In the Invisible Gorilla, Chabris and Simons briefly address how pilots with heads up displays sometimes missed planes in plain view on the ground. The technology gets pilots eyes looking in more or less the right direction to see more, but looking is not seeing. Beyond the illusion of attention, the heads up display does not help them see what they are not expecting to see. In the conclusion of their book, they state the following:
Technology can help us, but we must first be willing to acknowledge that automated judgments may sometimes be better than our own judgments-a difficult and controversial step. Still, we do not think that technological innovation can entirely solve the problem. A complementary approach to replacing human judgment might be to change our environment so that our limitations become irrelevant. In other words, if we know the limits of our cognition, we can redesign our surroundings to avoid the consequences of mistaken intuitions.
This sounds much like an airplane cockpit and points out some of the issues. Kathleen L. Mosier is an expert on this subject and I am going to look at her paper, “Searching for coherence in a correspondence world.” She not only helps better understand the nuances of the cockpit, but she also looks at the two primary constructs that enable us to evaluate our decision making. As she notes, Ken Hammond is the coherence/correspondence guru, but Mosier uses the framework well.
In this paper, Mosier traces the evolution of the aircraft cockpit as an example of the change of a probabilistic environment into an ecological hybrid–an environment with both probabilistic and deterministic features and elements. Mosier makes the case that judgment and decision making in a hybrid ecology requires coherence (rationality and consistency-justifiable decisions) as the primary strategy to achieve correspondence (objective accuracy), and that this process requires a shift in tactics from intuition to analysis. Although the examples in her paper are from aviation, the argument is applicable to any high-tech ecology, including medicine, military operations, or nuclear power.
Flying an airplane has been a primarily correspondence driven task achieved primarily through integrating multiple fallible indicators (Lens Model). Originally, flying was a physical task requiring some strength and accurate perception through the senses. Pilots often relied on railroad tracks to guide them. Soon, more and more information was placed inside the plane (altitude indicator, airspeed indicator, alert and warning systems), and this reduced the need for out-of-the window perception. Now, glass cockpit aircraft can fly in low or no visibility conditions. In these situations, seeing outside the cockpit is not even possible.
In a hybrid ecology, Mosier explains that cues and information may originate in either the electronic deterministic systems or in the external, physical environment. Input from both sides of the ecology must be integrated for good decisions to be made. Mosier says that characteristics of the electronic side of the ecology do not lend themselves to intuitive tactics.
These electronic systems have two notable characteristics: precision and opaqueness. The precision and reliability of the electronic systems are so high that the ecological validity of the information they present approaches 100%–if and when, Mosier notes, complete consistency and coherence among relevant indicators exist. Opaqueness means that what the pilot often sees is an apparently simple display that masks a highly complex combination of features, options, functions, and system couplings that can produce unanticipated, quickly propagating effects.
Although correspondence is the ultimate goal in the cockpit, Mosier states: “the primary route to correspondence in the physical world is through the achievement of coherence in the electronic world.” This means knowing what data are relevant and integrating all relevant data. Coherence competence also means knowing the potential pitfalls that artifacts of the technology, such as mode errors, hidden data, non-coupled systems and the recognition of inconsistencies in data that show lack of coherence. A tragic example of this was the flight of an Airbus 320 in 1993. The pilots apparently intended to make an automatic approach using a flight path angle of -3.3 degrees, but they executed the approach in the wrong mode where -33 created a vertical descent of -3300 feet/minute and the airplane crashed into a mountain.
Another issue is that the analytical processes required to achieve coherence are much more fragile than intuitive processes and small errors or details can result in disastrous correspondence errors. Moreover, coherence achievement can be hindered by technological decision-aiding systems which often present only what is deemed necessary. Thus, in their efforts to provide an easy-to-use intuitive display, designers often bury the data needed to retrace system actions. For example, Mosier notes, the trend has been to present data in pictorial formats, a design feature which has been demonstrated to induce intuition. Attitude indicators, collision avoidance warnings, etc. tend to do things like turn from green to red to allow quick detection of some abnormal states.
By making things look simple and intuitive, the decision maker can be lulled into forgetting the numerical data that signify different commands or values in different modes. Mosier states that the design of systems technology and automated displays needs to help the pilots recognize when an intuitive response is inappropriate, and that more analysis is needed. And certainly, the pilot must know how aircraft systems work and know the functional relations among the variables in the system.
This article was published in March 2009, three months before Air France 447 crashed into the Atlantic Ocean. With regard to that disaster, Dr. Mosier was quoted as follows:
“Training can not and should not be a fix for a lousy design.”
I would be interested to learn more about her thoughts on the particular case of Air France 447 if anyone has seen anything.
Mosier, K (2009) Searching for coherence in a correspondence world. Judgement and Decision Making, Vol 4, No. 2, March 2009, pp 154-163.