Naturalistic Decision Making
In our web enabled, data driven society, intuition is making a comeback. Paragons of the corporate world such as Jack Welch (who titled his memoir Straight from the Gut) are hailed not just for their success, but their readiness to bet on their instincts. Conversely, politicians are reviled if they are perceived to govern by opinion polls.
The idea that intuition prevails over data is comforting; a triumph of man over machine. However, the reality is more complicated. We certainly don’t admire everybody’s instinct, just those whose decisions pay off. One who makes the leap and fails is not a hero, but a fool.
Do we really want to rely on a gut decision from someone who just had a bad lunch?
While romanticizing about intuition is a formula for disaster, the fact is that rational decision making isn’t pragmatic or effective most of the time. Fortunately, decision making has been an active area of study and, despite some false starts, true insights have been won.
The implications of a vast body of research are that not only is intuition important, but also that “gut” decisions can be improved with training. Moreover, by understanding how consumers make decisions we can market to them more effectively.
A Failed Rational Model
One of the early models for effective decision making comes from a much cited paper by Peer Soelberg of MIT. He described a “generalizable decision process” for finding “ideal solutions” in a 1967 paper which provided an outline of 6 steps.
- Problem Recognition
- Problem Definition
- Planning
- Search
- Confirmation
- Implementation
The paper was logical, exhaustively detailed and well received. However, Soelberg’s model and others like it never gained widespread use outside of academia. In reality people rarely use such models because an “ideal solution” is very rarely what you need.
Most of the time, you just need to get something done; a problem solved. You have other things to do and more decisions to make. You find a solution that works and move on.
As an example, think about a pilot who just lost an engine. He has to make a decision about what to do – lives are at stake. He has no need for an optimal solution; one that will keep him from crashing would be more than enough.
Now imagine you are that pilot and review the 6 steps outlined above. How many would you skip?
A 6th Sense?
With funding from the US Military, Gary Klein set out to find out how people actually make decisions under pressure and whether the process can be improved upon. He was initially shocked at what he found: People who make life or death choices often do so with seemingly no process at all.
One fairly typical story was that of a fireman who responded to a routine kitchen fire. While they were spraying water on the fire, the lieutenant gets a bad feeling and orders everybody out. Seconds later the floor collapsed. If not for his order they would have been killed, or at least very seriously injured.
The fireman attributed his life saving “gut” decision to a natural “6th sense” he had about fires. However, during subsequent interviews, some cues became apparent. The fire was too hot for a kitchen fire, it wasn’t responding to the water hose and it wasn’t noisy – all of these things were out of place.
Something felt wrong because the situation violated the fireman’s conception of what was normal based on his experience with previous kitchen fires. His”6th Sense” wasn’t innate, it was learned.
Where a “Gut Feeling” Comes From
The Fireman’s example above is supported by scientific evidence. Renowned neurologist Antonio Damasio believes that we store important experiences as somatic markers in the brain which are regulated by the amygdala, an area of the brain involved in both emotions and memory.
In other words, when something violates our conception of what’s normal we really do have a “gut feeling.” It is through experience and learning that we build our conceptions. In effect, many of our lightning fast impulses are developed over time. That’s why top athletes need to train hard for years.
Damasio also points out that our instincts work faster than our conscious thought – like when we jump out of the way of an oncoming bus and only seconds later realize what happened. When it really matters, we don’t have time to think, only to act.
This explanation also explains the instincts of corporate titans such as Jack Welch. Inside GE, Welch was famous for his ability to digest complicated financial statements, a skill conspicuously absent in his early career. He also pumped billions into GE’s Crotonville Management Center and spearheaded GE’s Six Sigma program, an extremely quantitative process.
For someone who values going “straight from the gut” Welch certainly has worked hard at it, and that’s the point. Intuition isn’t something we’re born with; it’s something that we acquire. Moreover, it’s only effective when we have to make decisions within the realm of our experience.
(For a real life business example, see here).
How to Improve Decision Making
Our instincts are wired for survival, not planning. They help us recognize danger and reward, but don’t actually prescribe solutions. Our emotions help us weigh factors in a decision, but they don’t explain them. They alert us to past lessons learned, but don’t tell us anything about the future.
In order to make decisions effectively we need to both access past experiences and imagine how the future can differ. The evidence points to concrete steps that we can take to improve decision making.
Make Decisions: Like anything else, decision makers get better with practice. The more decisions you make, the better more proficient you’ll be.
Make Mistakes: While failure is difficult, it’s through experience that we build up the somatic markers that enable us to recognize situations quickly and react to them.
Simulate: An important part of Gary Klein’s findings was that while people who have to make life or death decisions routinely didn’t weigh options against each other, they did mentally simulate the consequences of each initiative. If the risk was too great they would move on to another option till they found one likely to result in a successful outcome.
Review Past Decisions: The US Military puts a big emphasis on After Action Reviews, where every facet of an operation is analyzed and rethought.
Our decisions are in large part prisoners to our own experience. By enlarging and clarifying our own personal database of experiences we will have more to draw on when the time comes to act.
Marketing Implications
The research also reveals a lot about how consumers make decisions and can therefore help us to market more effectively.
Most Purchase Decisions are Highly Intuitive: Consumers have to make too many decisions to utilize Soelberg’s rational process on a regular basis. They rarely seek out an ideal solution, just one that works. Researchers call this satisficing.
Product Cycle Helps Determine Emotional Content: While an emotional component is present in any decision, consumers will gather more information about important decisions they make infrequently.
The more important the purchase decision, the more optimized the process will be. While there will still be an emotional component, rational factors will be more important for purchases of great expense or impact.
Familiarity is Essential for Brand Preference: Somatic Markers are built through continuous experience. Brands that stay in touch with consumers have a distinct advantage. The research also suggests that new brands need to build associations that are already familiar to prospective clients.
Better is not Always Best
For most choices we have to make, an adequate solution will do. One that is simple, available and cost effective will be preferred over one that requires new learning.
We are hard wired to accomplish tasks, not to optimize them.
– Greg
Greg;
Another thought-provoking post. To me, thinking and decision involve using mental models to classify the issue into meaningful clumps of information which can then be manipulated as a chunk. These models are simply a representation of something else, but constructed in such a way as to allow a form of simulation to take place, which will reveal behaviour (or characteristics, or performance, or whatever it is you wish to explore about the object being modeled). In case all this sounds too abstract, think about a Doll’s House, or a Doll, or a toy car or toy boat. Each of these exists to allow a simulation to take place – in this case the act of playing with the doll in the doll’s house allows the child to simulate her life at home. Ditto for the toy-car or boat simulating the act of driving or sailing.
In your fireman’s example, he subconsciously noticed a number of things about the fire (lack of noise, excessive heat, etc.) and because they did not jive with his mental model of the way the fire should be behaving, he called for everyone to clear the room.
As a marketer then, we try to build mental models of our Ideal Prospects, (Personas), so that we can then use them to simulate a part of the marketing or sales process. We use the mental model to try on a pitch, or an offer, and using the mental image of the Prospect tends to make us more clearly able to simulate the way the Prospect will react to the message.
Eric,
Thanks for your input. What your saying has a great deal of historical precedence. Plato discussed at length “ideal forms” and Kant talked about how we use a priori beliefs to color our perceptions.
What I think differs in the work of Gary Klein and others is the emphasis on how those mental models are built, which allows us to actually improve decision making. Apparently, the US military is a big believer in this (at least to the extent that they funded most of Klein’s research).
– Greg
Greg;
Thanks for the additional info. I did not know that the US DoD funded much of Klein’s research. Makes sense – no doubt that if you know how to build your model in a conscious way, you will do a better job than if you just use intuition. Not that intuition doesn’t play a part, as you so succinctly pointed out, but adding reason and logic to it just makes it that much better.
Bye for now.
Hi Greg
Interesting post. Just one question – isn’t it possible that in emergencies – the brain goes through the entire 6 steps of the Rational Model that you have talked about – in a jiffy? The brain has a vast database of experiences and situations – and in the case of the fireman too – the Search and the Confirmation stage told him that the conditions in that particular incident were not normal – and that is why he probably ordered his men to leave at once.
Ajoy,
It’s possible, but the research shows that they don’t. What seems to happen is that people run through possibilities in an iterative, trial and error process. When they find a solution that seems like it will work, they go with it and don’t consider other options until they run into further difficulty.
– Greg
Greg,
Great argument, well written, as usual. Have you read “How We Decide” by Lehrer? What you’re saying here sounds very familiar after reading his book. I am trying to get a number of my clients to read it, because it is possible, then, to have a highly intelligent conversation with them about their marketing and why we sometimes suggest “awareness” campaigns when they are really looking for sales. I think that the fact that the whole familiarization process takes time and then results in action means that there will always be some lag between marketing and sales–something that will always be hard to measure. This makes it very difficult to postulate (or measure) a one-to-one correspondance between marketing and sales–at least for anything more complex than a simple purchase.
Adam,
No, actually I didn’t read the book, but other people have told me of the similarities to my post (which makes me think I should have chosen a different title to avoid confusion).
– Greg
Greg, You are batting 1000! I was just pondering this topic this morning, reading “Digital Body Language” by Steven Woods who is one of the founders of Eloqua. A good read on the topic is an older book “How Customers Think: Essential Insights into the Mind of the Market” by Gerald Zaltman.
Woods makes an elaborate analogy between the evidence of buying stage, role, etc. that prospects leave as they wander across the Web and the perceptions we pick up in face-to-face interaction. He is of course promoting his company’s black box but he makes many provocative points. The underlying logic is very simple and basic which he reveals in his case studies, not in the elaborate rationale. These simple rules of thumb and heuristic processes he outlines in case studies are like intuitively processed body language, which is to say, close enough and quick.
How we decide is something we can all learn more about simply by becoming more aware of our own behavior. All the evidence I have seen indicates that most decisions are made on a limited set of key factors. Your fireman’s story is a great case in point: hot, unresponsive to water and too quiet –GET OUT.
I have been repeatedly shocked by young, supposedly educated adults including marketing professionals who tell me they are not influenced by advertising, only want facts, and are self-guided in their search for information.
Roger,
Thanks for referencing the “Digital Body Language.” I’ll try to check it out. The neurology of marketing is still relatively new and moving fast, so I’m sure new insights will be gained and the field will continue to advance.
Regarding your last comment, I would say that it’s especially marketing professionals who think advertising affects them. The seem to think that it affects everybody except them, because they are so sophisticated. It’s probably why most marketing people are so bad at their jobs.
– Greg
Hello Greg,
Good post and valuable comments too.
I want to make a focus on decision making into companies : many are using BI tools (mining structured operational data) as basis to make decision, others are using CI tools (mining unstructured external data) and some are using both solutions separately.
So far, i didn’t heard about an application that combines BI & CI features with high capability to mash-up data. In others words, is it to possible to build an application that help managers make decisions based on “intelligent” mash-up of internal and external data ?
An interesting post to read about future trends of BI solutions : http://www.enterpriseirregulars.com/5706/the-top-10-trends-for-2010-in-analytics-business-intelligence-and-performance-management/
Amine,
Sorry to disappoint, but must confess that I’m completely ignorant of both technologies.
– Greg
Greg, you really should read Joshua Lehrer’s book “How We Decide” that Adam cited. It’s an easy read and summarizes current knowledge of decision making. It validates the value of intuition and shows how to use it best.
Joe,
I do plan to read it. I’ve taken a quick look and he uses slightly different sources than I do, so I’m sure it will be interesting.
Although, I should say that I’m always a little bit cautious with journalistic accounts. They tend to try to make everything fit the story line and you lose sense of the process of discovery and present conclusions rather than facts.
– Greg
I agree with your comment about journalistic accounts of technical things. However, Joshua Lehrer is also a trained neuroscientist having worked in the lab of one of the best known neuroscientists so he knows what he is talking about. Plus he does have the gift of being able to write so that anyone can understand.
Joe,
I stand corrected. His book is on my list, so I’ll get to see for myself:-))
– Greg
Greg,
Thanks for an interesting post. Lehrer’s books is on point, but there are whole fields of knowledge about “instinctive” decision making Jonah doesn’t get to. I try to address them in a book I have coming out in March, “How Risky Is It, Really? Why Our Fears Don’t Always Match the Facts”. While it focuses on the way we make snap decisions about risk, the book touches on the kinds of decisions about ANYTHING to which your post refers. A few details follow;
First, there is the wiring of the brain. Information for external sources, or even internal thoughts and memories, goes first to an area called the thalamus, which partially processes it and sends it quickly on to other areas. One of these, right next to the thalamus, is the amygdala. This is the part of the brain where fear begins, and this wiring makes adaptive sense. We want to know as soon as possible if something could be dangerous. The amygdala gets information before the rational thinking parts of the brain even receive it! We are hard wired to fear first and think second. That makes the kind of rational decision making your post discusses kind of difficult. Even after the cognitive cortex enters the process, emotions and values still carry more weight than reason. As Joseph LeDoux, a pioneer in the neuroscience of fear, wrote in The Emotional Brain, “…the wiring of the brain at this point in our evolutionary history is such that connections from the emotional systems to the cognitive systems are stronger than connection from the cognitive systems to the emotional systems.”
Then there are the heuristics and biases, the mental shortcuts to decision making we use when we only have partial information and have to make up our minds, which is how real life usually is. this is the work of Daniel Kahneman, Amos Tversky, and many others, no sometimes referred to as behavioral economics.
Re: risk, there are a set of psychological characteristics that make something feel more or less scary, no matter what the facts say. This is why some people are still afraid the vaccines causer autism, or why many people feel safer when they are driving than whey they fly.
Finally, there is our underlying cultural views on how society should be organized. As social/tribal animals, we make decisions in ways that reinforce the beliefs of our tribe (party/religion/family/local community/nation) so our tribe’s influence rises, which helps our chances for survival.
This is all well summed up be D’Amasio in Descartes Error. He writes about Elliott, who had a brain tumor which severed the connections bweteen the cognitive prefrontal cortex and the lower limbic/emotional areas of the brain. Elliot passed all the cognitive tests with flying colors but was dysfunctional. He couldn’t make a single choice, because choosing meant putting a value on one option over another, and the values part of his brain couldn’t talk to the part that had the facts. Facts were meaningless, in the purest sense of the word.
Sorry for the longish post, but thought this might add to the conversation. I also write about this, re: risk issues, at onrisk.blogspot.com
David,
Thank you. Descrtes Error is indeed a fantastic book. It was one of the books featured on on my Digital Tonto reading list🙂
Evolutionary psychologists like Richard Dawkins and EO Wilson also have much of value to say.
Thanks again for your comment.
– Greg