Don't drop the anchor: Recognizing and mitigating human factors when making assessment judgments under uncertainty

Abstract

One of the most important tasks a petroleum geologist undertakes is to recognize and incorporate uncertainty concepts into their geologic interpretations and estimates of volumetric parameters. This is crucial, because the first requirement for sound decision-making is an accurate and unbiased evaluation of uncertainty in hydrocarbon volume estimations. Reducing uncertainty is a worthy, but secondary, objective. However, not only is the information upon which geologists make such judgments imperfect, incomplete and potentially biased, but the pernicious effects of certain human factors can amplify bias and create unrealistically narrow ranges. A wide variety of human factors which influence such judgments have been revealed over recent years by applying psychological research techniques to geologic interpretation issues. Two of the most powerful and damaging cognitive effects are anchoring and overconfidence (Tversky & Kahneman, 1974). Anchoring refers to the tendency for one piece of information, such as a "base case", to inappropriately constrain estimates. Overconfidence is the tendency to estimate ranges that are systematically too narrow with respect to the state of knowledge. Together, these result in inappropriately precise predictions of hydrocarbon volumes ranges, with a concomitant degradation in accuracy. The phenomena are found to be powerfully ingrained, even in the minds of experts (Morgan & Henrion, 1990; Northcraft & Neale, 1987; Welsh, Bratvold, & Begg, 2005), who frequently have a disproportionately large influence on the final estimates. This paper describes and reports observations from both a program that ExxonMobil, with the Australian School of Petroleum, has embarked upon to increase awareness of these biases, and to develop tools to help mitigate their effects; and observations from the oil and gas industry as a whole. The ExxonMobil program is implemented through formal training and informal mentoring. Geologists are provided with exercises to test their natural predilection towards overconfidence, and susceptibility to being anchored. Upon revealing the influence of the bias, it is intended that the individuals take away an internalized caution against falling victim to these subtle traps in their future interpretations. The durability of such learning is not necessarily permanent, however, and frequent reminders of the phenomena are imparted in order to reinforce and refresh. This is intended to inculcate mitigative behaviors in the members of the geological community over the long term. Copyright 2011, Society of Petroleum Engineers.