IX Coach

The Model of Hierarchical Complexity, Explained

How tasks are ranked by coordination, what the 14 orders are, and how to use the idea on a hard problem

What is the model of hierarchical complexity?

The model of hierarchical complexity, developed by Michael Commons and colleagues, ranks tasks by how many levels of coordination their actions must perform, and classifies a person’s stage as the highest-order task they complete successfully. It lists 14 orders, from 0 (computory) to 14 (cross-paradigmatic). Scoring is based on the structure of the task rather than its content. The 2014 scale study on this page lists the developer as its first author.

Michael Lamport Commons, of Harvard Medical School, and his colleagues developed the Model of Hierarchical Complexity (MHC) as a non-mentalistic account of developmental stages, meaning one that explains stages through the tasks a person can complete rather than through inner mental structures. Piaget explained change with age as the growth of such structures. This model treats development as a sequence of tasks that become more hierarchically complex, with less complex tasks completed and practiced before more complex ones. The complexity of a task is set by how many levels of coordination its actions must perform, where each higher action organizes the lower ones beneath it. The model specifies 14 orders, and a person’s stage is the highest order of task they complete successfully. The practices below are ways to apply that lens to a problem you are stuck on. None of them is a test of your own order, and the model does not claim to measure what kind of person you are.

Practices

How these practices connect

The model locates difficulty in how a task’s sub-actions are organized, and its stages describe successful performance on the most hierarchically complex tasks, “rather than intellectual maturity.” Each practice here works on one part of that organization, as this page applies it: what the parts are, which must come first, how two parts are coordinated, what counts as done for each, and which frame the whole thing sits within.

How does the model of hierarchical complexity decide how hard a task is?

It ranks a task by how many levels of coordination its actions need, not by its content. In the paper’s terms, complexity is the number of “recursions” (repeated layers in which a coordinating action operates on lower-order actions) that its coordinating actions must perform. The 2007 introduction says scoring rests on “the mathematical complexity of the hierarchical organization of information,” and gives the example that an order-three task has three “concatenation operations” (chained steps, each combining the one before). The introduction names task analysis as one major basis for the theory. The practices that map sub-steps and combine two skills are this page’s own applications of that idea, not claims the model makes about them.

What is the difference between horizontal and vertical complexity in a task?

Horizontal complexity is the amount of classical information a task needs, counted in yes-or-no questions, so more separate items means more horizontal complexity. Vertical complexity is the number of recursions (repeated layers of coordination) the coordinating actions must perform on those items, so a task can have few items and still be vertically hard. When you have many simple items, the response is to group them; when you have the items but cannot see how they relate, the response is to practice the relating step. That response is this page’s inference.

What should I practice first when a harder task keeps failing?

Find the exact step where the harder task fails, then ask which simpler action that step depends on. The 2007 introduction says less complex tasks must be completed and practiced before more complex tasks can be acquired. Practicing that underlying action alone until it works reliably, then returning to the harder task, is this page’s application of that claim to a single failing task.

How do I know when a multi-step task is actually finished?

The 2007 introduction treats tasks as quantal (all or nothing): they are either completed correctly or not completed at all, with no intermediate state. Applying that to your own work, you write a yes-or-no check for each sub-step and count the step as done only when the check is answered yes. This removes the middle ground where a step feels nearly complete while parts of it are still missing.

Does the model of hierarchical complexity measure intelligence?

This page’s reading is that the developers describe it in two ways that do not fully line up. The 2007 introduction says the highest stage represents successful performance on the most hierarchically complex tasks rather than intellectual maturity. The 2014 paper describes the model as a framework that represents intelligence, measured by the complexity of tasks a person accomplishes. Neither statement is a self-test, and a stage here is a claim about task performance, not a verdict on a person.

How do you choose between these?

Choose by where the stuck point is. If the whole task feels too big with no clear edge, start by mapping its sub-steps. If you can name the two things you cannot make fit, practice the step that combines them. If you have plenty of material and no idea how it connects, check whether the problem is too many items or too little relation between them. If a harder version keeps collapsing at one specific basic step, go back and practice that step alone. If you keep saying “almost done,” write a yes-or-no check for each step. If the same frame keeps producing the same failure, step up one level and place that frame inside a wider one.

What these approaches do not settle

The model is a model of tasks: it scores the structure of what a person does, and this page does not offer a way to find your own order. The number of orders is not stated the same way in every source. This page follows the 2007 introduction, which lists 14 orders from 0 to 14. The 2014 paper by Commons and colleagues prints a 16-row table, “16 Stages and orders of hierarchical complexity,” running from order 0 (calculatory) to order 15 (meta-crossparadigmatic). Its test of equal spacing reported the orders were equally spaced, and said a significant departure would have needed more than a quarter of an order. One of its six authors is listed at the Dare Institute, and the model’s developer is its first author. The 2007 paper says: “Few individuals perform at stages above formal operations.” The model explicitly differs from Piaget’s mental-structure account, so it is one competing account among several. No study testing any of the six practices as an intervention was found in the sources checked for this page.

Practice this with IX Coach

Practice this with IX Coach

Frequently asked questions

What is the model of hierarchical complexity?

The model of hierarchical complexity, developed by Michael Commons and colleagues, ranks tasks by how many levels of coordination their actions must perform, and classifies a person’s stage as the highest-order task they complete successfully. It lists 14 orders, from 0 (computory) to 14 (cross-paradigmatic). Scoring is based on the structure of the task rather than its content. The 2014 scale study on this page lists the developer as its first author.

How do you map the sub-steps before judging the task?

The 2007 introduction says: “One major basis for this developmental theory is task analysis.” The application below is this page’s own: using task analysis on a personal task to find which coordinating step is blocking you. The introduction does not make that claim. This page infers that once the sub-steps are visible, the difficulty can be located in one specific step rather than in the task as a whole.

How do you practice the step that combines two skills you already have?

The 2007 introduction works through 3 × (4 + 1). Getting it right needs a distributive act that coordinates adding and multiplying, so it is one order more complex than either alone. The paper says that people who can do both “display a greater freedom of mental functioning.” The practice here is this page’s application: isolate that coordinating move and rehearse it on purpose.

How do you tell apart too much information from too little connection?

The 2007 introduction distinguishes horizontal complexity, the amount of classical information a task requires, counted in yes-no questions, from vertical complexity, the number of recursions (repeated layers of coordination) that the coordinating actions must perform on the elements. A task can be heavy on one and light on the other. Naming which kind you have points to a different response, which this page infers: sorting and grouping for the first, practicing the relating step for the second.

How do you practice the simpler skill underneath a failing one?

The 2007 introduction states that less complex tasks must be completed and practiced before more complex tasks can be acquired, and presents this as the account of the developmental changes it describes. In the model, each higher order organizes and transforms the lower-order actions beneath it. If a higher task keeps failing at one specific point, this page infers that the point is often a lower-order action that has not yet been practiced to the point of working reliably.

How do you define what finished looks like for each step?

The 2007 introduction says tasks are quantal (all or nothing): they are either completed correctly or not completed at all, with no intermediate state. The model uses this to treat stages as hard and distinct. This page applies it as a self-check: a yes-or-no completion question for each step removes the fuzzy middle where you can keep telling yourself you are nearly there.

How do you step up a level when one frame keeps repeating?

The 2007 introduction says: “Some adults are said to develop alternatives to, and perspectives on, formal operations.” Formal operations (Piaget’s stage of abstract, rule-based reasoning, the highest stage in his model) is where that model ended. The introduction describes such adults as using formal operations within a higher system. In the model, a higher order organizes and transforms the lower-order actions rather than simply adding more of them. This page applies that idea to a personal frame: you keep the frame as one part of a bigger organization instead of the only option. The introduction does not discuss personal frames.

How does the model of hierarchical complexity decide how hard a task is?

It ranks a task by how many levels of coordination its actions need, not by its content. In the paper’s terms, complexity is the number of “recursions” (repeated layers in which a coordinating action operates on lower-order actions) that its coordinating actions must perform. The 2007 introduction says scoring rests on “the mathematical complexity of the hierarchical organization of information,” and gives the example that an order-three task has three “concatenation operations” (chained steps, each combining the one before). The introduction names task analysis as one major basis for the theory. The practices that map sub-steps and combine two skills are this page’s own applications of that idea, not claims the model makes about them.

What is the difference between horizontal and vertical complexity in a task?

Horizontal complexity is the amount of classical information a task needs, counted in yes-or-no questions, so more separate items means more horizontal complexity. Vertical complexity is the number of recursions (repeated layers of coordination) the coordinating actions must perform on those items, so a task can have few items and still be vertically hard. When you have many simple items, the response is to group them; when you have the items but cannot see how they relate, the response is to practice the relating step. That response is this page’s inference.

What should I practice first when a harder task keeps failing?

Find the exact step where the harder task fails, then ask which simpler action that step depends on. The 2007 introduction says less complex tasks must be completed and practiced before more complex tasks can be acquired. Practicing that underlying action alone until it works reliably, then returning to the harder task, is this page’s application of that claim to a single failing task.

How do I know when a multi-step task is actually finished?

The 2007 introduction treats tasks as quantal (all or nothing): they are either completed correctly or not completed at all, with no intermediate state. Applying that to your own work, you write a yes-or-no check for each sub-step and count the step as done only when the check is answered yes. This removes the middle ground where a step feels nearly complete while parts of it are still missing.

Does the model of hierarchical complexity measure intelligence?

This page’s reading is that the developers describe it in two ways that do not fully line up. The 2007 introduction says the highest stage represents successful performance on the most hierarchically complex tasks rather than intellectual maturity. The 2014 paper describes the model as a framework that represents intelligence, measured by the complexity of tasks a person accomplishes. Neither statement is a self-test, and a stage here is a claim about task performance, not a verdict on a person.

Research

Related concepts