Cognology is an emerging feature of the introduction of AI tools and techniques
Cognology is the counterpart of ‘technology’. Where what we think of as technology has often been associated with physical labour and physical work, cognology is about mental work, thinking and reasoning.
Of marshmallows and boiled frogs
Where AI can be useful to help us address and even avoid the internal heuristics that guide and derail thinking processes, where the immediate pushes out the future, where bad thinking like bad money pushed out the good. This is the essence of the learning from the marshmallow test.
And what of the boiled frog, perhaps like Buridan’s ass, immobilised by either indecision or too many choices. AI can provide a useful third way to avoid indecision by offering a fresh perspective. We must always be aware of zombie thinking, which like dead zone regulation, persists and we are the weaker for failing to bury them (and keeping them from coming back again!).
Bad decision making: the art and science of solving the wrong problem really well
Sometimes, decision makers and problem owners use advisors whose problem-solving mind-sets are inappropriate to challenges which are not well-structured. In other cases, problem owners take advice from people selling a canned solution rather than listening to you and your needs. Because we like to think that people know what they are doing, we are dismayed when the advice is mundane or simply wrong headed.
And from their advice, we often do the wrong thing really well.
There are four types of Cognitive Challenges applicable to all decision making and reasoning and where AI will have a role
- Simple & Repetitive challenges require the ability to offer strong delivery of results consistently over time. AI can provide templated and pre-structured solutions to common challenges shared by many people or organisations.
- Complex & Repetitive challenges require the ability to offer very strong problem solving that brings together technical capabilities and knowledge with AI powered knowledge models where standard solutions may not be suitable, and for which AI is suited in reorganising information.
- Simple & Non-recurring challenges require the ability to extrapolate from experience to create a solution to a fairly simple problem, but one which has novelty. AI widens the scope of what is known in a manageable way.
- Complex & Non-recurring challenges require the ability to create new solutions, where strong new ideas are needed to create new solutions that haven’t been seen before for challenges that haven’t been seen before. There is tendency in humans to try to turn this fourth cognitive challenge to one of the other types; this has its roots in both heuristics and the behavoural tendency of advisors to prefer problems they understand or have models and tools to sell. These types of challenges can morph into new ones, called “wicked” problems, which have the bad habit of looking like they can be solved, but for which solutions usually spawn new challenges. AI, properly instructed, can sift through the complexity of information what makes a wicked problem wicked and provide new approaches and insights and perhaps tame this particularly difficulty cognitive challenge.