Nick, Peter and Fraser talk about what the presentation of shop fronts tells up about signalling and quality.

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 taking a look at the state of Hillary Clinton's health. So Peter if you can lead us in, tell me why are we talking about

Hillary Clinton's health and why does it matter? So I think it was this week in the very height of the campaigning for the presidential elections Hillary Clinton was taken ill and taken to hospital with I think suspected pneumonia and there's quite a lot of media footage of her being pretty feverish, sweaty and stumbling around trying to get into a car and things so she was obviously quite ill and this would possibly be not a good thing when you're running for president because you want your president to be both sort of fit and able in body and mind and able to take on such a complicated and important office. So that's I guess that's the simple question

answer to the question you know why do we care about her health. I think one of the things interesting is just from from this side of the Atlantic is I don't think we care that much over here about the health of our politicians. Maybe we don't like them very much. Yeah I'm not sure if they love them that much in the US either but but I think it's I think it's I understand it's sort of common practice or it's usually in the US to sort of for the for the president or the potential president to always release their health records and actually their financial records as well and over here we don't really do that and I guess it's to do with being head of state as well as being head of government.

But well I think that I think there's a there's an interesting cultural difference there which I don't think we want to get into much but I think the American the American culture is more about being on show and being demonstrably this is what we can do this is who we are whereas UK Western European culture is much is usually a bit more conservative reserved and it's you you you observe people's behavior and outcomes more than advertising who they are.

Actually I think this reminds me of a something I read about about London it was a book called In Search of London by H.V. Morton where he talks about the fact that European royal families have these enormous schlosses you know half the city taken up with a huge castle and whereas whereas the British royal family have St James's and St James's Palace in Whitehall is considered you know it was widely regarded as a bit of a joke among European royals it was like a tumbledown townhouse but he makes the point you know that actually the British royal family has been going for longer while revolutions swept all of these other things away and I think that sort of brings us on a bit to what we were going to talk about even though yeah okay we might think it's a bit weird for the Americans to be so obsessed with with whether Hillary Clinton's got pneumonia but we all do things like this where we you know we judge we will judge a candidate for a job by you know the shoes he's wearing or how nice his suit looks which is not you know people are we're voting for the president so because they make good decisions we're not voting because they're healthy and like but likewise you know we employ someone presumably because they're good at their job and unless their job is to stand there looking nice in a suit um why do we look at things that do not appear to be superficially related to the thing we're interested in?

Like and so I guess is it is the question that we are trying here effectively we're talking about signaling okay and is it the case that the signaler is looking to signal or is it is it more um sorry guys um a reminder I've got a podcast recording today um yeah so my point is so is it a question that the signaler is consciously making these signals or is it is it that they're doing it subconsciously and the as the receivers of this signal that we are

primed to to receive this for what we're looking for? Well signaling theory um is a study of exactly this um and uh it's it's just that the the the sender and the receiver are the two parties in any sort of signaling transaction um they they both are in a competition um often it's sort of it may be related to um whether or not there's a viable mating candidate or a sort of collaboration candidate if you're hoping to do something build a bridge or whatever um and uh it's it's present in all in all parts of sort of at the animal kingdom and very prevalent in our own um and and signaling theory suggests that you can uh both infer from somebody else's signals and imply with your own signaling some unobservable intrinsic quality about about the about yourself um and so you so these signals are often proxies for um for indicating some other

kind of quality and it there is the signaling theory does tell us what the characteristics need to be for a signal to work so to work in this context right here we have this this sort of a competition i want to find people who are good say so i'm looking for high quality people but low quality people would would like to be mistaken for high quality people that's the problem that there's an asymmetry of information the high the low quality people know their low quality but i i can't tell the difference that for a signal to be successful it's got to be costly but it has to be more costly for low quality people than it is for high quality people so when you start looking around for signals that that appear to do this job you you actually start finding them everywhere in fact there's a uh a u.s um uh theorist called robin hansen who who believes that signaling is really fundamental to human interaction but if you take an example of something like wearing a nice suit um you know it's it's hard it's easier for someone with lots of money to get a nice suit so that tells you that person's got lots of money well we're not employing them to have lots of money but it's easier for people who are good at their job to earn lots of money so so people who are good at their job one way or the other find it easier to wear a nice suit so that sort of works as a signal yeah um you look at something like having a posh accent uh that's uh easier for people who went to a good school to have um you know for whatever reason it just they're raised like that if you're raised in a certain way it's very very hard to disguise it um and then you can think about that in the context of other types of signal that we use the suntan is an interesting example where um you know a long time ago it was more expensive to not have a suntan because you know most people had to go and work outside um and so if you had a suntan it showed that you were probably a manual laborer of some kind um and then of course you know as people more and more people worked inside even people who weren't uh you know rich or or high status worked inside that no longer became a signal it no longer became important for women to paint their faces white and um then suntan started being a signal of being rich enough to go abroad um you know then having a suntan was a high status thing at that point because it was more costly for for high status for low status people to get one and then we get to where we are today where it ceased to be an indicator of anything and now suntans are um if anything you know a a signal of low status so you know that that's the key thing we should expect to see signals where there's a value in differentiating between people and uh it's harder for the low quality but it doesn't even matter what it is it could be totally unconnected but as long as it's harder for the low quality people or the low quality things uh to to send the signal then

it's going to do its job okay so where does this leave us how do we use this information what what's the aleph thing what is the what's the aleph angle on this what do we what do we do with this

information i think the uh i think the answer is a lot actually it's something that we can think about in lots of uh uh in lots of domains so you know if you're trying to find a good shop for example now if it's this if we're in that similar situation so if there's high quality shops and low quality shops um you've got to think about what if i'm choosing this shop on the basis of the way it looks um you know what what signals are going to be harder for low quality shops to send and and that's where things like you know michelin stars are a good example i mean you know it's it's it sort of seems obvious to us that if you get a michelin star and you're a restaurant you must be good but why do we bother with that uh it's because it's much much harder for for rubbish restaurants to have michelin stars um and uh you know so so if but whereas green grocers for example they don't need to look like expensive shops from the outside because you can tell the quality of the things that they've got uh very easily you know there's no it's there's no asymmetric information problem so so this kind of tells us that the kinds of the kinds of things we should expect to to need to worry about signaling when it's hard to tell what the quality of something is but where there is still a high and low quality um uh dimension to it so for example you might expect dry cleaners um not to particularly need to signal by having a flash looking shop um because uh dry cleaning is neither high quality nor low quality but as long as they don't destroy your suit it's either dry cleaned or isn't so what so that so what so on a practical

level what this could potentially let you do is produce a an assessment framework for maximizing the amount of information you gather from observing um the signaling given off by other people or by businesses as part of your your due diligence process about whether or not you're engaging their services um and and such a framework would wrap in and wait accord rate weight appropriately what um factors are most likely to be prevalent within that context be they the cultural or the

technical so this i think this is what i was going to lead towards so is what you're saying there that we can build a model that will predict let's just say in the case of shops that will predict what kind of shops will where someone will look nicer than others and not even look nicer the thing is

of bring it back to where we started then peter so what does this tell us about hillary clinton

for example so um the the as i alluded to earlier you off you want some reassurance that who you're voting for is going to be able to perform uh as you as you need them to perform in their job and president is uh no different it's probably one of the hardest and most complicated jobs in the world uh with one of the great huge amount of responsibility so you know you're an indication of someone's physical health um is a direct indicator of how likely they are to survive the term and be functional and function well within it um but at a biological level your physical health and your mental health are extremely linked in fact you know argue with this is all the same thing and it's just a very artificial division between the two so if you're if you're physically unhealthy that you're more likely to be mentally unhealthy as well and not make good decisions and not not not perform in that job so well um so that that so this this demonstration of health and this um going out of the way to look bold and strong and um and is is very tied to being uh the perception of of a good leader um that you know when we were when we were living in caves and on the plains you know the tribal leader would probably have been a combination of the smartest and the strongest person amongst the group um and people would sort of look to them to

make decisions and and lead them into into battle so even though and i mean this doesn't this is no longer something where we we need to look for an explanation an explanation that makes this rational for this specific case even if hillary clinton had a bad cold i think on some level because we have evolved to read signals we would still start to wonder about her her mental health simply because we've evolved to pick up that connection between between um physical and and mental health um but these things are very old i mean some of the things peter is talking about uh like taking risk taking for example is you get a lot of risk taking behaviors um particularly along like men who are competing for women uh it's been found that women like risk taking in men and uh men obviously like to take risks you know they like to um sort of write they drive less safely and you know do extreme sports um and the and that the reason is that you know the theory would say well people who are stronger and fitter um and more fertile uh they are just it's less of a risk to them like they they're less likely to die because they're more likely to survive these sorts of trials um so so i think you know on an evolutionary level that it sort of explains why we care about the health the physical health of presidents because we've evolved to you know quite rightly read that as a as an indicator of mental health

even though in this case it isn't yeah yeah also what's sort of striking in this case is that we can look at someone with a cold or whatever it might turn out to be and go hmm not sure about this um and yet at the same time look at someone who's bright orange with a bizarre sort of mop top sort of hair everywhere and given what you've talked about the link between physical health and mental health and just go doesn't someone who wants to project that much um virality virility rather someone who wants to project project that much virility we don't you know we don't sort of question or maybe we do question over compensation there is this issue of

counter signaling which i thought it might be might be a bit too far for this particular podcast but the the idea is that if you are sufficiently high status and have sufficient a number of other signals you no longer need to worry about the the signal that everyone else is using so um you know the classic example is not wearing a tie you know when when um people uh wore ties to differentiate themselves from manual workers who didn't because they couldn't because they were operating machinery or whatever having a tie was a sign of being a professional um you know but the thing is that someone who's a boss with a big office and uh you know they don't they don't have to do that anymore they've got enough signals to show you that they're high status so they don't need to wear the tie it's where you get the boss without a tie and his underlings who are wearing ties you know and that that's so so in the case of Donald Trump uh you know actually it's quite a powerful signal to say i don't need to i don't need to look like you know all of these other people i'm Donald Trump my name is on my aeroplane you know i don't need to conform to what you think uh the you know a successful politician looks like and and that is you know again it's not contradicting it's sort of explained by the the same theory there's another good example of that

um in um industries where your output and your your prowess is highly measurable take for example software engineering where your thing either works or it doesn't and you can very easily measure how many things you have produced um there's a there's a amongst that culture job interviews are often extremely casual and people often dress down to go to the interview trying to find a dirty ripped t-shirt because that sends the message that i'm all about the code not about my appearance you know i spend my entire life in front of the computer that's my main focus maybe i should

take that tip if i'm doing some job interviews myself or maybe that's where we're going i've been dressing down in the wrong industry that's what i've been doing wrong um wearing ripped t-shirts to bank interviews well i i do you know what not to banking interviews but i once i work i used to work for Chatham House and um which he was quite illustrious dare i say it's slightly stuffy institution sometimes and my wife uh once had stopped me on the doorstep as i was leaving for work because i was wearing um like a slightly garish hawaiian shirt okay and i was just i was i was going off to work that day if you had a nobel prize that would be fine yeah yeah and unfortunately that wasn't the case and but i i do remember feeling look i am you know i can do this i you know but she said no you you just really shouldn't go to work looking like that and especially not so uh so yeah um okay well look um can we close this off can we sort of seal this all off i mean

what would you want to any anything extra to add here well um i mean ideally we wouldn't need signaling so in an ideal world we would be able to tell the quality of something and you can see a lot of a lot of technology has sort of driven us towards being able to get instant feedback on things and and so we might speculate that you know the better sensors get the better we are able to just sort of tell how good someone is uh because there's more information about them um signaling should be less of a thing and i think you know if the the peter's model there of you know the things that we currently think of as important to demonstrate such as you know time keeping and um and uh and wearing wearing smart clothes might in future become less and less of a of a useful signal because they will you know that we'll be able to measure the quality of

someone's output directly okay and at which point the the the market will sort of decide who the winner is um and there won't be uh as much kind of as much signaling required so i think i think there's i think the dry cleaner example um because there's a sort of standard of except a standard of goodness it's either clean or it's not yeah um and striking has probably sourced their customers from immediate local areas so a few blocks around so as long as there's not too high density of dry cleaners a dry clean there that is good will succeed um so there's they don't need to message about they don't need to shout about their service as much so all i need to do

then is is you know from a business point of view if i'm looking for is just just go on google maps or something and work out where there's a where there's a hole where there are no dry cleaners pretty much i can't go wrong just it sounds like there's boom yeah there's no marketing costs

there's you know that's a very good strategy that that shops employ they do a sort of they do a statistical analysis of an area i've seen this work for bike shops destination bike shops and destination pubs where they do some research work out where uh there's a gap um that's within a sort of commuting distance either by you know by car or by foot or by bike yeah and deliberately choose to put a shop there for no other reason than it's somewhere like people are likely to go

well that sort of makes sense on one level but all the more so i'm saying it's even easier if you're a dry cleaner right because you don't have to worry i i suspect setting up a destination you

still have to make sure you don't destroy people's suits i mean let's just you know let's get this

in perspective there's always that but i think it must be easier to open and run a dry cleaners than it is a destination hipster coffee cafe bike shop i think well am i am i wrong i don't know

i don't know why i'm asking you guys dry cleaning shop are probably quite high yeah you're doing a

disservice to dry cleaning there might be much more in it than than than you're giving it yeah maybe i'm overly respecting i suspect you know unless you unless you suddenly found a new town that didn't have a dry cleaner in it i think the market's probably reasonably saturated fair point

it could be because it's relatively easy right anyway um we've we've run out of time but just before we do are we signaling to each other right now i mean i actually literally i'm singing i'm pointing at you peter but are we signaling anything to each other i i think it might be time for a coffee okay that's a quite a strong signal yeah okay look let's wrap there um thank you very much gentlemen um thank you as always for listening to the cognitive engineering podcast i've been here with nick hair and peter coghill of aleph insights and until next time thank you