Reasoning and Diagnostics Pt 1

Barney Donohew explores logical and critical thinking

Published:  17 February, 2017

Diagnostics is all about decisions. And what is a decision? It is a conclusion or resolution reached after consideration. Therefore, efficient and effective diagnostics is about drawing the right conclusions at the right time. How do we do that? Amongst other things, by making sure our logical and critical thinking skills are up to scratch. This series of articles aims to help us with that by looking at the principles of human reasoning.



I want to add a little extra context and motivation. Why? Because this is going to read like a textbook chapter (sorryÂ…!) and it might not be obvious that taking the time to understand human reasoning will help us fix vehicles. However, seeing as we go back to basics whilst fault-finding, why not do it now when weÂ’re trying to understand our logical and critical thinking? You wouldnÂ’t learn to identify all the systems and components in a vehicle without learning the physics needed to understand how they work together, so why would you learn how to make diagnostic observations without making sure you were correctly piecing them together to draw your conclusions?


Should we care? Yes, because diagnostic technicians are undervalued and their logical and critical thinking abilities, necessary for effective diagnostics, are often overlooked. Also, when reasoning, humans have a habit of making the same types of mistakes. If we can be made aware of them, they might be avoided. Then, we will have improved our reasoning skills and become better diagnosticians.


Our most important diagnostic tool is our brain. Most of us are aware of the benefits of training, which in some ways is like adding or updating a software program within our brain, bringing with it new fault-finding knowledge and skills. What about our brain’s “operating system”; are our logical and critical-thinking skills working okay behind the scenes? It won’t do any harm to check for any bugs and make sure everything is running smoothly.



How do humans arrive at conclusions? We make inferences. An inference is a step within reasoning, where we use existing information to generate new information; i.e. getting “four” after “putting two and two together”. When explaining our reasoning, we present the relevant information as an argument; a collection of statements (premises) supporting a conclusion. Usually, our arguments present either a guaranteed conclusion (deductive reasoning), a probable conclusion (inductive reasoning) or a possible conclusion (abductive reasoning). Other kinds of reasoning exist but we’ll focus on these.


I keep mentioning critical-thinking. What do I mean? If we are drawing conclusions, we need to check or improve the validity and strength of the arguments behind them; that is what we do when we apply critical-thinking. How though do we apply our critical-thinking to check the logic of arguments? IÂ’ll show you but the process can feel like wading through treacle as it needs discipline and precision; just like the fault-finding process! Hence, if we donÂ’t exercise and train these traits thenÂ… well, we might as well get out the diagnostic dartboard...


Where to begin? WeÂ’ll start in this article with deductive reasoning, use it to establish some basic principles, and then continue in a subsequent article with inductive and abductive reasoning. This order doesnÂ’t infer that any one kind is any more effective or useful than another; we use them all, both during fault-finding and in our everyday lives.



In an ideal world, we would like to guarantee that our conclusions are true. Where we can use them, sound deductive arguments will provide this guarantee. Consider the following simplified argument we might hear within our workshops:


The engine needs fuel to run. The engine can run. Therefore, the engine has fuel. From the implication of the statement “the engine needs fuel to run” within the situation in which “the engine can run”, the argument guarantees its conclusion; i.e. assuming its supporting statements are true, its conclusion is guaranteed true. This is defined as a valid argument. Validity is necessary because having true supporting statements doesn’t guarantee a true conclusion, e.g. as in this example:


The engine cannot run. An engine needs fuel to run. Therefore, the engine has no fuel. In this argument, other factors might be preventing the engine from running, so it is invalid; a potentially true conclusion isnÂ’t true enough. From a fault-finding perspective, we need to be aware of a deductive argumentÂ’s validity, otherwise we might mistakenly accept a conclusion as being true, when there are alternative explanations.


As suggested above, validity is not the only important issue. The following is a totally valid argument, if the supporting statements are true, as it would guarantee its conclusion: All hybrid engines have flux capacitors. Flux capacitors warp space-time. Therefore, all hybrid engines warp space-time.


Therefore, for our arguments to be sound they must have true supporting statements and be valid, as in our first example above. If we can totally accept sound arguments, does that mean we must discard all others? For example:


The engine is sounding terrible. The valves must be bent. Therefore, the engine is knackered. This argument is neither valid nor sound: it doesnÂ’t logically demand that the conclusion is true and the second supporting statement is not guaranteed true; sometimes easily rectified problems can make an engine sound terrible and is it a fact that the valves are bent? Despite this we canÂ’t totally reject the conclusion, it could still be true. We just need to challenge ourselves to provide a better argument to justify it.


Our take away summary: deductive arguments must guarantee the conclusion, as there must be no room for any alternative explanations. If they can guarantee a conclusion and all the supporting statements are true, then the argument is sound. Sound is good!


Critical thinking and deduction

We’ve covered a lot of ground and it might have felt confusing – even though it is supposed to be logical – but this is just human nature getting in the way as we are programmed to take shortcuts and jump to conclusions. To bring us back to reality, let’s use an attempt at diagnostic deduction to show us exactly where critical-thinking helps:


Faulty coil packs cause a misfire. The engine has a misfire. Therefore, a coil pack is faulty. Seem familiar?! The supporting statements are true and the conclusion is plausible, and very tempting to accept, as we “know” it’s a common fault but, we can tell that this argument is invalid, as something else might be causing the stated misfire. So, here is the point: without any further consideration of any other facts, critical thinking can tell us that a decision to fit a new coil pack based on this argument would be invalid. Therefore, this type of critical thinking is the logical underpinning to asking “what would you test next if the part you were considering fitting did not clear the observed symptoms?” I.e. before we are even to consider any tests, we must first determine that our argument is unsound.


When to use deduction

We usually employ deductive arguments when the scope of our diagnostic deliberation is narrow, such as during specific system tests (e.g. “DPF regeneration does not start until the exhaust temperature is above 650 ˚C. The exhaust temperature is 450 ˚C. Therefore, DPF regeneration has not started.”) or, specific component tests, which hopefully could be the last test in your search if it has become sufficiently narrow (e.g. “A volt drop above 250mV between these two points on this wire would indicate a high resistance fault. The Volt drop is 3 Volts. Therefore, the wire has a high resistance fault between these points.”).

What next?

But what do we do if we donÂ’t have perfect knowledge regarding the way a system works (and it is arguable that we can never have perfect knowledge) or if we are at the start of a diagnostic investigation and all we have is a symptom that we canÂ’t easily connect to a possible system or component failure? This is where we can use inductive and abductive arguments to help drive the process forward. We will examine induction and abduction in future editions.


Related Articles

Most read content


Sign Up

For the latest news and updates from Aftermarket Magazine.


Where should the next Automechanika show be held?


©DFA Media 1999-2019