The GAT Skill List

The purpose of a cognitive skills tests appear to be measuring cognitive skills.  Or does it?  This is the first of a 2 part series evaluating just what the heck a cognitive skills test is measuring and how to beat it.

For the last few years I've been happily working with a framework that seemed so obvious to me.  I am double checking and it before I get to obvious we have to wade through a dark and murky swamp.

The Big Picture

#1 Cognitive Skills + Interest + Will To Do It = academic and life success.

The aspect of teaching cognitive skills without destroying interest or crushing will is called coaching. If you google my blog and the terms "crying" or "whining" I'm referring to coaching.  The last post is half on an article on this concept, and a discussion of coaching will complete the picture that this article begins.  (I'm going to be adding to the coaching article over time and will warn everyone when it is complete.)

#2 Cognitive Skills + Interpersonal Skills (e.g., empathy) + Intrapersonal Skills (e.g. self control) = The Full Skill Set

If you child is missing one of these skill sets there is work to be done.  I think band and Cub Scouts/Girl Scouts are really good activities for little kids to address the last 2 skill sets.  Executive Functioning programs are good for preK training if you have $30,000 to spend on pre school; if not, projects, crafts and art help.  Music in general is great for the first and last skill set.

I don't think team sports at young ages are good at all for any of these skill sets.

#3  The Classical Education
For a really inspiring discussion of the classical education, read the first 30 pages of Susan Wise Bauer's The Well-Trained Mind.

Here is my butchered summary:  Grades 1-4:  Cram as much information into their brains as possible. Grades 4-8: Think analytically.  Grades 8-12: Think logically, and grades 12-16 (yes, there is overlap in grades) I haven't gotten this far yet.

If you pair this with Piaget's Theory of Cognitive Development, then you can see that there are things expected of older children that would be impossible for younger children.  This is obvious.  What is not obvious is whether or not later skills like logic need new cognitive skills or just refinement of an underlying skills.

Since my focus is grades K-4 and I only dabbled in 5+ (because I found out the SAT is using problem solving skills and cognitive skills to improve their test), this theory will have to be explored later.

I'm interested more in training the brain at ages 5-9 to handle info and teaching higher order skills, but I can assure you that I've done as much information cramming as anyone into my kids' brains.

#4  Is 99% attainable for long periods of time with a normal kid?

Every 4 years the bar moves and the list deepens.  Cognitive skills training at age 5 only buys so much for a 9 year old, especially if he hasn't practiced his skills for 5 years.   So far, my plan is simple.  Find out what the skills are and what the level is, and provide suitable learning material.   Test prep is a great opportunity for this effort.

But at some point the only people who really need 99% in anything are those who really want it. When I get to the point where I'm helping with sign ups for AP Calculus, AP History, AP Literature, AP Civics, Orchestra, Play Practice, Cross Country and I don't know what else, we're going to have a goals planning session and make some choices.  But that is many years away and I've changed my mind before.  If I'm dropping $250,000 for college, we're back to 99% and this is not negotiable.

#5  Tests
There is a really great article here summarizing the NCBI's grouping of skills.  What I find most interesting is the discussion of problem solving on the PISA, that international test on which US students deliver a spectacularly inept performance.   The paper also evaluates graduate placement exams.  Cognitive skills and tests go hand-in-hand.  A cynical parent might be motivated by the need for their child to pass tests.  A benevolent parent wants their child to have the same skills that test makers use to predict success.

Types of Research
Research comes in 2 flavors.  The first half is totally made up intelligence crap that provides a lot of useful sounding words, and the second approach is watching kids work, succeed and fail and applying these words to what is really happening.  I think you know what side I'm on.

There is a third branch of research and that is super secret GAT tests which are good at predicting academic outcomes, and to some extent, life outcomes.   I spent quite a bit of time gleaning cognitive skills from this area and apply it to problem solving skills.   Angela Duckworth's Grit research almost fits into this category but she made it publicly available from the get-go and is the fourth inductee in the Competitive Parent Hall of Fame behind Jo Boaler, Susan Wise Bauer, and George Poyla.  (David Lohman is on the fence.)  The difference between cognitive skills and GAT tests is that GAT tests use a much longer list called "cognitive abilities".  I note these on the chart below.

Blooms Taxonomy
I was hoping to start here, but this is a depricated construct.

Poyla's Problem Solving Method (1945)
This is pretty close to my approach to cognitive skills.  I have lined up cognitive skills to the problem solving steps and this is my focus for teaching.  If a problem solving step requires a cognitive skill that isn't there, we work on it.

Some math workbooks start with a very detailed section on problem solving strategies.  I was very excited about this section until my son applied a strategy to successfully solve a problem without knowing what he was doing, then I ripped the section out of his book.  I've noticed these books work step-by-step with scaffolding.  I've talked to parents who's kids used these books and struggle to get a high score on the COGAT.  I'm not using them anymore or recommending them.

Critical Thinking
I think there is a big overlap between Critical Thinking skills, Cognitive Skills, and Problem Solving Skills.  Any 2 should almost define the complete system with the addition of Working Memory. Working Memory is a necessary prerequisite to using these skills.

The critical thinking list strikes me as overly detailedThese appear to be things derived from lower order cognitive skills.  For example, a critical thinking skill is "observing similarities and differences" which should follow from the higher order Problem Solving skill would be "stare at the problem until you understand it".  I think the COGAT test agrees with me but I'm not sure what they would call that skill.   Maybe "ability to compare things", maybe "see details in things".

The List

I'm going to take Poyla, a critical thinking list, and an Executive Functioning list and organize it into my biased construct with cognitive skills.  Finding a list of cognitive skills is impossible so I created an unathoritative list from the lists available that I think appear on GAT tests. There is a great list on SharpBrains.com that looks like it is reading-comprehension-centric but I don't use it.

My interpretation of Poyla is over simplified so that it applies to 5-9 year old children.  I would have a different presentation for different age groups.

I'm surprised that no one has done this before (curse you cognitive researchers) but if researchers had been on the ball for the last few decades I wouldn't have started this blog, my writing skills would stink even worse than they do, and my slightly above average children would never have had a shot at cheating their way into a GAT program nor succeeding in it.

There is a much looser connection with Grit than I would like.   Duckworth's research is on a longer term connection between work and fortitude and outcome.   I need a really short term connection, like the time it takes to sit for a test.  The Grit column has weak associations so I didn't include Grit in the chart except as notes under Executive Functioning.

It wasn't until I sat down to do this comparison that I realized just how bad the list of cognitive and critical thinking skills are.  They are either too narrow, incomplete, or more commonly too broad to base a curriculum on.  I have found that problem solving skills are well defined and concrete, easy to diagnose and best of all, make a great approach to beating a GAT test or getting an A in a course. For this exercise, in the back of my mind I'm thinking about GAT tests and how different problems would use what skill, and this approach is tailoring my framework.

Note that the "Understand the Problem", "Solution Strategy", and "Execute Strategy" sections have such a high degree of overlap that many of the activities within either section could easily happen in the other.    In fact, for most children the first 3 steps are a blur of activity.  For example, on a pattern problem does pattern recognition happen when understanding the problem or solving it?  Part of the problem is that cognitive skills tests have well defined rules and the solution strategy is almost a foregone conclusion.  The example tells the child exactly what to do.  The other problem is that children are not properly trained to spend most of their time analyzing the problem and usually just jump in to solve it - prematurely I might add.

Poyla Problem Solving	Critical Thinking	Cognitive Skills	Executive Functions
Understand the Problem Read the problem Understand the problem Use all elements in the problem Fill in missing or implied elements Eliminate distractions Highlight key elements Make assumptions	Analyzing Information Seeking Context	Visual Spacial Processing Working memory Attention to Detail Identify Patters Identify Similarities/ Differences Inductive and Deductive Reasoning Inference Resolve ambiguity (GAT)	Ignore Distractions Perception Flexibility (thinking modes) Pattern Recognition
Solution Strategy Look for short cuts Eliminate incorrect answers Apply examples Draw a [mental] picture Break problem into steps Find subset that solves the problem Try an easier problem, work up to harder one Look for patterns Solve it backwards Guess/check or try all options	Problem Solving Decomposition Algorithms	Working Memory Problem Decomposition Apply examples (GAT) Eliminate answers (GAT)	Problem Solving Decision Making Planning Apply rules Sequencing (Project Management) Set Goals (Grit and EF)
Execute the Strategy check progress Try different approach Abandon assumption, try again	Classification Analogous reasoning Similarities/ Differences Applying rules Predicting Transforming Reasoning Generalize	Working Memory Meta Cognition Spacial Reasoning Analogous reasoning Mentally fold (GAT)	Self Control Overcome setbacks (Grit) Maintain Interest (Grit) Diligence (Grit)
Check the Answer Check solution Try again Start over	Evaluation	Rethink assumptions when answer isn't ther (GAT)	Finish the Job (Grit)

The solution strategy section is the biggest gap for Critical Thinking and Cognitive Skills literature.   The Poyla list has an important set of learned skills.   Experiments with EF programs proved that kids with EF training do much better on GAT tests (by a ridiculous margin).   Conversely, Problem Solving skills don't really define "Understand the Problem", yet cognitive skills break this section down. This difference is most pronounced on non-verbal tests.  After 1st grade, all tests have a greater portion of verbal and quantitative questions.

I think the main difference is that Poyla's list as I represent it doesn't care what the problem is and is silent on the assumed skill set.  In practice, it is usually geometry theorems.   GAT tests have a well defined set of skills for problem solving, but don't care about problem strategies because the problem strategy is pre-defined.  For GAT tests, the most important competitive advantage comes from the first and last sections - understanding the problem and checking the answer.  I know that from coaching.  Anyone can learn to rotate triangles.  It's much harder to teach patience,carefulness, and diligence.

There is a list of critical thinking skills that can be found here that looks like it applies almost 100% to reading comprehension questions.   My list primarily applies to all other question types but not reading comprehension.  Ironically, I feel strongly that reading comprehension is a great practice for non-verbal and cognitive skills tests.   The two reasons are working memory and understanding the problem, which combine to form about 65% of the magic.   Reading comprehension is great practice for both skills.

This brings me back to my goal of Test Prep Math and why I'm spending my week off working on it. My working theory is that a child who is patient (understand the question) and careful (check the answer) and diligent (tries again) will have an advantage both in school and on any test over a child who does 1,000's of figure matrices or arithmetic problems.  In fact, hurrying through a worksheet with lots of problems may teach speed over patience, carelessness, and boredom.  Working memory is essential and working memory is not exercised with fast, easy problems.   I'm closing the gap.

For first grade, this is what I have in mind.  Take a figure matrix.   Throw out the format, because it's copyrighted.  Come up with something that has the same cognitive objectives but is so convoluted that when the child asks "Did I get the right answer?" the parent has to spend 15 minutes staring at it to see if the child was correct.  I'll work on it this summer.