Nick, Peter and Fraser make some irrational assertions about decision-making.

For more information on Aleph Insights visit our website https://alephinsights.com or to get in touch about our podcast email [email protected]

Hello and welcome to the Cognitive Engineering Podcast produced by Tell Me Studios for Aleph Insights. In this series of podcasts we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm Fraser McGruer and I'm here with Nick Hare and Peter Coghill of Aleph Insights and this week we're discussing why you two might intervene to stop me from drinking but not intervene to stop me from rock climbing. So Nick, let's go ahead and answer the question. Let's imagine I'm an enthusiastic rock climber who perhaps also likes to drink a bit too much and one day you two take me aside and say Fraser you need to stop drinking. Is this a rational action on your part? That's what I want to ask.

So the first thing I suppose we need to look at is what we mean by rational and I think there's probably two senses which we need to consider. The first one is the kind of technical economic sense of what rationality is and that's quite straightforward. Economic rationality is about taking actions that on average will maximise your utility, the objectives you're trying to achieve. What is it you're trying to achieve and are your actions doing the things that will produce the highest level of that thing that you're trying to achieve. So if you're a firm you're trying to maximise profits, the rational way to behave is to always take the action that will maximise your expected profits and if you're interested in rock climbing it probably makes sense to maximise the amount of rock climbing you do. Likewise if you're interested in drinking you should probably drink more. So the purely economic sense of what rationality is is about taking actions that deliver the things that you want right. However that is slightly different to the sort of popular definition of rationality in which we tend to talk about things being rational if they're in some sense in your long-term interests. So someone who drinks constantly to the extent that they lose their home and their marriage and their job would be said to be behaving irrationally and we would want to stop that right. I mean generally people think we want to stop that but it is problematic to explain why that's different. Someone who is taking risks by drinking a lot may just really like drinking as far as they're concerned that may be more important than all of those other things. Likewise someone who does a lot of rock climbing and therefore risks their health in that way for some reason I don't think we do feel quite the same and it is quite difficult to explain what the difference is between those two from the economic theoretical sense this is just about

doing things that you enjoy that are risky. Okay so even though it's difficult let's talk about well let's try that I mean what is the difference then between rock climbing and drinking too much is that a fair question to ask because you've said it's you can't answer it's difficult but

well let's try can we? I think there's difficulty in defining what the utility function is because if as Nick says if you really like drinking and maybe destroying your health and your personal life in through drinking is a good thing because it gets you lots of fringe benefits which are much more difficult to quantify that aren't necessarily the popular accepted modern western values of not being a headless maniac who's just destroying himself for fun but the so I think that's that's where the difficulty comes is well if you really do like drinking and drinking is kind of your thing then it doesn't really seem very much different to somebody who rock climbs and rock climbing and thrilling themselves and the adrenaline and all the physical activity is their thing and I think economists get often slated for trying to model people as rational econs, robotic like people who are trying to maximize very narrowly defined things like their lifespan and their health and their income but I think that's unfair to economists because economists are what will say that it doesn't really matter what the utility function is you just as long as it's defined correctly it's really the nanny state who has tries to maximize things along very narrowly defined things but there is a difference which is drinking

let's say even in moderation is inherently detrimental to your health right it part of it is it's it's not good for your health whereas something like rock climbing if you do it correctly it's not a detriment to your health there's just a risk that it might be yeah but

but you've already assumed there that one of the things we're wanting to maximize is health whatever that is right however you define that the assumption that one of the things we ought to be maximizing is health has already imposed a some kind of objective set of values on behavior you know and why should we care about health more than we care about other kinds of pleasurable outcomes I mean there's a very famous quote from David Hume where he was seeking to sort of understand what the difference was between the beliefs we have and the desires we have and he said it's not contrary to reason to prefer the destruction of the whole world to the scratching of my finger right so what he's saying is it's not there's nothing inherently better about health or in his formulation there's nothing inherently better about the scratching of your finger uh compared to the destruction of the world the what the only way you can say that one is better than the other that the scratching of someone's finger is just objectively less bad than the destruction of the world is if you assume a set of values about things now now here I I think and I think this is what what this comes down to is that we we sort of basically think that health is the a right sort of thing there's a moral component that you ought to uh try and um try and be healthy but you ought not it should be somehow bad it's something there's something cheating it's cheating to take drugs like alcohol and enjoy yourself doing it that way somehow I think people feel like it ought to be hard work to get to get the things that you want yeah and rock climbing is a virtuous kind of activity whereas uh taking drugs isn't a virtuous kind of activity um I mean my my question to to someone who disagreed with that was let's imagine a drug with exactly the same outcomes as rock climbing so there's a very you know a small percentage chance of it severely injuring you but other than that you know it had no detrimental effect on your health would that would you would would you consider that to be as okay as rock climbing and if not why not okay but if it could replicate

exactly then I would personally I would say yeah there's no problem with a drug okay but look I want to come back to you on something because you said something which is a statement that only a philosopher could make right and I get fed up with hearing these things I don't actually I don't mean fed up now I mean just generally you know especially going back to university but you know the what was the statement by Hume uh repeat that statement that idiot uh yeah it's not contrary to reason to prefer the destruction of the whole world to the scratching of my finger sure okay so what that does to me one that tells me two things one that was definitely a philosopher who said that because that's the sort of rubbish that only they say right um and two um well actually looking a bit more logically then that sort of for me that that brings into question the value of reason okay because it was I think the premise where there was according to reason or something like that so that just makes people think well clearly we need to be thinking about other things than reason because I put it to you two that I don't think either of you actually believes that statement would really take that to heart yeah I mean the point is that's what Hume is saying

the limitations of reason right okay pure reason yeah he's making a statement about objectivity

saying that you can't you know if there's nothing nothing objectively better or worse than anything else okay the better or worse bit comes in when you're imposing your own values on things right and the the maximization of people's health and things these are quite sort of high-minded liberal-like attitudes which are relatively new um you know public health is a new thing um in the in human history so it's it's it's the and these are believes believes that these are important are imposed on the system by us okay so I was getting upset and hot undercover hot

under the collar for no reason I was actually in agreement with Hume was in agreement with me is

that is that the case yeah but it but it's still a really useful distinction to have to hand to say look there's a there's there's the things that we believe about the world so you know let's assume that drinking is bad for your health I mean you know there's various reports that one glass of wine a day might not be but let's just assume that drinking is a direct trade-off between something enjoyable which is getting drunk and something bad which is which is damaging your health now um you know the if you care someone who cares a lot about their health their rational behavior would be to not drink very much someone who doesn't really care about their health uh their their rational behavior will be to drink lots now uh the question is is one set of those values more is one of those sets of values sort of in itself irrational are there such things as irrational values and I think it's very very hard to support that idea that that there are sort of values that are just objectively better that it's better to want to be healthy than it is to want to be drunk um why do you think that well it's just very hard to make an argument that says that that's objectively better for the reason that actually simply to say this set of values is better than another set of values requires another pre-existing set of values and that's just another set of values and I think you know I think where you where it can be

justified where the anti-drinking campaigns and high levels of tax on alcohol and things can be justified is when you have a belief that there are parts there are there are people who out there who have access to alcohol who can't necessarily weigh up the pros and cons for themselves and they they need state institutions and things to help them with that and the the most cost-effective way of doing that is to hike the tax up and make it less accessible and to have a sort of encourage a societal attitude that disapproves of drinking rather than to outwardly ban it so not to not to remove it completely but to discourage it in order to help those people who struggle with these these complex calculations for themselves

yeah and I think here this is this is where this there's a there's a real puzzle a problem actually about sort of when when people's utility functions can change over time and reformed alcoholics have you know a very good example of this where you have they they know that when they're not drinking they want to stay not drinking right but that is very possible for their that for their utility function to change and for you know for whatever reason maybe they're a bit depressed or something you know reach for a first drink and of course you know that that's one drink is too many but two is not enough they would rapidly they recognize that they have the potential for their utility function to change into someone who all they want to do is drink at the expense of everything else and they've got to make a judgment they've got to decide which of those people they would rather be and that's a really tricky thing to decide and it's you know because it does require taking a stance and saying well I've got two potential you know is it I suppose you know is it is it better to be an alcoholic who's drinking a lot or a reformed alcoholic who isn't drinking at all and you know it's still ultimately making a value judgment you have to there's no there's no objective right answer but but that the I suppose the sober person is trying to make sure that they stay in a with a sober person's utility function I mean it's a bit like saying you know would would it if you let's say you like opera and you spend lots of money on the opera would you rather enjoy carrots just as much as you enjoy the opera and and that you know that would be a much cheaper thing to indulge you

indulge yourself with okay yeah well look are we going down the path I mean something I want to explore have we have we sort of talked here about cognitive biases at all have we sort of got onto that or not or okay well look let's talk about that and then there's something else I want to come on to but yeah where does cognitive bias fit into this or even or even societal bias so I

suppose that that there you have certain biases which might lead you to seek a quick fix and so therefore should think about the short term and that might bias you against long-term benefits like health and another reason why interventions are arguably a good idea is it well have I gone

down have I gone down the wrong avenue there no this is really this is really key right because it's it's the question when someone does something that's let's just say self-destructive okay smoking is a good example assuming that smoking is bad for you then now when you when you're smoking you want to keep smoking right yeah if while you're in the short-term withdrawal you want to keep smoking but I think it's easier once you haven't smoked for a while it's kind of easier to not to carry on not smoking now the problem is like is is that desire to keep smoking when you when you're currently in a state of active addiction is is that is that the result of some sort of bias are you are you biased can we say that it you know you're irrational you have an irrational desire it is quite hard to support that view it feels very intuitively compelling to say you know you're just being very short-termist I'm only thinking about tomorrow or the next five minutes and my need to have some nicotine but if I could somehow force myself to think long term then then I would realize I don't want to carry on smoking sure I remember having that internal conversation myself yeah so but but I mean but who says you should think long term that that's the issue is like well why should you think long term why is that better than thinking short term yeah at the moment you're a person who happens to like thinking about things in a very short-term way so at that moment are you being irrational if you carry on smoking this this question of whether or not I mean actually another example might be something like anorexia where you know is is someone who's anorexic have they got an irrational are they behaving irrationally are they behaving against what their interests actually are or do they just really really not want to eat um you know is it is it irrational behavior are they failing to maximize something or are they or are they uh correctly uh being rational in pursuit of a preference that we just

think is probably a bad preference okay look so to put this crassly I mean are we are we slagging off rationality is that what we're doing are we saying that you know because that's kind of what we said with Hume that and I think what you said he was saying is that reason has its limits okay so are we saying that you know just sometimes because it's what I'm tempted to do just go oh sod it just you know just forget all that just plow ahead and do what's fun kind of thing um and let's not beat ourselves up about it or to put it another way I remember I know there's a famous sort of example you know saying well who says that we live in a rational world we don't because if we lived in a rational world if you could plan it and execute it you know um one third or two thirds of the world would be taken up with soya fields or something like that you know so we could all feed ourselves um but that probably isn't desirable even if you could do it um so am I being simplistic is that what we're yeah I mean that's what I'm tempted to say just I think we're not slagging

off rationality per se we're we're what we are disagreeing with is people doing things and imposing their own values on people in the name of rationality saying that that okay they're imposing their own objective on somebody else and calling it more rational than something else okay uh than the subjects rationality um I think that's what we're disagreeing with yeah I mean

no rationality if you want to get the things you want right and you do because they're the things you want right you should be rational in the pursuit of those things sure rationality is merely about doing the things that give you the things you want on average um so so yes I mean you obviously it's it's really I mean it would be very hard to argue that someone ought to be irrational that someone ought not to try and do the things that uh that they um that they you know which will give them the rewards that they're after of course they should that's you know if not then you're not acting in a way that is going to deliver the things you want um so so you know rationality but the point but it's this is about where the limits are you know that rationality doesn't tell you what you ought to want that bit comes from somewhere else okay so look we need

to wrap up but I don't know if that's a summary in itself but one thing I'd be interested in is at least one institutional um example you have of this well to to to read to bring up a theme that

I bring up quite a lot there's um the rationality problem is at the heart of a lot of the kind of um tension within modern business and in post-industrial world the the business who has employees wants the employees to do as much as possible for as little as possible yeah the employees have more or less exactly the opposite aims they want to do as little as possible for as much as possible yeah um so the the employee is incentivized to maximize uh their their income while minimizing the amount they do um and that's that that's a totally rational thing so that both parties are acting completely rationally um but uh they are they're at odds so good you know good management and leadership is about finding the right balance between the two so that the business and the individuals both profit um from that relationship that makes sense so yeah no

yeah just that I think I think you can look you can find examples of what you might call organizational irrationalities there's organizations who fail to behave in ways that will that will maximize profit um I mean one one example might be uh Kodak you know their failure to um move into digital photography when when it was it was obvious to people within the organization that that was the way that things were going but somehow the organization simply failed to respond um you know so there is that the phenomenon of organizational irrationality and again this is about you know failing it's not about getting an analysis wrong or anything this is about knowing what the right thing is to do but somehow not doing it another example um which I've seen cited is the first world war and the way that that unfolded which um which led predictably I mean people could tell that the things that they were doing the different the sort of ultimatums uh the invoking various treaties um would lead to this war that in fact nobody wanted yeah um the the war just the first world war just sort of happened um no individual who's making decisions wanted that to occur so I think that's another another interesting example of you know arguably of of organizational

irrationality so I mean it sounds to me like one way to sum it up would be to say look uh being rational is not necessarily the end game it's not necessarily the the point and that rational uh decisions can lead to uh sub-optimal outcomes it sounds like to me no I'm wrong you're shaking your head well tell me why I'm wrong no because these aren't rational decisions that's

the point if okay we think they're rational but they're not well no no I think what we're saying

is this is a situation where there isn't well I think I think maybe maybe organizing some forms of organizational irrationality apparent rationality come from at a micro level those little individual decisions that people make that in aggregate cause a war or cause a business not to take a uh take it on an opportunity they appeared rational at the time at a very micro level um and there's no but there was there was no sort of unifying strategy that was rational that gave people a framework for making these these tiny decisions I mean I could sort of see

where I went wrong because actually it if it all depends on defining your outcomes right and defining your wants and so therefore if it's in line with that it can't be rational sorry it can't be irrational that's what I mean um got it okay I look forward to the test um so um right we'll wrap up there so thanks chaps um so thank you very much to Nick Hare thank you very much to Peter Coghill um that's everything for the moment for from uh the Cognitive Engineering podcast thank you and until next time bye