Recently I was reading a paper about how anchoring bias has a strong role in how different geologists interpret data, because they base their interpretation on data they associate with a location. For example – Dover is known to have chalk cliffs, therefore if you see a cliff in Dover, it must be chalk. Which lead me to think ‘can geologists ever operate without using anchoring bias?‘Before I launch into my reasoning for this, I should probably define what anchoring bias is. Anchoring bias or focalism, is a type of cognitive bias that usually happens in decision making, whereby the first piece of information (the anchor) becomes the basis for all subsequent decisions – even if this is illogical. Most often the phenomena has been tested using numbers, to quote an example given by Grau and Bohner – when participants are asked: Is the Eiffel Tower higher or lower than X meters? Followed by the question: How high is the Eiffel Tower? The answer to the second question will invariably be influenced by the value of X given in the first question. It has often been mentioned that anchoring bias is easier to demonstrate than explain (Strack and Mussweiler, 1997) which certainly seems the case, with explainations varying between selective accessibility (the anchor provides a point from which to test hypothesis and the data for it is more easily accessible – Mussweiler and Strack, 1999) to emotional state (those who are ‘sad’ and who generally demonstrate less bias – aka a more realistic view of the world – than those who are happy, seem to be more susceptible to anchoring bias – Bodenhausen, Gabriel and Lineberger, 2000) to expertise and experience (whereby experts base forecast and other data extrapolations on previous values, which may in some cases lead to very inaccurate results, Campbell and Sharpe, 2009), which bring me back to the geologists.
The reason I thought of geologists in particular being susceptible to anchoring bias is because, having worked as an Identification Officer at the Natural History Museum in London, I know that hands down the first question I ever asked someone who brought me something geological to be identified was ‘where did you find it?’ In fact I would even go so far as to say that if someone didn’t know where an object was from initally, it would be practically impossible to give a detailed ID. Yes I would be able to tell them roughly what their object was, but for detail I needed a location. This is true of all geology, where you are defines the types of rocks, mineral and fossils that you expect to find. Most intersetingly in this case however, it is just as likely to define what you WON’T find. Anchoring bias in geology precludes us from choosing certain rock types, fossils or environments of deposition in an area simply because that’s not what the maps say. And we love our maps.But is there a problem with this? I mean, as researchers show, it’s likely that all experts experience some form of anchoring bias (in fact we all do) – so what’s the big deal? Well I guess as someone looking at communicating geoscience and trying to understand how people perceive geology, the influence of bias is a big deal in building trust, and we don’t even take anchoring bias into account. If we unknowingly encourage anchoring bias in interpreting our data, then are we skewing our results and presenting them with more confidence than we should? I wonder how many geologists would change their interpretation of soemthing just because the loaction changed? And yes, context is important, but do we let it control our interpretation too much? It may be that for geoscientists, anchoring bias is just a fact of life, but I think we should all be aware of it and try to take our own natural biases into account when communicating our data.