The physicist, Neils Bohr, said: “prediction is hard, especially about the future.” The statement was made in jest, but no matter what the fortune tellers, prognosticators, priests, or experts say, humans cannot perceive the future, but can only infer it. These inferences enable us to assess risk and react to opportunities presented by the environment. These inferences range on a continuum from the automatic to the conscious, of deliberate actions. Nevertheless, contrary to popular opinion, the vast majority of inferences occur either entirely or mostly on the subconscious level. We incorporate environmental inputs and make predictions without any conscious knowledge of it. For example, perception and memory are largely inferred reconstructions of the outside world. Our brains infer the details of a historical event or an image. Experiments with visual illusions and false recollectionsuncover the automatic inference mechanisms underlying memory and perception, and how they lead us astray in our predictions. Almost all of the most important inferences occur without any consciousness and we cannot spend a minute of our waking life without making mostly automatic inferences. On the other hand, we can spend hours or even days without ever engaging in deliberate reasoning.
We often confuse inferences and reasoning in part because inference is commonly thought of as a process in which you reach a conclusion based on evidence and reasoning. However, 19th century enlightenment philosopher David Hume saw the folly in this definition and stated, “nature must have provided some other principle, of more ready, and more general use and application; nor can an operation of such immense consequence in life, as that of inferring effects from causes, be trusted to the uncertain process of reasoning and argumentation.” He seems to be in fact correct as humans use a wide variety of specialized, modular inferential mechanisms that work on specific tasks. A module is defined as an “information-processing mechanism that is specialized to perform some function.” Each of these modules has a specific function associated with solving evolutionarily significant problems (i.e. survival and reproduction) from our history in thepaleolithic. These modules are sometimes connected to each other and to certain inputs and feed each other information but sometimes they are not. Therefore, restricted information flows (information encapsulation) through many of these modules as needed and bypass other unconnected modules. This modularity explains why our brains (or us) can contain so many inconsistencies in our judgments and decisions. [Why Everyone (else) is a Hypocrite].
The world is somewhere in between being completely chaotic and unpredictable and completely predictable and deterministic. If this was not the case, no relevant inference could ever be drawn either by logic or probabilistic systems. What makes predictions possible are dependable regularities or patterns in the environment. Similarly, the Emergency Department is often times on the brink of chaos and outcomes of patient presentations somewhere between completely predictable and utterly random. Like the Paleolithic, in such an environment, decision making and predictions are often not the results of deliberate reasoning but the outcome of automated, largely subconscious general purpose inferential mechanisms that handle specific types of information based on an expert’s level of experience. These inferential mechanisms are a blessing but can also be a curse in the Emergency Department as the predictions are far more often right than wrong. However, these predictions are wrong inevitably because the decision maker was fooled by a cognitive illusion and made an automatic, subconscious decision without the consistent vigilance that makes emergency care nebulous and humbling. In fact, it is unrealistic to expect a human (even a highly trained one) to exercise that kind of self-vigilance from patient to patient and shift to shift. Therefore, from a perspective of an Emergency Physician, there must be an external system in place that can identify these pitfalls in real-time at the point of care.