastery to keep pushing it up.

Profile B – Single-domain master

• Slightly lower compute (peak C ≈ 0.85),
• One domain’s mastery ramps from ~18 to ~40, then stays high,
• V(a) overtakes Profile A by mid-30s and stays clearly above it through midlife,
• Declines only gently in later life.

Profile C – Polymath (4 domains)

• Compute just under A (peak C ≈ 0.9),
• Four domains accumulating over time (staggered ramps),
• V(a) starts only modestly above A early on,
• Then climbs steadily and ends up far above both A and B in mid- and later life.

5. Relation to existing theory

The model is compatible with, and conceptually derived from, existing theoretical traditions, but it intentionally collapses them into a low dimensional representation.

• The C(a) term corresponds broadly to the shared contribution of fluid abilities to many tasks. It echoes gf in the Cattell Horn tradition, though it does not attempt to model separate components such as reasoning, working memory, or processing speed.
• The G(a) term reflects the broad accumulation of crystallized abilities. It is not tied to a particular test battery, but captures the idea that vocabulary, general knowledge, and culturally valued skills typically increase well into adulthood.
• The mastery term M(a) = α Σ mᵢ(a) reflects Ackerman’s emphasis on domain specific knowledge as a central component of adult intelligence, influenced by personality and interests that direct long term investment.
• The multiplicative form V(a) = C(a) × (G(a) + M(a)) aligns loosely with investment theories in which fluid abilities support learning that builds crystallized abilities. It also reflects the intuition that processing capacity scales how effectively knowledge can be applied, rather than simply adding to it.

The novelty here is not in the components themselves, but in their explicit aggregation into a single, communicable scalar with an additive stacked term for multiple domains of mastery. This scalar representation is designed for conceptual clarity and communication, not as a replacement for established psychometric frameworks.

6. Limitations and possible extensions

This model has several important limitations.

1. Lack of measurement specification
   The functions C(a), G(a), and mᵢ(a) are not defined in terms of observable measures. The model does not distinguish between different fluid indices, types of knowledge, or different levels of mastery within a domain.
2. Oversimplified aggregation
   Real world performance in a particular task depends on specific subsets of abilities and knowledge. A single scalar V(a) ignores this structure. In practice, some tasks are constrained almost entirely by compute, while others rely primarily on domain knowledge.
3. No role for non cognitive traits
   Personality traits, motivation, self regulation, and interests all influence learning, performance, and the pursuit of mastery, and are integral to frameworks like PPIK. They are omitted here for simplicity.
4. Health and pathology
   The model does not distinguish normal aging from pathological processes such as neurodegenerative disease, nor does it incorporate the influence of education, socioeconomic status, or health behaviors on trajectories of C(a), G(a), and mᵢ(a).
5. Functional form
   The decision to make compute multiplicative and knowledge additive is a conceptual choice, not an empirically validated functional form. Other forms, for example interactions between specific abilities and specific knowledge domains, might better capture some phenomena.

Despite these limitations, the model can be useful as a didactic device. It encourages explicit separation of processing capacity from knowledge, highlights the role of sustained mastery, and makes clear why age related declines in fluid abilities do not imply a simple, monotonic decline in overall cognitive value.

7. Conclusion

The simple multiplicative model presented here defines an individual’s “brain value” at a given age as the product of a compute component and the sum of general knowledge and stacked domain mastery. It is meant as a conceptual tool that integrates three well supported ideas from the cognitive aging literature. First, fluid abilities peak relatively early and decline gradually across adulthood. Second, crystallized abilities typically accumulate into midlife and beyond. Third, domain specific knowledge and expertise are central to adult intellectual performance.

By representing domain mastery as an explicit, additive term that can be accumulated across multiple domains, the model illustrates how sustained investment can compensate for lower baseline compute, and why polymaths who achieve high competence in several areas can have disproportionately high aggregate cognitive value. Conversely, individuals with high compute who do not pursue mastery may under realize their potential.

Future work could develop more detailed versions of this framework, link it to empirical indicators, and explore how personality, motivation, and environmental factors shape the trajectories of C(a), G(a), and mᵢ(a). Even in its simple form, however, the model may be useful for reframing public conversations about cognitive aging, expertise, and the value of lifelong learning.

If you have any thoughts, critiques, suggestions or improvements to this model PLEASE, PLEASE leave a comment! I don’t know how exactly this model might even be used by anyone other than myself, so let me know what you think.