Human Factors Minute is now available on Spotify: Check it out here!
Nov. 12, 2021

E225 - Should We Trust Computers Over Our Doctor's Expertise?

Recorded live on November 11th, 2021, hosted by Nick Roome, &  Barry Kirby.


| Recorded live on November 11th, 2021, hosted by Nick Roome, &  Barry Kirby.

 

| Programming Notes:

| News:

 

  • Doctors should rely less on ‘mental shortcuts’ when deciding patient care
  • Image Source: Pexels

 

 

| It Came From:

 

 

 

| One More Thing…

Let us know what you want to hear about next week!

Follow us:

Thank you to our Human Factors Cast Honorary Staff Patreons: 

  • Michelle Tripp

Support us:

Human Factors Cast Socials:

Reference:

Feedback:

  • Have something you would like to share with us? (Feedback or news):

 

Disclaimer: Human Factors Cast may earn an affiliate commission when you buy through the links here.

Transcript

Welcome to human factors yeah your weekly podcast for human factors psychology and design. Hello everybody welcome back to another episode human factors cast this is episode 200 25 recording this live on 11/11/2021 I'm hosting chrome I am joined today by Mister berry Kirby. Good morning how the W. I am I am devilishly good very. Thank you again for midnight midnight midnight and what mood lighting with us. We got a great show for you tonight we're actually talking about man versus machine and who to trust when you go to the doctor's office and later we're gonna answer some questions in the community about what to do when recruiters are weird what do you want to see from a start menu end the most important questions when hiring a UX designer and or researcher but first we have some programming notes here. I were excited to announce that our coverage of ergo acts is started. With the previous out now so we've we've done a preview of ergo acts. Yeah I we had some scheduling conflicts during the conference itself so we'll have to wait for that 30 day window before we can actually see the talks but if anyone went to go axe and wants to leave us a voicemail about your experience please do drop us a line you can go to our website and click the little voice mail in the bottom right hand corner and in other news we have joined team sees so if you are unfamiliar team sees is an effort. Across many different content creators with the goal of removing 30000000 pounds of trash by 1/1/2022 from the ocean so they're basically they've equated $1 of donations to 1 pound of trash removed from the sea and this is started by mark Robert and Mr beast we are now part of team sees. And to kind of make sure that this effort does not go unnoticed we here at human factors cast are gonna be producing 8 standalone human factors minutes available for everybody so this is also available in your feed now you heard the first one this week it was about what team season is and what our plans for teams these are kind of what I just told you but you can listen to it now. We have 8 episodes coming out through the end of this year and like I said that's free to everyone normally our human factors manager for patrons but this is too big of an effort to just keep to that small pool so go check it out team sees.org. And I think I think we'll leave it there let's go ahead and get into the you know what you're here for here for the news. That's right this is part of show about human factors news very what's the story for us this week. So this week we're looking at whether doctors should rely less on mental short cuts when decided patient care so I was with Ms and analytics now commonly used by professional sports in sales forecasts Langdon isn't lending decisions and car insurance providers research is suggesting this type of doctors to stop relying on they use a few mistakes when making decisions about patient care would lead to cognitive resources so your mistakes the middle show because we talk about using health electronic health records more more than 86000 infant deliveries they found the delivering physicians what influenced not only by the indications of the current patient so the one that that committee working with but also the outcome of the most recent delivery. The noticing the physicians use these battles. But the the physicians use these mental shortcuts like experience expertise or training but the research demonstrated it's a common tendency even through the most more experienced doctors. They've offered several suggestions including the use of artificial artificial intelligence decision support systems to help physicians overcome their reliance on maladaptive hubristic all the real decision rules. So Nick can you convince should doctors give up that good feel for ARE. Yeah that yeah I think so it makes sense to me. Doctors are humans humans make mistakes A. I. has progressed over the years by a lot. And I do think the future of of health care is artificial intelligence I really do. You know I think there's there's that happy medium of having sort of the doctor play a supervisory role when it comes to artificial intelligence and that's kind of the future is where the system can look at sort of hundreds of thousands of data sets and determine what's best for the patient and as long as nothing's officially the doctor goes all right yep good let's do it. And they can kind of use their expertise to to judge that fishing this right at that anyway what what are you thinking about this article. All right my goal is to go along with what you say but also then maybe play devil's advocate with that because. What is the future and will go into we'll go to the the the and a bit more detail I'm. Is it day yet is it actually truly thinking in a way that connects do diagnosis in a way that humans can use big sticks up that's true but actually the way that the the human brain works is still way more powerful in the way that he dives into problem solving and be able to pull together up struck these information only one of the quotes out to the article when you read is most the time here sticks to save time and resources and they produce pretty good outcomes but in some 6 situations pretty good is not good enough Sir even then the sort of admitting that the way the humans do things does work most of the time. But then I I'm going to send the overarching and a couple of episodes of house this morning I don't know if anybody watches house but that is. Looking at the M. a group of people who do in this advanced diagnostics and and I begin to read the we had problems the the the not standard and so they have to use all sorts of of their own come to ability to solve the problems or at least a call to the script writers they do but the. That's kind of the kind of thing is is the you that level of diagnosis and and having to do the investigation reveal and what a day I'll be able to do that in its current form today my good feelings that no it wouldn't you know we did it we just don't know enough but that's not to say we not terribly far away from the future. M. and then my only last autumn is doing that checklist is going to have an impact in people's times in inside people's live at the time that it takes to go through a checklist where if we have the hero 6 we recognize what we're looking at and just dive straight to it which check listing doesn't really allow you to do. So I don't know I'm I'm I'm I'm going to sit on the fence as well as painful as it could be and say that yes we could do it but we're not there yet but I think is it I think this is a good point to maybe use dive into some of the the H. R. principles around it so to divert some of that yeah I think let's let's start with maybe here mistakes in the health care and then get into sort of algorithms piece of healthcare so let's take it from the psychology side of things and yeah they they mention these mental shortcuts those are heuristics and it there's some research that suggests that some of the most common decisional sort short cuts that are used in medicine. The field of medicine overall ours our availability anchoring and confirmatory here a stakes where. You know representative ness overconfidence and band bandwagon effects are also prevalent in medical practice so we're looking at I just listed off a whole bunch of names there will go through them one by one and talk about them but those are typically the most common in medicine and medical so we want to talk through these. Here is sticks or I guess mental shortcuts one by one very. Yeah we condemn any I'll kick it off with the availability heuristic Sir it's the the the I guess best way to describe it it's the embassies based on you know things that you already know things the familiar to you things that. Lend themselves to what you add to your current knowledge and and then for whatever you're saying fits with you that's with yo yo model and so because that information is available to you then you'll automatically jump to it. Yes let me have anchoring which is kind of latching on to that first thing that you hear or understand that first piece of information right so. In a way I think this article is almost. Suggesting that anchoring is happening with the most previous. Delivery at least in this case they're they're kind of using that as a as an example and and we can talk about that when we get to more on the article discussion but I'm kind of taking that one piece and making holding it as the most true when when making a decision. Yeah I am you can also dive into complementary so doing something and then whatever so actually taken actually has the effects of that so things that confirming what you what you think you will you know whether you know it might not be right but Sir because he keeps on confirming the your initial thoughts then you're going to and yourself and into that a lot more because it keeps it keeps on. Give me the confidence that the what you're doing is the right thing. Yeah I mean you have the representative nous here Ristic this is kind of. This is kind of looking at an event and comparing it to what already exists out there and and basically. The thinking or using that mental short cut to think that what is true in one case is representative of many cases. And and again this is kind of with that previous delivery you might go okay they had a caesarean it might be true that SS Arian is a better. Mmhm. Sort of method of. Birth in in this case because it was I use that last time therefore it might be good this time it's kind of representative of the sample of that doctor. I I think I think I didn't okay job describing that one. It's it's basically admitting probabilities okay what you did to these those easy as you think what yeah yeah maybe my my overconfidence A. N. in my did my judgment to be able to deliver this is it far outweighs my the accuracy of the way I'm just grabbing these I'm not is a great example of overconfidence basically of comedy is the you have more confidence in your judgment don actually the the accuracy more in more. The of what you're actually delivering. Download all the at the acoustic your judgments itself so you're basically just being to mail the about. Yeah the last up here we have the bandwagon here Ristic a bandwagon a factor this is you know if everyone's doing it then it must be right you know everywhere we talk about crypto in the pre show everyone's investing in crypto it must be right. Mmhm so that can happen in medicine too so you know one example would be if there's a procedure that has worked in several cases as kind of a breakthrough procedure it. It might be that everyone sees that as the new big thing and. Maybe there's some additional study that needs to happen before. Yeah that is in fact to this day the status quo basically jumping on because everyone else is doing it. So I think that's that's a pretty good overview of some of these he respects that we're talking about in the healthcare field you know and I think just overall. Here is 6 are mostly positive but again kind of it depends factor right it can have sort of these negative consequences if it's not being used correctly right we've talked about them a little bit in a negative light here in the sense that that they're almost at biases biases in in this case but I think in in a lot of cases that actually does end up. Affecting patient care positively the fact that they have these mental shortcuts that they can take to get to where the patient needs to be can often you know sh it is a short cut it's it they're they're taking that step to. I'm. Cut and not cut corners I add that sounds negative but it is too it it's it's that the speed accuracy tradeoff that we're talking about here how quickly do you get patients care. Depends on the speed but by which you can diagnose those problems and so they are using these mental shortcuts to cut through some of that. Mmhm does that process noise. Yeah another I think you're right that because I think the again it goes back to what we sent earlier in that. I'm particularly general practitioners all people who were in say the emergency rooms and things like that have to make really quick decisions really quickly based on fundamental that one experience of the medication so soon as they see something that looks like something else the Dave that they've had experience also it it's maybe based on so the familiar so blatantly in on that availability all or anything like that. 99 percent of the time that's going to take you down the right route to be able to save lives because that's that's what they're left the game isn't it the the bandwagon effect again is is really I think we'll be quite strong and good quite well used because 99 percent of the time it'll be the right thing to do I think the the point of the article is eighties perhaps saying. The the it's not names at the time isn't as good as it isn't as good as 100 percent the time so it is that it is 96 right off so I think there's a an element here of just how our algorithms used in in health care and how he's a I go to get to that so is it worth diving into some of that do you think. It is I I do want to just wrap up the here mistakes bit right I mean you know we talk all the points that I completely messed. Well well I do want to just wrapping up the ins and say you know that there is this huge it depends factor right there's the. The personality the level of expertise and experience that a doctor has that might influence their decisions it also is influenced by context and conditions of the situation that they're in and I think all this is important to remember as they're taking into account some of these heuristics they're looking at these situations through the lens of all this context and that's something that I think is going to be important to capture when you're looking at artificial intelligence and machine learning right and so that those are kind of. I'm some mitigating factors that we can use to reduce the likelihood of some of these. Some of these algorithms are your mistakes. In the doctor's heads being used incorrectly so yes let's talk about algorithms and machine learning what you jump into it because you you you got you brought us a great reference here. Well I'd I'd I do like to and the money the vast amount of pay you give me for common do these international against bins I am yes No we I was actually looking up but it's a really nice the coincidental that I'd let the weapon are allowed to try to weaponize last week with the child and she belongs and human factors and then he just just white paper. Comes to be a from a special interest groups that I'm a member of. So see I'm I'm just old but he qualified to be talking about this except the fact I've just I've been bringing actually a lot of the DHS sites rather than the the the A. I. side but there is some. The use of artificial intelligence and artificial intelligence is such a which is what worries me the mom is such a catch all term because there's many different types now old. We don't think the common the activist at intelligence so things like machine learning. And things like that and a new field which I think was not not nearly as it's being told what more mall is artificial general intelligence so it's worth defining the difference between the 2 sides this intelligence is very specific and if it's applied in a very specific so case where is the artificial general intelligence is something that you should be able to engage with I'm just almost talk normally give it a whole lot of random facts and any come out with the right of the right sort of answers so you can be used in in the general case which for health care I think it's really exciting because that is the equivalent of the emergency doctor being given anything to be able to take it take it all information so. There is this could be seen as a 3 city 3 approaches in their bid 3 different types of AI algorithms that used at the moment and so if we dip into them so the if I had the first one which is busy supervised learning. Which is where the data is old sorted and organized before being presented to the algorithm and the algorithm is that to minimize is to minimize area minimize ever so we we training the day I like to do it's to to do its thing it to reduce the the differences between its own output so what is perceived to be the correct output. Your job is to unsupervised learning yes unsupervised learning is basically where it independently independently discovers the patterns in the data by itself and this is the scary one to me because it just kind of results in so how did you figure that out I just had a look through the data and finds a thing so yeah we're talking about how these different approaches. Are kind of used in these algorithms. For the purpose of. Thinking about how we feed it data from health care side of things right so when you when you have that supervised learning you have. Sort of these these training sets that doctors will go in and kind of a sign in and make sure that all the parameters are there where unsupervised learning is Hey we've detected that this patient has a heart disease and we don't know why but it's there and then you have the last one here which is reinforced brain force meant learning and this is basically using a reward function right so it's it's given that input data and then the output is measured according to that. The success of that output right and so that's kind of in simple terms the 3 different approaches and so when we're talking about A. I. and algorithms in this context I think we're mainly or least I like to think about this in terms of. Like a decision support system I think there's that is what I think of when we think about algorithms and health care we can certainly have these. Algorithms especially the unsupervised learning ones those are just crazy to me where you know it'll it'll come back with the diagnosis and the doctor goes yep that that's right how did you I didn't even and it's kind of shocking to everybody but the the doctor then ultimately makes the call but that algorithm help them get there and so I don't know do you want to talk about anything else with the house with the differences in algorithms and how they impact how you know health care yes this this one bit that came up so when when I went back to buy Virgil degree actually there was we saw studied about fishing jobs that put this going back quite a few years now. But one of the cases that was leading at the time. There was Alectra was day she was one of one of the leading in in the office does well at that time and she was working healthcare and it was a low back pain and date she was developing algorithms to diagnose low back pain and and that was all great but actually we shoot shoot this related to was the the initial bit of research went through and the. It basically meant that a lot of the times that she was that the algorithms would diagnose it low back pain the doctor going no no that's not low back pain that's something else that's you know eccentric xetra. When we all get that got me first then actually turns out that her all do agree the was right all the time and it was the doctors that were wrong so when you look at the difference between supervised unsupervised learning for added this is been so been in healthcare that's really important the stuff that he's been labeled and organized inside this is what the outcome is. If the diagnosis is not necessarily right Sir we need to be quite careful about how we teaching array eyes to to do things because with the still so much about healthcare that we just don't know so that's it's an interesting and it's such an interesting topic to be to be delicate moment because of the exact that reason. Yeah well let's talk about some of the issues in health care from from the artificial intelligence perspective right this is this is also coming out of that that white paper that you shared with us and we'll put a link to that white paper in the show notes too so everyone can go check it out. But you know it by there's a there's a nice graphic on there the kind of outlines all the different issues and we pulled out a couple of them here so automation bias. Let's just talk about a Mumbai once automation bias is an issue right we have this over reliance on sort of these aids in decision systems. And it's kind of the human tendency to take the path the the road of least cognitive effort right so you're you're basically looking at these results and going up that's right instead of critically thinking about whether or not that is in fact the correct outcome or if the kind of what went into that decision right we got to carefully craft not only the output so that way the doctor understands how the system got there but you know have them trained on the artificial intelligence system itself so that way they they know exactly what goes into it we'll talk about that with training too but not tackle this next one here yes that's what I said M. expression trustees William sending the output of the I and have confidence in that decision because if you don't if we do have any almost goes back to automation by says well that if the computer gives you and gives you an output and says it is the S. then we more like to go over them that's cool and move on with it but it would need to have a. Be the bills to interrogate the I. so it tells what how why and how we came to that decision on the level of trust that its goals in the output because it's fruitless will take you quite easily and so and say oh it says this step forward today so but even though he got side at 50 percent company tried to get it then we need we need to understand that as well to be able to make that part of our judgment. You can take them the next one yes the next one here is human AI teaming and this in a nutshell is just basically who handles what part and how graceful is the handoff between each of those parties right the human and the A. I. and it might be that maybe the A. I. handles most of it in the human is just kind of there to deliver the news that could be one model you also have another model where. The A. I. does part of it it asks the doctor about a you know a series of questions they provide some input and then it comes back with another output. And that would be kind of V. the hand out so we got to think about how exactly the doctor's interacting with the A. I. in this in this case and then we also have to think about sort of what parts the A. I. handles what parts of the human handles that's that's kind of that whole human AI teaming aspect just to make them gel together I think working with another human being everyone has their specialties where do you. Sort of focus efforts on the system. I'm not really leads into the next which is it which is around training because we go to this isn't gonna work on day one and we got to we'll everyone will decide where we go we go to train in practice working with the with the A. I. so they so do the the medical practitioners of the I. working well together doing conducting their own exercises conducted their own training and almost experimentation as well to understand where it's best employed how do you interpret the A. I. outputs easy I'll put exactly what your what you need to come out and help how you interpret confidence levels but what he gets to you Sir it's all practice makes perfect piece is that it's really important I think with the you would understand how to interact with the A. I.. Yes so there's interaction with that A. I. and then the next one here is well how does that A. I. N. doctors you know that that relationship between staff and patients and A. I. and whether or not you know if an AI has more of a dominant role does the doctor than do more work away from the patient and how does that impact the patient's perception how does that impact sort of. You know the bedside manner that that whole thing is really important for for doctors and so that relationship then changes when you introduce artificial intelligence into the mix because that doctor that needs to communicate that back to the patient they need to have some awareness of how that decision about their. Care was made and so the doctor needs to be X. able to explain the A. I. and so there's this really complicated relationship between the 3 it's a it's an interesting like try out right and then last up you wanna talk about this one this one 's fun yes Sir this is only the actually in him frustrated we we don't we can motivate very often but it's the ethics side of things because this is going to be absolutely. Critical to make sure that this is that it's successful so there's a whole bunch of bits where ethics need to be considered so firstly is in the design and the development of the I. I. I. itself that there is already proven to be a identifiable bias in AI models of depending on who's been designing and programming them so we need to make sure that them ethical issues I in that within within these these shoes around privacy so patient privacy patient health home and making sure that they you know people's records are nice income generated the the autonomy the practitioners so who has the final say if the if the M. A. I. comes out with a radically different output how does that compare with what what the doctor says I or the partition of size who wins. And for the Bentley told the balance on Sunday what will the but but if it's about minimizing harm to patients needs so the whole ethical issue is is providing a whole stream of research the still go some way to go yeah and who who was in the wrong if it's a mis diagnosis right you know the doctor's going off with the A. I said well the developers of the algorithm and at fault for you know malpractice because they you know as the it's a it's a whole complicated things so let's get back to this article here. And we'll just kind of go over the research one more time so they're they're looking at 86000 cases rack records of infant deliveries and these delivering positions were again kind of influenced by. The current patient but then also what happened previously. And so one example that they use in this. This article here is that when a when a position experience a negative outcome with vaginal delivery they're more likely to choose to deliver the next baby by caesarean and vice versa so you know this this whole. Or having complications here then the next one we're just gonna go straight to see section. Because it's you know it that that prior influences than influencing their decision. Yeah and I think but I think what is really quite needs about the the papers that that I can only doing the the the study said that that. The professional to acknowledge that this is a problem but it's not the positions filled it's a way to be trying to it's it's the it's a leading all difficult nothing else at the moment and so you know they they going to be the best tools of bagels but now with the advent of A. L. I. and under the and techniques so it isn't just a Ali but he it straight you know just because the algorithm doesn't necessarily mean it's artificial intelligence it's just that those are the tools out there I'm. We now have the ability not just to go simply with your good decision anymore. We got the good ability did to do it by the way is that full do we either need to start looking at doing that way doing all pushing a lot harder so the this official said that you know managers know the decision makers in business and elsewhere no longer simply go with the guts and the doctors often remain reluctant to introduce such information by making medical decisions for patients. Which kind of makes sense given that it's there it is their profession and it's also that room injured they have their own personal insurance if it goes wrong. Yeah I think I am I yeah I I don't know where we go from here. You know there's. There's still a lot we don't know about AI and how it works and. Really you know all those human factors issues that we talked about in that white paper that you share with us is is really considered considerations we need to make or take before. We implement this full blown you know I think slowly dipping our feet into the waters here is probably the best way to go and slowly changing over time verses this rapid implementation of all these new systems. But then it's also like will do just keeping with the medical analogies to you yank the bandaid off and just jumped D. A. L. right in I don't think that's the right approach personally but you know where do we go from here very well for 20 now we mean this in the state told development and trust because firstly the system's going to be these AI systems are going to be developed in a way that that trust by practitioners but they're also trusted by patients and in the grand scheme of things right now. They and he goes back to a severity around you know we we label everything ever sought to look into this a I bother. If it goes wrong we soon just because of the maturity of it we gonna throw out the baby with the bath water to keep him with us delivery at. See who keep looking back it's like professionalism you know we we we going to do that because this isn't just some big mistakes happen probably not through the fault of the I. but it lets you go back to you know some other issues will end up having I would be just too scared to engage with and if you if you don't choose to turn around and say well actually I'm going to use use this system this **** and help diagnose you and me it's already got bad press that people are going to be against it regardless just because we don't like technology so I think it's a watching brief I think we got is we gonna try Kathleen the evolution and the the design for and we a a a softly softly approach. But you put out by room to room yeah what will people think yeah I was gonna say billable let's hear from everyone else so we asked the question for tonight's social bots tonight's episode is all about doctors algorithms and analytics. Our social thought for this week who would you rather diagnose your medical issues N. A. I. A. with access to hundreds of thousands of patient profiles or doctor with years of experience and bury you won't believe this. But out of the out of all the votes we had 100 percent as a I. What I'm gonna say that's surprising. The present but at the same time you gotta consider our audience I think you know we have an audience that kind of knows. It's what's out in the field or at least doesn't infer yeah that that that's that's probably true well maybe have way more confidence in it you know he did yeah we just professed that the that that we do so hello let me ask you would you would you rather an A. I. or doctor. It depends on the walls are you did it depends I did not I did I did a sort of house version of it depends I think you know if I was gone if you something that I was going in for the walls and you know he's a it was kind of out of the mail stuff that he needs compatriot Alonso thing I think. I could tell C. N. A. I. bean you celebrities it because it would do that maybe Foles what's been really serious like maybe roam the brain or something like that some people said is there already then maybe the allies it's a good thing that because you know you wanted to pull in the until it all all the facts so actually that they both cases yeah our product of the I. D. that's interested in the go the. But I think I think I'd go they I too because because like I said it's it's scary how much a I can come up with something based on what it does and we don't know how it gets there. I do wanna bring up one comment here this is from Katy Jurich's open saying that correctly they say I'd actually choose option 3 health care provider who uses assistive automation to support diagnosis so there's the answer for the night should we trust computers over our doctor's expertise. Barry I say yes but I think ultimately it'll come down to sort of a mix of both and who handles what that's still to be determined but that's that's the answer for tonight folks. Any closing thoughts on this one berry yeah I I'm B. Katie Katie right yeah Katie's right all right thank you to our patrons this week for selecting our topic and thank you to our friends over at Kelley school of business at Indiana University for a news story this week if you want to follow along you can join me on office hours every Monday where I find these new stories we do post the links to the original article's on our weekly roundup center blog or you can join us in our slack or discord for more discussion on these articles we're gonna take a quick break and then we'll be right back with some things around the human factors community right after this human factors cast brings you the best in human factors news interviews conference coverage and overall fun conversations into each and every episode we produce but we can't do it without you. The human factors cast network is 100 percent listener supported all the funds are going to running the show come from our listeners our patrons are our priority and we want to ensure we're giving back to you for supporting us pledges start at just $1 per month and include rewards like access to our weekly Q. and A.'s with the hosts personalized professional reviews and human factors minute patriotic only weekly podcast where the host breakdown unique obscure and interesting human factors topics in just one minute patron rewards are always evolving so stop by Petri on.com slash human factors cast to see what support level may be right for you thank you and remember it depends. Yes huge thank you as always to our patrons we're always happy to have you here especially want to thank our honorary human factors cast staff patrons Michelle Tripp patrons like you keep the show running thank you all so much for your continued support now this is where I usually plug more about patriotic but I've been told by our treasurer that we are putting all of our eggs in one basket with a tree on so. I'm I'm I'm now I'm going to promote the merch store did you know that we have a merch store some neat designs over there that include it depends shirt show logo like on the hoodie I'm wearing tonight other cool designs based in human factors culture want to support the show and look good doing it go to human factors cast.media slash P. slash store. It can be done I thought I could sell it but then I don't need to stand up and show the shuttle that show the whole deal because yeah here you go up the I hop in the envy. Yeah I mean it I will say it's a very comfortable hoodie and I I I don't like doing all the self promotion but do this is a really comfortable hoodie we partnered with spreadshirt. And you know they're they're close actually quite comfortable and I do enjoy wearing the merch because of that reason. So we had a March provider in the past that wasn't too great comfort Weiss writes messy all right anyway enough of that let's go ahead and switch gears and get into this next part of the show we like to call came. That's right this research all over the internet bring you topics the community is talking about this week it's reddit do you find these answers useful give us a like where you're at to help other people find this content we have 3 of them up tonight we have let's just tackle them one by one here this first one here is what do you think are some of the most important questions when hiring a UX designer and or researcher is by oatmeal man John the user experience sub reddit they go on to write also do you think that a researcher position should be separate from a designer it's been in my experience that I've hired designers or researchers who are prima donnas and tend to get emotionally attached to their work so they cannot make a job objective decisions let's talk about this. Mailing address think we need to have a chat I'm so old because of these to be to begin with should should the reset be separate from a designer to be honest it depends on the size of project you doing in my name my experience I've tended to pull both together I could be doing researching one they are committed to designing the next in the forties big enough you can separate the balance. Even then I don't know whether there is a vast amount of benefit from doing that because most people will be able to do both both roles but I mean god forbid that you get an emotionally attached to your work and you get really bold bolting to the topic that you're doing I mean that's just a drama waiting to happen isn't it of course you do that everyone does that if you don't if you're not doing that but I suggest you do the job properly in the first place so mail manager I think you need to go now to take a long hard look at yourself and look at people who actually get motivated by the work they eat them if you if you don't get emotionally attached I'd be more worried. Sorry that may be slightly angry while I eat you're really you're really going hard on oatmeal into here let let's let's talk about this so I think I'll be able to does. The problem here is that there they see this emotional attachment as not being able to make objective decisions about their work. And you yeah I get that but I mean at the same time if they're good at their job they'll know it's one thing to be I think the confusion here is passion verses. A motional attachment because you can have passion and look at something with sort of this critical I still right in fact I do a podcast all the time I have an attachment to this podcast I love this podcast but I still look at it a critical eye and go okay maybe the story that night wasn't so great or okay we could have pulled in some extra stuff here to supplement that. Yeah I do it my work too I'm passionate about it doesn't mean that I can't think about it critically in fact I would argue that the people who think critically about it. Love it more because they want to just make it the best thing that they can be in those who are attached to the way things are are just in love with the way things are now and don't really want to be better they just kind of accepting the status quo anyway my advice here. What do you think are some of the most important questions when hiring a UX designer researcher. In terms of making sure they can do the job correctly I think. You know kind understanding what their process is I think that is the biggest. Sort of a question that you can ask them is what is your process because if you can get into their head if you can understand their thought process about how to approach it a decision or how to approach research how to approach a design what steps they go through to modify that research or design how they interact with users how they interact with developers how they interact with people on the team I think those are some of the most important questions you can ask. A lot of times I find that some of the. The stuff that we do in the day today kind of changes and we talked about this last week right our day to day changes quite frequently this is a matter of what tools were using at that time. I'm process can be relatively the same overall but. Again it just kind of depends on who we're talking to who were interfacing with but that part of it is so critically important that when you're trying to bring someone else on the team you want to make sure that they can gel with the way that things are done in your organization and that they can gel with you if you are a hiring manager. Anyway that's my $0.2 on any last the last notes. Yeah I think V. maybe get some examples of stuff to do before if that's if that's possible not always possible but yeah that but the whole Jelling into your team just because they work well on a on another team does this I mean the work will of yours because you have a different dynamic every dynamics different and yet it. Don't be necessary hung up if that processes don't align to what you already know why the cause effect quite successful the might just have a different way to do it doing things that's also quite good fun. All right let's get into this next one here recruiters like you don't have enough experience but let's keep in touch and connect in a couple of years this is by throwaway person to to to on the user experience I've read it and they go under right so you want to reject me right now gain experience at a place who will give me a chance then come back to you when I suddenly gained more experience make it make sense recruiters don't know how ridiculous they sound sometimes very have you encountered the situation where there's let's just talk about this from the perspective of ridiculous demands from either somebody who's hiring or a recruiter have you experienced this yes and. I don't necessarily think it's it's an out of order thing especially human factors will because we are we have such a small domain really but there are more people coming in and if we could just look at you and say well actually for the position I'm looking at the moon because they could be looking at multiple positions a lot of the time so you might not have the right experience for the position they're looking at trying to fill right now but we will keep you on the books for the future it's something else more useful comes up. Then but then that's not necessarily a bad thing on the receiving end of it it's frustrating because you are sitting there going well actually I've got I've got skills I've got ideas of call this sort of stuff yes I know I have the experience just why should so either if you think I'm good enough to give me a shot at it so but it's also some of the. Suit recruiters and HR practitioners that the note by note that the teams that they do have a fairly standard process that they go through to to get people on board so as much as I think I agree it's frustrating and it's hard to hear I don't miss everything gets the the recruiters fold I think it's it's unfortunate business I'm afraid. Yeah I mean like I I've definitely seen the advertisements were there like okay we want a. Someone to work for us for $20 an hour and you know I have 10 years of experience and also know how every single language of coding and also know how to do research effectively and also know every design program I've seen those and those are absolutely ridiculous yes. The requirements for a job that I think do you come down on that okay we don't know what exactly we want this role to be that that's what that reads to me as. In this case here where there. Rejecting based on. Years of experience I think. From a recruiter standpoint Hey I mean. I guess I get it but at the same time it's it's not it doesn't feel great from the person who doesn't have the experience. I'm and. I to me what this reads as is the person who reached out to you their job is recruiting and if they see you as somebody who fits another position that similar to that they don't want to burn the bridge with you that's what I'm understanding. So they're looking for somebody with slightly less experience maybe instead of a senior role you're looking for more of a junior role then they want to keep in touch with you because you might be a good candidate for that position I'm not saying that that's what's going on here that's kind of just what it sounds like. Based on the verbiage that this person is using reject is pretty harsh so I'm wondering if maybe they applied to something in the recorder screen them and that that was seen as a reject rejection instead of a a screening I don't know that there's a there's a subtle difference between the 2 but I think it's an important distinction. So yeah I I get I get the frustration from from this person but at the same time knowing how things work sometimes I get it from the other perspective too I don't know it's not a very good answer. And I'm sorry for that but I I think. You know maybe maybe I don't maybe don't burn bridges that's that's my advice don't burn bridges if they do have something for you later they might reach out. Any other closing thoughts on that one very yeah I guess I've I've been on the other side as well I've been that person to say you know you just not the right fit for us right now I read like you but you don't therefore you know that that the type of specific job I'm wanting done you're just not you don't have the experience I think you'll be able to go on with it but I would you know would love to keep in touch so yeah I've I've been there I don't think it was trying to be that's all these I would hope that would be able to get us to because at the end the day the sign yeah we wanna keep in touch with you that's a good thing that the disk the this can be something that like brought your personality and and generally you they just want to have a bit more experience because it is not miss any time experience it's just practical experience of being you know being in the in the rights all roles to do the the right levels in your to to do to do things what it is just life I think. Yeah it's rough I let's get into this last one here what changes do you want to see in your operating system you pick this one I'm really excited this is bright sexy I got 69 I'm H. C. I. subreddit did you pick it because of the user name I'd I'd I'd love user data didn't read about it but I think that's just something else. They gone I'm taking a human computer interaction course in the first project is about creating a different start menu for operating systems and then testing and analyzing the results of this new build needs a man for inspiration from you all you can propose simple alterations are completely new innovations such as creating it as a central pie menu or any other form you can imagine to improve the functionality and approachability the start menu's berry now you pick this one for a reason so I'll I'll let you just go ahead and answer this one. Well I salute because I thought it's quite useful with the but with that with the launch windows 11 and they've gone they've changed the start menu from being left hand I'm good on the on the low about to be in central which on the face of a I can see that it's it's it's different. We give that you know it's it's a step change but then I can also see from a design perspective if it's all one thing from that from the center how's that going to look at the house the U. tabs on display out that we've got the spinout from both sides and not moving from left to right mother it just starts triggering lots of things in my 8 with my just design and set up so saying there is the settlements that I don't know how they're going to work. In terms of wall into the actual question what I would see my start menu I'm. I think if we want to be truly radical about double what have the start menu a toll you you only need it that when you actually use it so if we get more more to touch screens and actually just one person you wanted to disappear until you to to to to to screen and you stop having a a menu pop up that I like the idea of having the the pie menu of of XP you star in some previous designs of done I am very recently where you touch and then you get you can stop spreading out around it. But I think for me it's about making sure that you have access to the the the corporations that you do above it in a really dynamic way so if you normally use word and excel and not so I think that's what they that that's what comes up first and then he dived into that a bit more but also having the the documents that you've used in the maps really accessible really quickly so I think having that that that level of dynamism but then also driving into into some static will be will be read to get. Can you fight without a comment on the from M. tech a tech sergeant Jan he says and that's why he they like Lennox multiple desktop environments you don't know if you don't like what you have use another one or make one of your own which is kind of faith you are in into the if you're into the the the next side of things and the world is your oyster and that's all thanks the flip side is also true though that the your ability to break your you if your new voice results of please also call them quite strong. That's almost that's almost like a the apple android arguments as well. I got I will give you yes I don't give you a lot more flexibility which we I think we we spoke about last week was on a police is very much you play it by that run by the rulebook which I think actually you don't the vehicle gives you a lot stronger. Usability not respected less flexibility so. Yeah yeah I would this is what you missed this is a tough question for me to answer so like designing the dream start menu is interesting because I think what you said is why why have one. Inferred for truly radical right Y. 8 why even have one I think if a system can understand the context by which you are we're working in and understand the next time you need to take it is ideal because then it's just kind of doing things on its own it's it's basically predicting your next move and from there it will pull up the appropriate. Program or whatever it is that you need. You know I I think when when I think about it from my perspective here I do think about sort of the the approach where. I. In in I know I am how games do like a start menu they they switch the mode on you it's not something that pops up overlays like mac or windows it's something that changes your entire mode you stop playing the game. To go to the next thing that you're doing and I think that for me works because it puts me out of this motive I'm working and I'm picking the thing out that I need to do. Mmhm and it's it's tricky to understand exactly what you need at that time but I think context based. I don't know programs might help. And then there's also like your favorite so I don't know there's there's a couple ways to go about it but I think the the coolest thing to me is is the predictive peace of knowing what is I'm. What what's coming next in the work flow if you're if you're making a Google document well you might need to do research on a topic and so you might need to go to a different tab and or or pull up you know a PDF and so it might know based on the content that you're writing what PDF's are relevant to that based on the stuff in your desktop and so it might just suggest those automatically. You know as an overlay and I think that's that's kind of a dream right. I do want to get to a couple of these comments here in the chat. Mads dune says I just I just love the synchronization of mac an apple I know how to work computer but I don't have the time to delve deep into the topic and learn more which is probably why I like the simplicity of mac an apple and then text search engine. Provides an alternative here alternatively you can use different interfaces for specific uses once gaming one for office work it cetera and I think that that's a that's gonna getting at that desktop. Perspective right like having having set desktops for certain modes right like this in my gaming desktop with all my apps for gaming and this is my and I like that approach I think that works I haven't figured out how to get at work king on mac yet I know it exists. You know I I do it for windows certainly you know I have my podcast desktop. He was he I I'm invented that yet I need to because yeah yeah vote for exactly this on another set of I like them for doing this podcast a like do I have the sort of like to do my my podcast on the other side of the road for everyday working and stuff like that so yes I need to play ball all right well let's go and get into this last part of the show is just one more thing it's really where we just have an opportunity to talk about one more thing Barry what do you have up this week. So this week has been so cool because I actually don't feel doing usability trials with real users we okay in the opener and actually just face to face tough it's been so nice to be able to do that it's it's such a. Such a refreshing thing and along with that of the desert but we did the podcast last week I didn't get my usual sleep in because straight down to the. To deliver to lectures to the human factor that craft maintenance again that was all face to face to win the elections are begins of out of all been a bit remote subjects going to guest lectures and so it was really nice to being able to to see people and and get be able to react to them and engage with them that was really cool well it's. The woman thing really that that's really important and and I need your advice on really is do you great windows 11 because it's come up on the desktop said it's ready it's downloaded click this button and you all your life will be revolutionized. But all you just scares me I don't know whether I totally should because yet because if it doesn't work and you know to have a you know I'm I'm the settled with my a mixing desk and and they stopped near the bottom if it doesn't work you can't go back in I don't know I I'm I'm waiting I mine is also ready to go ready ready to install I'm waiting on mine I think. The my biggest hesitation is the same thing as you there's a lot of settings that I have ready that I don't want to mess with and I am planning to do it when. Media outlets like tech media outlets say it's okay to do you know there's there's there's some certain features that I'm waiting for that aren't quite there yet and then also. You know it's it's a lot of set up when you upgrade and hopefully they've streamlined a lot of that but. You know do it when you have sometime that's that's the big thing right so like maybe over the holidays if you got like a week off or something just. Take some time update your system and then slowly fix it over the week. Yeah it's not gonna be super pressing right yeah it's the I think for me it might have to wait until the new year right I thank unless we can do some. But as you can do I mean the the the left of our goal is quiet quiet desa moan you know it's relatively new it's it just got it this year so it shouldn't have a problem in that respect but it's the like say forget the nuances behind it so. We shall see him I might it might be a long weekend jobs or 2.on the Friday and then fix it. Exam on the Saturday and Sunday hope you all are going to work again by the Monday. There you go and it's not me what about you my one more thing it is a little thing called M. Taylor and this is a this is gonna sound like an advertised I swear it's not I'm gonna put an affiliate link in the show notes you can help the show out but this is not paid for by them I just wanted to highlight it's a cool service and it really it just it's one of the first times that some felt like magic so. Basically what M. Taylor is is you get naked in front of a camera. And you'll turn around yeah hang on you you get naked in front of a camera you turn around a couple times and the send you clothes that fit your body perfectly and they use they basically use the camera feed to use that to feed algorithms and AI that then come up with a bunch of different measurements and then you know they cut out based on those measurements and it it feels like magic it feels like magic because you put in this order and then. You know like I'm I'm a big guy and so like finding clothes that fit me well and that I feel good in is hard and so it's just it's amazing you get this you just get the clothes in the mail and. And it just fits you and if you if it doesn't they do like the they'll they'll go and sort of do it again for free for you which is really kind and a great business model the clothes are a little pricey. But you know I think for for the comfort and for. The convenience of having something custom fit to you is huge. Like I said it sounds like an advertisement I just was delighted by the experience and I wanted to share it with you all. I will I will put the affiliate link down there you can get 20 percent off. If you want that type of thing by no means do you have to again it's just up the show and not advertisement not paid for. I think what that that's gonna be ever today everyone if you liked this episode we invite you to check out another episode that we did 223 on how artificial intelligence can improve learning content wherever you're listening with what you think of the story this week for more in depth discussion you can always join us on our slack or discord communities. You can visit our official website sign up for our newsletter to stay up to date with all the latest human factors news feel like what you hear you wanna support the show there's a couple things you can do one leave us a 5 star review that's you can do that right now go and leave us a 5 star review or if you're watching listening to tell your friends about us word of mouth really helps the show grow I can't tell you how many people said I only found you because my friend mentioned you they listen to your show and 3 consider supporting us unpatriotic you have the financial means of doing so or fire March I don't know do that because our treasurer says anyway as always leads to all of our social and our website or in the description of this episode I don't think Mr Barry Kirby for being on the show today where can our listeners going find you they want to talk about algorithms. Algorithms then you need to find me on Twitter at Basel disco okay or Camillus into me interviewing people on my podcast whether to the human factors podcast at 1202 podcast.com. As for me I mean her neck road you can find the streaming on twitch every Monday from 4 to 5:00 Pacific for office hours and across social media at Nick _ Rome thanks again for tuning in human factors cast until next time. 8 the defense.

Barry Kirby Profile Photo

Barry Kirby

Managing Director

A human factors practitioner, based in Wales, UK. MD of K Sharp, Fellow of the CIEHF and a bit of a gadget geek.