Expertise

Rather self evidently – expertise in flying involves being able to ‘fly’ the aircraft ‘adequately’, but once you move past this necessary skill few people can accurately define what it is that differentiates expert pilots from the rest of us – even though we all believe we can readily recognise a expert. Motherhood statements such as ‘being ahead of the aircraft’ really don’t get us any further as they are the result of aviation expertise, not the cause of it. Before continuing please recognise that in this article the terms expert and novice are being used relatively and actually exist on a continuum rather than being dichotomous, so most pilots would fall somewhere between the two extremes displaying both expert and novice skills in varying amounts.

Whilst it is rather obvious to say that experts know more than novices, it is less well recognized that experts know things in a qualitatively different way. Over years of what is termed “deliberate practice”, expert’s knowledge has been transformed into what are called abstract schemas, or models of how the (aviation) world works. As a result their knowledge has become integrated and interrelated in ways that make it easier to both recall and use. In addition because this knowledge is stored in integrated schemas or chunks, experts are able to bypass the cognitive bottleneck of our limited working memory capacity. Normally our working memory can only hold and process around seven units of information at a time, however it makes no difference whether these are isolated pieces of information or large chunks. Hence increasing the size of the chunks that are being processed effectively frees up mental capacity.

Related to this is experts have a greater and more abstracted range of what are termed scripts, or routines, that they can draw on. What this means is that when experts recognise a situation they have a range of predetermined procedures that they can use to deal with the situation, or because of the abstracted forms of these scripts, can readily mesh several scripts together if required. Once again this relieves the limited capacity of our working memory to deal with other tasks as these scripts can effectively run subconsciously. An easy way to think of this concept is the use of flow patterns for pre-departure checks and how easy these make it to properly check and set a whole range of items, which would be much harder to memorise and set individually.

Building on this we come to the next distinguishing trait – experts have finer and more discriminatory perceptual skills. In other words experts can see (and notice) subtle changes or patterns that are invisible to others – until it is pointed out to them. In addition this also allows experts to see through the ‘noise’ in a situation, and determine what is important and what can be ignored. This also explains why aviation’s ‘situational assessment’, which is what generates ‘situational awareness’, is so difficult to train – it is largely a function of expertise.

The next major difference is that experts have a greater sense of what is typical in a certain situation and what isn’t. As has already been mentioned, over years of experience experts have developed abstract models of how their professional world works and how it fits together. As a result, experts can quickly recognise a situation as fitting into one of, or combinations of, their abstracted models – even if the situation is normally abnormal. Importantly, these models also allow experts to accurately predict what should and shouldn’t happen; hence they can quickly recognise if things are “running off the rails” and take early and appropriate remedial action. This allows experts to quickly spot the difference between small impending changes that are important, and ones that can be ignored.

In addition, experts possess “negative knowledge” – they know what not to do, or what won’t work in a situation. Once again this has an important effect on the mental bottleneck of our working memory. By allowing us to limit our search for solutions to problems we encounter, we are left with more available capacity to use in other areas.

Finally, it has been shown that experts have what the researchers call “anticipatory knowledge”. This isn’t inert knowledge or simply predictions about what might happen; rather anticipatory knowledge is stored as scripts or procedures where experts have prepared themselves for action – even for the low probability high threat events. Once again, should such a situation occur, anticipatory knowledge will reduce the cognitive load on the expert pilot since they already know both what to do, and how to do it.

However the story doesn’t stop here, since the above differences also mean experts and novices are able to make decisions in totally different ways. Considerable research in a wide range of fields has shown that experts usually make decisions in a fast and forward thinking manner that requires little cognitive processing, using recognition primed decision making processes. In these experts first spend time understanding the problem at a strategic level and when the problem is properly understood a satisfactory solution essentially becomes self evident – a process that can be virtually instantaneous. As a consequence of this, the only way to teach experts to make better decisions is firstly to help them understand the inputs into their existing decision making processes and where they are likely to make mistakes, ie to make them more expert; and secondly by encouraging them to use their spare cognitive resources (if time is available) to look for a better solution, or to play devil’s advocate in reviewing their current decision and resolving any potential ambiguities and / or discrepancies.

For example, an area where highly trained pilots often make poor decisions is in what’s termed ‘plan continuation errors’ – where they should have taken an alternative course of action but for some reason continue with their existing plan despite mounting evidence to the contrary. Whilst there can be many factors in these types of errors, something that isn’t as well known is that our working memory capacity reduces to 2 to 4 ‘units’ for more complex processing operations. Reordering competing goal hierarchies would meet the definition of a complex mental operation, which is what is called for to avoid plan continuation errors. Hence, the vital importance of planning for that missed approach, or exactly what you will do if you get high and hot on an approach onto a rain soaked runway in marginal conditions, when your spare cognitive resources are likely to already be limited or in overload. Recently a human factor specialist has coined the term OVERLOC to describe this perilous state where reasoned decision making becomes impossible – cognitive overload loss of cognitive consciousness. This is a major dynamic threat that is rarely raised in Threat and Error Management discussions.

Because novices don’t have the same abstracted, deep and integrated knowledge that experts do, experts’ recognition based decision making processes are usually not available to them. Hence, they are forced to use rearward looking, means-end analysis types of decision making processes. This means they have to start with the desired state and work backwards to figure out how to get there from the current state. In addition to having to use these slow and cognitively taxing decision making process, novices also will focus on the surface features of the problem – simply because they don’t have the depth of knowledge to do anything else. In other words, you can’t teach novices to make decisions in the same way as experts do. Research has shown that they are already trying to do this but simply can’t make the connections. However you can approach this from the other direction by helping novices become more expert, as quickly as possible; which also means training them in the skills to increase the rate of their own experiential learning.

It is worth noting that when experts come across novel or unusual problems they essentially become ‘educated novices’ in relation to that particular problem, so are forced to use similar decision making methods to novices. Fortunately they do still have their integrated domain knowledge plus expert metacognitive skills to help them. (Expert metacognitive skills are an essential component of expertise and refer to the ability to monitor and think about your own thinking; further discussion of metacognition is beyond the scope of this brief article).

So if you have ever wondered why some people can make flying look so easy, the simple fact is that it is easy for them (most of the time)!

This is because all of the above expert qualities allow for the more efficient use of the cognitive bottleneck of their limited working memory capacity (where all our mental processing takes place), leaving spare capacity for other uses. How this spare cognitive capacity is then used is another subject in itself, but essentially this will determine whether their aviation related skills continue to further develop or stagnate. Of course this comment is made with the assumption that the expert has sufficient expertise to fully understand what is happening around him or her in the first place, as distinct from a less expert pilot who may be happily oblivious to his or her impending predicament as they inappropriately attend to less germane or even non operational tasks. As James Reason wrote, citing Wagennar and Groeneweg – “Accidents do not occur because people gamble and lose, they occur because people do not believe what is about to occur is at all possible.”

Readers may find it interesting to review the CBS interview with Captain Sollenberger linked below, for an illustration of how many if not all of these skills are used when a real expert was faced with an unprecedented situation.

https://www.youtube.com/watch?v=ixUbdeXCp0M&feature=related