Discussion: View Thread

Cognitive Biases

  • 1.  Cognitive Biases

    Posted 02-08-2021 02:26 PM

    This is a fascinating and I think fun topic and I am hoping others will weigh in with their own experience, particularly as seen or experienced in engineering practice. I first became aware of this topic in the early 1990 when first exposed to the concept of decision quality (a future topic) and in particular the disabling role of anchoring.

    As  a brief background, the notion of cognitive biases was first identified by Amos Tversky and Daniel Kahneman in work published in the early 1970s. The role cognitive biases play in everyday life has now become far ranging from Behavioral Economics to baseball's Sabermetrics. They are also also making their into engineering, e.g., February's free paper: Value of Information on Resilience Decision-Making in Repeated Disaster Environments.

    Examples of cognitive biases - also referred to as heuristics - include:

    • Anchoring - Why we tend to rely heavily upon the first piece of information we receive?
    • Availability - Why do we tend to think that things that happened recently are more likely to happen again?
    • IKEA effect - Why do we place disproportionately high value on things we helped to create?
    • Representativeness  - Why do we use similarity to gauge statistical probability?

    Finally, if you read and liked Michael Lewis' book Moneyball I highly recommend his follow up book the Undoing Project. The  provides the why behind former. There's also a nice article from the New Yorker The Two Friends Who Changed How We Think About How We Think that serves as a great intro to the overall subject of cognitive biases.



    ------------------------------
    Mitch Winkler P.E., M.ASCE
    Houston, TX
    ------------------------------


  • 2.  RE: Cognitive Biases

    Posted 02-15-2021 06:39 PM
    I'm really interested to see if anyone has some examples of this in their engineering experience.

    One way I have seen it is in developers or other clients having an extremely difficult time accepting changes from what they were initially told. (For example, if unknown geotechnical conditions are discovered on site, they may want a second opinion even though this information is being delivered by the same professional that delivered the initial analysis based on the borings.) 

    For this reason, I think it is critical that we, as engineers, are very careful how we present initial information since that first piece of information is often held so tightly as fact.

    ------------------------------
    Heidi Wallace EI, P.E., M.ASCE
    P.E.
    Tulsa OK
    ------------------------------



  • 3.  RE: Cognitive Biases

    Posted 02-22-2021 09:16 AM
    Part of your job is salesmanship to prepare the client for potential setbacks beyond your control. By nature, developers are Type A and in a hurry to complete a project. Setbacks bug them and you are the easiest person to vent on. That goes with the territory. With time you learn good clients from bad clients.

    ------------------------------
    Gordon England P.E., D.WRE(Ret.), M.ASCE
    PROJECT MANAGER
    Cocoa Beach FL
    ------------------------------



  • 4.  RE: Cognitive Biases

    Posted 02-22-2021 10:51 AM
    I can tell Gordon has been doing this for a while and I appreciated his comment " Setbacks bug them". I've literally told my design team on more than one occasion "I'm paying for your time and expertise, right now its just time and I need you to listen to me vent for a moment."

    For our younger engineers.....Yes, setbacks bug developers...like a lot!!! LOL

    This is because development projects are a lot like gambling.  You see....the goal of a development project is to make a profit on an investment. So what are the choices one has for a profit from an investment. A developer could put money in the stock market and leave it there, or could take their money to Vegas and put it all on red. But they choose to develop real estate for the illusion of control and because they take pride in the works and/or genuinely enjoy what they do. We hire experts to help navigate the rough waters. We get frustrated when our captain's run the ship through rough seas. We know they don't control the seas but we hope we they are good enough to see it coming and navigate appropriatly.

    ------------------------------
    Jesse Kamm PhD, PMP, A.M.ASCE
    Senior Vice President of Construction Management
    ------------------------------



  • 5.  RE: Cognitive Biases

    Posted 02-18-2021 10:04 AM
    I like what Heidi said. I definitely have seen some examples of anchoring bias as well in my own line of work. One of the things we do with hydrologic models is to re-calibrate them as more data becomes available to us. It is not uncommon for clients to be surprised and confused when this additional information changes some of the flow values that come from these updated models, even when the results are more favorable to them than they were initially.

    ------------------------------
    Christopher Seigel P.E., M.ASCE
    Civil Engineer
    ------------------------------



  • 6.  RE: Cognitive Biases

    Posted 02-18-2021 01:19 PM
      |   view attached
    I agree that engineers should learn more about cognitive biases.  A good place to start is "Thinking, Fast and Slow" by Daniel Kahneman, one of the two friends referenced in the original post.   The SEI Engineering Philosophy Committee (of which I am presently the chair) held a session at the 2013 Structures Congress where I presented a paper on this subject.  I have attached it for those who might be interested.  

    Bill Bulleit, PhD, PE
    Michigan Tech

    ------------------------------
    William Bulleit P.E., M.ASCE
    Professor
    Michigan Tech University
    Houghton MI wmbullei@...
    ------------------------------

    Attachment(s)

    pdf
    Struc13TooFast.pdf   3.85 MB 1 version


  • 7.  RE: Cognitive Biases

    Posted 02-23-2021 05:49 PM
    Hi Bill, thanks for sharing your paper! I downloaded and look forward to reading. Please give my regards to Professor Mattila. We were classmates at Cornell.

    ------------------------------
    Mitch Winkler P.E., M.ASCE
    Houston, TX
    ------------------------------



  • 8.  RE: Cognitive Biases

    Posted 02-18-2021 01:20 PM
    Heidi and Christopher - these are great real life examples of anchoring and thanks for sharing. I hope others will continue to add their own experiences. 

    One i will share in the meanwhile is the IKEA effect. During my career i worked with many individuals who were always quick to knock the ideas of others. At the time, we referred to this as the not invented here syndrome. Upon reflection, I think this was a manifestation of the IKEA effect. If they did not think of the idea, it was not worthy. This was tantamount to disproportionately placing a higher value on something of their own creation. 

    Mitch


    ------------------------------
    Mitch Winkler P.E., M.ASCE
    Houston, TX
    ------------------------------



  • 9.  RE: Cognitive Biases

    Posted 02-19-2021 02:41 PM
    How much does our professional cognitive bias influence our practise? Let's take transportation design as an example: We make traffic and population growth projections, often linearly extrapolating a current short-term trend decades into the future. Time and again, after funding massively expensive projects that severely alter the landscape and character a city, we look back and realise that that growth just never materialised. We widen streets and roads with the expectation their expansion is needed now, but after spending millions of dollars of the municipal budget, realise that the growth never came. Or worse, it does come, but decades after we built the infrastructure such that in its first lifecycle it didn't get used.

    As a profession, I think we're incredibly biased, especially since we're supposed to be the "experts." But engineering, more often than not, is rule-of-thumb and best practise, not scientific law. (I would argue that the problem solving required to find solutions to unique and interesting engineering situations is one of the aspects that draws us to the field, so this fact is not a negative, but helps keep us interested and engaged in the profession.) Too often, instead of approaching a situation humbly in search of a solution, we turn to a manual and treat it as the Bible, even though most manuals explicitly state that they are just guidelines, not gospel.

    ------------------------------
    Joel Dixon P.E., M.ASCE
    Oklahoma City OK
    ------------------------------



  • 10.  RE: Cognitive Biases

    Posted 03-02-2021 09:01 AM
    Edited by Tirza Austin 03-02-2021 09:00 AM
    That's a good point, but I don't think I've ever been on a highway more than a year or two old that was that overdesigned - the traffic finds them.  An underused urban street can be made into a nice linear park.

    As a young engineer, I was working on a master sewer plan for a small city.  They were experiencing very rapid growth, but we were directed to design based on the zoning.  They had a lot of undeveloped land which had been zoned for fairly dense uses.  Based on zoning, they could support a population well over ten times the existing, but even at the rapid growth rate they were experiencing the population would not reach that level for fifty years, and at a more normal growth rate it would take centuries.  When I pointed that out to the planning director, he looked at me, mouth agape, for several seconds and then asked "So, you think we have too much residential zoning?"  My reply was that we should not be planning massive interceptor sewers to support a population that may never exist, but should plan for flexibility to support what was likely to happen during the life of the facilities we were planning.

    ------------------------------
    William Forbes MASCE, PE, ME, BCEE
    Senior Principal Engineer/Vice President of Engineering
    Forensic Analysis & Engineering Corporation
    Virginia Beach, Virginia
    ------------------------------



  • 11.  RE: Cognitive Biases

    Posted 03-02-2021 01:05 PM
    I beg to differ with Mr. Forbes. Rarely is infrastructure over designed. Smart planning looks 10-20 years in the future when it will be almost impossible to upgrade roads or sewers to catch up with growth. In the 80s-90s Austin tried to stop growth by freezing new roadways and not upgrading existing highways for a decade. The thought was that with no WWTP capacity and crowded roads, development would stay away. Development happened anyway with individual subdivision package plants etc. The roads filled with more cars. During rush hours the highways were shut down for hours. Austin has been playing catch up every since, not being able to buy land to widen highways or build new ones. They will never be able to plan 20 years into the future now. Their traffic is some of the worst in the country around the clock.
    It is more cost effective to overdesign now than to go back and tear up homes and roads in 20 years to catch up.

    ------------------------------
    Gordon England P.E., D.WRE(Ret.), M.ASCE
    Brevard County Storywater Dept.
    Stormwater Engineer
    ------------------------------



  • 12.  RE: Cognitive Biases

    Posted 03-06-2021 09:49 AM
    Edited by Tirza Austin 03-06-2021 09:48 AM
    Please don't let me be misunderstood!

    Mr. England, you are not disagreeing with me.  I was responding to a comment by someone who said transportation systems were way overdesigned.  I disagreed that ever happened.  My comment was that you should design for expected growth, and design with the flexibility to accommodate greater growth if it occurs.  But, you can't design an interceptor sewer for a million people when there will be less than 100,000 users during the life of the sewer.  It won't work well, maintenance will be a problem, and the community can't afford it.  You can design for future expansion.  Your example of Austin essentially designing for no growth apparently in the hope that it wouldn't  happen is the antithesis of flexible design.

    ------------------------------
    William Forbes MASCE, PE, ME, BCEE
    Senior Principal Engineer/Vice President of Engineering
    Forensic Analysis & Engineering Corporation
    Virginia Beach, Virginia
    ------------------------------



  • 13.  RE: Cognitive Biases

    Posted 03-07-2021 09:12 AM
    Sorry to misunderstand you


    Gordon England





  • 14.  RE: Cognitive Biases

    Posted 03-08-2021 10:12 AM
    Edited by Tirza Austin 03-08-2021 10:11 AM
    Gordon, Over designed or over designed? The key operative in your comment is the design of the life cycle. Facility design entails capacity and physical durability. Highways designed for larger capacity can indeed be over designed. But this effect can be minimized and even negated by designing increased durability into the finished product. Typical highway designs are incredibly wasteful in terms of life cycle use of resources. A typical 20 year design life usually requires corrective maintenance far earlier than in 20 years. Pavement pot holes, distressed concrete bridge overpass abutments and highway joints, and frozen bridge bearings are all too common. Highway designers imagine that entrained air is the magic ingredient in their concrete, whereas increasing the cement content and reducing the water cement ratio in the design generally leads to a more durable product. Also, avoidance of reactive aggregates such as limestone and dolomite in saline and acid environments by use of basalt greatly increases the life of concrete and is usually cost effective where used.

    What do you plan to do with a product that has reached the end of its design life? Was the US Capitol Building 'over designed'.? Do we rip it down and do it again?

    ------------------------------
    Irving Schlinger P.E., M.ASCE
    Consulting Engi
    Irving Schlinger P.E.
    Chester NY
    ------------------------------



  • 15.  RE: Cognitive Biases

    Posted 03-08-2021 11:22 AM
    Yes, highway guys live and die by the design life. DOTs have much money and can afford to replace roads periodically. Down at the city or county level there is never enough money for maintenance, expecially in highly developed areas. I spend much time and money trying to fix bad pipes and ditches and widen roads between houses or in crowded areas. This is most disruptive for civilians. I would rather spend more money up front to double the design life of my infrastructure and not come back in 20 years. Depends on your point of view

    ------------------------------
    Gordon England P.E., D.WRE(Ret.), M.ASCE
    Stormwater Engineer
    Cocoa Beach FL
    ------------------------------



  • 16.  RE: Cognitive Biases

    Posted 03-08-2021 03:15 PM
    Edited by Tirza Austin 03-08-2021 03:14 PM
    Replacing previously constructed facilities is killing us; and no, Highway Departments do not have plenty of money to replace existing facilities. Their capital programs are usually backed up by years due to scarcity of capital.  Rebuilding the same thing repeatedly, using the same standards and criteria, and expecting a better result is a sign of design failure, planning failure and usually, near sighted stupidity. Would anyone seriously suggest replacing the 140 year old Brooklyn Bridge? Roebling's design, and the subsequent construction by his son should be a model for how to create roads and bridges. The Pensacola Bridge which collapsed due to a barge collision and is now under reconstruction is one of the latest examples of this type of wrong headed thinking. Roebling designed and build a suspension bridge for the Delaware and Hudson Canal at Barryville N.Y. on the Delaware River over 150 years ago. The Delaware River is notorious for ice dams and large floating ice during the deep cold winter conditions prevalent in this area. The bridge has easily withstood site conditions due to the careful design which includes upstream treatment of the supporting piers. A penny wise and pound foolish design in Florida did not foresee that a bridge over a busy navigation channel in a notorious hurricane location might be subjected to a ship or barge out of control. 'An ounce of prevention is worth a pound of cure!' Poor Richard's Almanac, B. Franklin.

    ------------------------------
    Irving Schlinger P.E., M.ASCE
    Consulting Engineer
    Irving Schlinger P.E.
    Chester NY
    ------------------------------



  • 17.  RE: Cognitive Biases

    Posted 02-22-2021 11:53 AM
    Mitchell's phrase, "... many individuals who were always quick to knock the ideas of others" is a good summary of human behavior regardless of profession. I would decribe that behavior as "They don't know what they don't know!" To be fair though, it takes more than a lifetime of learning to "Know what you don't know."

    ------------------------------
    Samuel Ng Ph.D., P.E., F.ASCE
    RETIRED
    Plymouth MN
    ------------------------------



  • 18.  RE: Cognitive Biases

    Posted 02-23-2021 02:47 PM

    There are some fascinating discussions on this important topic (thanks to Mitchell Winkler for initiating it). As we understand it, cognitive bias is a person's subjective reality that results from his or her own pattern of thinking that may not be in sync with the characteristics of the object, or of a problem. It is a systematic pattern of deviation from the rationality of judgment. In the context of societal relationship one can say that, cognitive bias represents the observer's state of mind – his or her mind-set or conviction – which has nothing to do with observed.


    Let us attempt to see the issue cognitive bias from a different perspective – in terms of solving a problem – engineering or otherwise. For such cases, and individual invariably starts with searching for tools in his or her toolbox, which may contain at least three:

    1. The past experiences that lead to the development of one's intuition – often termed as rule-of-thumb, heuristics, commonsense or simply as gut-feeling. They can be very powerful – especially in the preliminary phases of a project – and if the tools of experience fit the requirements. Many routine and well-established engineering projects usually fall into this category.
    2. The standards, guidelines and design codes are developed by experienced professionals utilizing experience, principles and theories. They are intended to provide a general guideline. They come with a disclaimer – because they are not based on site-specific conditions of a certain project. But if certain codes are accepted within a jurisdiction, an engineer is legally bound to check on them as a minimum requirement.
    3. The third is very important for large and complicated projects that require more than routine procedures. Here all-out efforts such as those combining research, advanced computational routines and experiments are needed because depending on the rule-of-thumb approach or design codes may simply prove inadequate and too risky.

    Whatever tools an engineer chooses, his or her commonsense judgment is always required – but not steeped with cognitive bias rather with seeing and comprehending the problem or project as is.

    Often, we become so much involved in something that we forget to see a problem in simple terms. I remember, one of the giants in coastal engineering, RG Dean (1931 – 2015) asked a simple question to a presenting author during a conference session. The author's presentation was highly elaborate in formulae and mathematics – and he was so consumed by them that a simple physics based question appeared very difficult for him to answer.

    -----

    Dilip

    Website

    ORCID ID

    Google Scholar



    ------------------------------
    Dr. Dilip Barua, Ph.D, P.Eng, M. ASCE
    Vancouver, BC, Canada
    https://widecanvas.weebly.com
    ------------------------------



  • 19.  RE: Cognitive Biases

    Posted 03-25-2021 08:54 PM
    I agree, it's a fun topic. I wonder where would 'pretending that what we don't know is not important' or 'is approximately equal to one' would fit in this (as in assuming a value for a coefficient in a formula instead of evaluating what effect a reasonable range of this coefficient would have on the results). Formal statistical procedures and sensitivity evaluations are helpful for this. It's essential to simplify things, but I think we are very subjective in what and how we simplify out of the equation.

    ------------------------------
    Natalya Sokolovskaya P.E., M.ASCE
    Wynnewood PA
    ------------------------------



  • 20.  RE: Cognitive Biases

    Posted 05-03-2021 09:05 AM
    Edited by Tirza Austin 05-11-2021 04:42 PM

    e.g., Some call it cognitive bias, others defense mechanisms.

    And then there are both positive and negative applications of each.

    Appreciation to Mitch and others for surfacing such mission-critical socio-psychological internal mental operations that are within each of us. However, "Shooting from the hip" at them will not deliver the value they offer to our people and their work.

    Observation:

    The value of such knowledge, once we are educated and understand what such insights mean and do not mean, the better will be our collective abilities to engineer our future, today. This again is just one more subject not presented to engineers who are tasked with making decisions as well as evaluating the decisions of others.

    Stay Healthy!

    Cheers,

    Bill



    ------------------------------
    William M. Hayden Jr., Ph.D., P.E., CMQ/OE, F.ASCE
    Buffalo, N.Y.

    "It is never too late to be what you might have been." -- George Eliot 1819 - 1880
    ------------------------------