Thursday, 22 October 2009

Can we trust our judgement?



So my book has still not arrived, partly due to Amazon's slow despatch system and Royal mail strikes which ironically raise a decision making question* which I will point out after I summarise what I learnt from the reading I have done so far.

I have read chapter's one and two of the book (to my relieve it is a fairly compact book but rather informative nonetheless) and also the article by Dhami, M.K. on Psychological Models of Professional Decision Making. The first chapter of the book was a useful introduction to terminology and a general idea of what we refer to Judgement and decision making. It also touched on the debate about what rationality really is and what characteristics are required in order to be rational. We, as humans, have adapted to our environment through evolution and can make rational assessments about the world such as we make the judgement that that the moon doesn't actually change size just because it appears bigger one night and smaller the next. However, for certain judgements we require cues to help us make judgements such as motorway speed indicators when coming of at a junction etc. The most interesting thing about this debate is the decisions we make based on bad judgement when it comes to gambling etc. So it made me think; how rational are we and to what extent are our decisions misguided by wrong judgements?? And if we aren't able to rationalise properly with all the information we have, then what information are we basing our decisions on??

There are ways in which we can try and figure how it is we come to make these decisions. The Social Judgement Theory looks at what factors are important to people and what 'weight' do hey place on certain information. If I am going to be completely honest I am still getting to grips with some of the maths involved in the methods they use to predict these weightings but I think I have got the main idea of what they show. What I do understand is that using previously gathered data we are able to work out roughly the decisions that a certain group of people such as doctors or judges will come to given a certain amount of information, we do this by creating Actuarial models and working out how much weight they place under each cue. The interesting bit is that people show to be inferior to actuarial models when it comes to predicting decisions, this findings show that people are inconsistent and don't place the same weighing to cues in every case.

I must point out that when I was first reading all of this, I did think "so what?" I mean we are only human and we are bound to make misinformed decisions. But there is a real problem here, what about when these decisions are important; what about when someone's life is at stake; what about diagnosing patients on irrelevant information. We could just replace people with actuarial models but that would have a very bad reaction from the media and I am sure people wouldn't want to be replaced by a statistical equation. Even so, there is so much evidence out there showing how bad we are at decision making.

Another question is: what are we actually doing if we aren't using all the information available. When we have to make 'fast and frugal' decisions what models do we use? The 'fast and frugal' heuristics seems quite convincing as they point out that we use short cuts when we have limited capacity. We could be asked to make a choice on a question that we don't know the answer to so we would search through our heads and we would choose the first one that sounds familiar (i.e. recognition heuristics) if we cant recognise it then we search a bit further in our heads and look for the most likely then so on and so on (you have probably read about them so no need for me to list them all). The evidence for the use of these heuristics which includes the Dhami,M.K. article pretty much convinced me.

What I have been able to summarize from my reading s that we are not very consistent and good at making decisions when under pressure and the use of such shortcuts in important situations highlights a real issue. Measures should be taken to insure that, in situations such as trials and medical diagnosis, decisions are being reached using all the relevant information available and not just one or two cues.

Anyway I thought I would leave my question till the end as it puts a bit of real life context to these debates.

* I actually decided to buy it on Amazon as I thought it would be more efficient and the link was right there on the webpage, however I did know from previous experience that it can take a bit long and also I was aware of upcoming postal strikes.... had I taken all this information into account I probably would have bought the book when I arrived at uni and got it by now (BAD DECISION MAKING ON MY PART)

2 comments:

  1. With regard to Dhami's study of the magistrates, any thoughts about how we might make them think more analytically? Or should we make them think more analytically?

    ReplyDelete
  2. Well I suppose a constant reminder that they should be taking all cues into account, or maybe we could simple show them what the equation could predict and they could use that as one of their cues? However, it could be argued that we should just let them make the decisions that they are making now as it make people more at ease if it was a human and not a mathematical equation that should be doing the decision making???

    ReplyDelete