Chris McInnes explores the role of cognition in warfare and asks whether we understand the types of cognition needed in the future, and what we might be able to learn from other disciplines.
On 9 June 1982, the Israel Air Force (IAF) destroyed 87 Syrian military aircraft and 17 out of 19 Syrian surface-to-air missile batteries in the Beka’a Valley, for the loss of almost no IAF assets. The Israelis held several advantages – the initiative, generally superior equipment, and higher quality personnel – but it was the manner in which the IAF combined its forces to fight for, and with, information that produced the lopsided result.In short, the Israeli strike forces could see, communicate, and understand the environment and events, while Syrian forces were blind, dumb, and confused. One observer recalled seeing Syrian fighter aircraft flying figure-eight patterns as the battle raged because they could not communicate with their ground controllers or take their own initiative. The image of Syrian military aircraft orbiting aimlessly while the Israelis dismantled and destroyed their ability to operate is an apt metaphor for the whole battle.
Much is made of the Israeli’s total dominance of the electromagnetic spectrum during the battle, but this was only one part of Israel’s success. To be sure, the IAF’s ability to control what could be sensed and communicated via the electromagnetic spectrum during the battle was the razor-sharp tip of the information spear. But the shaft of the spear – which supported and guided the point – was the superior training and education of the Israeli personnel. As one Israeli officer observed, “they [the Syrians] could have flown the best fighter in the world, but if they flew it the way they were flying, we would have shot them down in exactly the same way. It wasn’t their equipment at fault, but their tactics.
Through superior training and education, the Israelis were better prepared for the battle. To use John Boyd’s famous OODA cycle, the Israelis were better oriented for the battle and were thus able to observe, decide, and act far better than the Syrians.
The same was true of air combat during the Korean War. American pilots’ training was the foundation for their dominance over their opponents; technical differences between the F-86 Sabre and MiG-15 were less important than how those differences were exploited. This is why John Boyd believed ‘orient’, the second ‘O’, was most important in his OODA cycle, and is why a fifth-generation force must be cognition-centric.
Cognition is the process or action of acquiring knowledge and understanding through thought, experience, and the senses. The reason a fifth-generation force must be cognition-centric – rather than platform, information, network, or decision-centric – is because it is cognition that determines the quality and utility of each of those things. It is cognition that will determine how well a force functions when the information and network are compromised or denied. It is cognition that will cause a junior pilot stepping through their decision-centric checklist – in circumstances for which the checklist was not designed – to pause before firing the ‘fox 3’ that potentially starts World War III. Conversely, targeting an adversary’s cognition can cause delays, confusion, and – perhaps most effectively – profound misunderstanding of a situation. When the Israelis denied the network-centric Syrian forces access to their networks, it was Syria’s failure to develop the cognitive capabilities of its force that resulted in futile figure-eights.
The rise of automation and machine intelligence reinforces, rather than reduces, the centrality of cognition in future warfare. In addition to considering human cognition – which the Israelis exploited so superbly – planners must now account for machine cognition and human-machine cognition, and the interaction between different types of cognition. Algorithms are the ‘orient’ of a machine’s OODA cycle so, what Peter Layton has called “duelling algorithms” will remain as much a contest of cognition as that which occurred over the Beka’a Valley in 1982.
In the coming years, the ADF will introduce unprecedented capabilities to target and influence other people’s cognition, while defending its own. A variety of non-kinetic options, including the EA-18G Growler and nascent Australian offensive cyber capabilities will complement and enhance traditional kinetic weapons. Developments in network defences and management, as well as electromagnetic spectrum operations management, will give the ADF a greater ability to sense and display what is happening in non-physical environments. There have been numerous discussions and seminars about the impact that these new capabilities will have on the ADF, and the need for personnel that can understand and exploit them.
What has often been overlooked in these discussions is the types of cognition that militaries need to optimise the utility of these capabilities as a whole. This is less about developing specific knowledge, and more about the way in which people process new information so that they can understand enough about all these new capabilities, operating environments, and challenges, to rapidly combine these factors in ways to maximise their likelihood of success. Militaries are good at building specific knowledge, as this is the foundation of technical mastery; technical knowledge can be defined, measured, and tested reliably. And the discussions of new capabilities tend to centre on how the organisation builds more subject matter experts. Efforts to enhance broader cognition tend to centre on appeals to read and engage in debate. These are useful and enjoyable pursuits, but the approach is less than scientific.
Considering which types of cognition are best suited for future warfare is important for two principal reasons. Firstly, if the organisation can identify the modes of thinking that are optimally suited to operating in a networked, multi- domain force that fights for, and with information, then it stands a much better chance of being able to educate its people broadly to optimise those skills. Secondly, knowing what cognitive attributes posture an individual to contribute optimally to a force’s OODA cycle may allow the organisation to identify individuals early and groom them for key decision-making and decision-support functions. The option of selecting personnel for key functions based on aptitude would appear to pose quite a disruptive challenge to military workforce models based on hierarchy, time, and experience. But in many ways, applying this approach throughout individuals’ careers is simply an extension of how people are initially selected for roles through the recruiting and training processes.
Researchers in multiple disciplines are endeavouring to understand how and why people process information to make decisions. The psychologists Daniel Kahnemann and Amos Tversky demonstrated that human decision-making is not purely rational, and is subject to the whims of often unpredictable and frequently sub-optimal cognition. Efforts to understand those whims and how they apply to economic decision-making have revolutionised the study of economics, and saw Richard Thaler, one of the pioneers of behavioural economics, awarded the 2017 Nobel Prize for economics.
Psychologist and social scientist Phillip Tetlock has more specifically explored the cognitive attributes that position individuals best to make decisions, specifically forecasts in Tetlock’s experiments, in environments characterised by dynamism and ambiguity. Tetlock’s work saw him characterise forecasters as foxes – those who know many small things, are skeptical about grand ideas, and adapt to the world around them – or hedgehogs – those who have one dominant conviction and endeavour to make everything fit that theory. Tetlock’s categorisation was not based on an individual’s level of expertise, but rather on the manner in which they processed information. Many experts were indeed hedgehogs as they sought to fit the evidence to their field of expertise, but this was not always the case. Tetlock’s research showed that foxes were, on aggregate superior forecasters than hedgehogs, but his key conclusion was “how you think matters more than what you think.”
This conclusion is a vitally important one for militaries as they seek to grapple with the implications of future warfare and the types of cognition needed for optimal outcomes. The question remains open: esteemed historian John Lewis Gaddis argues in his 2018 book, On Grand Strategy, that good strategists should be a combination of fox and hedgehog. Importantly, the cognitive attributes that make a good strategist may not be the same as those needed for other functions. The research outlined above indicates that expertise in a particular area, which remains the foundation of military workforce development, may not be the characteristic most desirable to perform key decision-making and decision-support roles in dynamic and ambiguous environments.
Moreover, as algorithms begin to play a greater role, the optimal combination of machine and human cognitions must be considered. Are machines to be the hedgehogs to the human foxes in a human-machine team? Understanding and developing the types of cognition – human and machine – needed to win future wars is critical to optimising the capabilities of the future force. Investment in equipment and training will enable a force to fight, but it is cognition – individual, collective, and machine – that will determine how effectively it fights.
A version this article originally appeared in the May-June 2018 edition of Australian Defence Business Review.
If you are interested in cognitive warfare, we recommend you check out this short article from our friends at The Cove.
Wing Commander Chris ‘Guiness’ McInnes is an officer in the Royal Australian Air Force. The opinions expressed are his alone and do not reflect those of the Royal Australian Air Force, the Australian Defence Force, or the Australian Government.