Human Hardware: Searching With The Basal Ganglia
In the last column, we explored how navigation can become a habit. Ann Graybiel’s work shows how regular paths become etched in our basal ganglia, limiting the need for conscious interaction. When we do something enough and the situation seems familiar, we can cruise through on autopilot, saving our prefrontal cortex for more urgent demands. […]
In the last column, we explored how navigation can become a habit. Ann Graybiel’s work shows how regular paths become etched in our basal ganglia, limiting the need for conscious interaction. When we do something enough and the situation seems familiar, we can cruise through on autopilot, saving our prefrontal cortex for more urgent demands. This is how you can drive home from work and really not remember how you got from point A to B. You didn’t need to think, so there were no conscious memory traces laid. The sub-cortical brain was doing all the work
Is searching a habit too?
In this column, I wanted to look at how this can apply to our use of search engines. One of the most frustrating things experienced by Yahoo, Microsoft and Ask, along with everyone else who is challenging Google’s search crown, is that the more they appeal to users to give them a try, through advertising, incentives or other approaches, the bigger Google’s search share seems to grow. It’s almost as if the ads and incentives were working in reverse.
Where there’s smoke there are habits
A possible answer can be found if we look at another area where habits play a huge role: healthcare. The anti-smoking lobby found a similar dilemma when they ran TV campaigns urging youth to not smoke. The ads were certainly compelling, providing hard hitting, rational reasons why it’s literally suicide to light up a smoke. But the ads didn’t seem to be working very well. In fact, in one study, they found exposure to the ads actually triggered a desire to smoke. What was happening?
The problem, as several studies began to uncover, was that the ads were targeting the wrong part of the brain. The rational appeals were aimed at our logical, rational brains, and just like Graybiel’s mice in a maze, the desire to smoke had long left the rational brain and had been firmly entrenched in the subconscious. Smoking is a habit, and by the very definition of a habit, we don’t think about it. We just do it.
To make the smoking dilemma even more perplexing, Graybiel has discovered that habits, or rather, strings of behaviors, often play out in “chunks.” A stimulus will trigger a habit that is comprised of a string of sequential behaviors. One example study found this might be the reason behind a cultural conundrum: why are the French less obese than Americans?
Why aren’t the French fatter?
I’m writing this column from southern France, so I can personally attest to the never ending parade of culinary temptations that present themselves here. French cooking is incredibly rich, as my gall bladder keeps reminding me. This is a country dedicated to gourmands everywhere. Just a walk past a cheese counter (Reblochon and Beaufort are my two personal favorites) can cause an hour of indecision. Charles de Gaulle once remarked that it’s impossible to govern any country that has 246 different kinds of cheese. Yet, despite all the cheese, cream, chocolate, and carbs, I’ve seen very few obese French people. The national rate of obesity in France is 7%, compared to the world leading American rate of 1 in 4 people. Why?
Researchers at the Cornell University found one possible reason for the so-called French Paradox. The French eat until they are full, whereas Americans depend on external signals from their environment to stop eating. In fact, our eating schedule is often tied to our TV watching schedule. We eat ice cream when we watch Desperate Housewives. Because we spend our lives in front of the TV, our watching and eating habits have become entwined in a big habitual “chunk” that plays itself out, night after night. The theme music of our favorite show triggers our desire to eat. In just one amazing example, the average menu that accompanies a Super Bowl party can easily top 7,000 calories per person, almost 4 times the daily recommended level!
It’s not just eating that we do by habit. Researcher Wendy Wood at Duke University found that up to 45% of our daily routines are done by habit, actions triggered by environmental cues in the places we tend to frequent. We do the same thing day after day, in the same place, without conscious thought. We go through our lives on autopilot thanks to our habits.
The habit of searching
So, is searching a habit? Does a browser window trigger the playing of a habitual “chunk” of behavior? From everything we’ve seen in our user interactions with search at Enquiro, we believe the answer is yes. Remember, the road to build a habit requires a constant environment that always presents the same stimuli and a repeated behavior that produces the same results. Those factors create a stable environment, and stable environments are the petri dish in which habits grow and thrive. When we can keep doing the same things in the same environment and get the same results, we don’t have to think about it anymore. The prefrontal cortex can relax and habits can take over.
More than just searching, I believe our overwhelming use of Google is a habit. We find ourselves using Google without thinking about it. And this is the dilemma faced by Google’s competition. Advertising and incentives (such as Microsoft’s Cashback) are aimed at the rational mind. But our choice of Google has become habitual. To get us to change the habit, you have to disrupt the environment that’s allowing the habit “chunk” to keep playing itself out. Unless Google suddenly starts delivering horrible results, forcing us to kick in the prefrontal cortex to suddenly evaluate our alternatives, the only other option is to interrupt the habitual behavior upstream before it has a chance to execute.
How Google became a habit
If you accept my hypothesis that our Google use has become habitual, it then becomes interesting to speculate on how Google came to be a habit. Remember, the important thing with habits is frequent repetition in a stable environment, and this offers clues to the formation of the Google habit.
Prior to Google’s emergence on the search scene, we didn’t have one engine that we could rely on for consistent results. I used 3 or 4 engines regularly, switching between Infoseek, AltaVista, Excite and Yahoo!. If I didn’t get the results I was looking for, I quickly changed to another engine and tried my search there. I even found that for different types of searches, some engines produced better results than others, but I use the term “better” in a relative context. The results were inconsistent, to say the least. Search engine use for me, and for everyone I knew, meant juggling between a number of different engines.
Then I discovered Google, still a beta project at the time. Google’s admittedly revolutionary algorithm produced better results than the competitors for every single search. Suddenly, one engine replaced 4 others. My searching “environment” stabilized.
Google also was the beneficiary of good timing. Just at the time that Google was stabilizing my search environment through the PageRank algorithm, there was also an explosion of web content available to be indexed. The number of websites was growing exponentially during Google’s beta period. One of the consistent findings of all Enquiro’s research has been that actual relevancy matters less than perceived relevancy, and for Google, the explosion of indexable content meant its users had a corresponding lift in perceived relevancy. Not only was Google’s algorithm better, but it was being applied to a much bigger body of content. As far as we were concerned, it seemed that no matter what we were searching for, Google did a better job of bringing back relevant websites.
Today, the question of which engine is the most relevant is probably an academic one. Our perception is that Google is the most relevant, which reinforces our habitual use of Google. And here we have another quirk of our human hardware. Once habits are in place, and we have a vested interest in making sure they stay in place. Habits save us from expending unnecessary cognitive effort, so evolution has built-in safeguards to preserve them. Once we’ve formed a habit, we tend to ignore information that threatens the habit and seek out information that reinforces it. This is another reason why advertising appeals largely fall on deaf ears.
Scanning search results on auto pilot
Habits play a role not just in which search engine we use, but also how we use them. As long as the interface remains the same, we don’t have to think our way through it. One of the more interesting findings in our past research came as a surprise to us. In one eyetracking study, we divided up our sample into two groups and gave each of them a different search task, but we showed both groups the same set of results. The first group we gave an information gathering task, and the second group we gave a commercial task that would have involved making an online transaction (in this case, booking a hotel room). Both groups were shown search results that included the same top sponsored results and the same algorithmic results.
We expected to see significantly different scan patterns in the same groups. We were testing a theory we had called “preslicing,” where we believed that the users would tend to skip over the sponsored results when their intent was less commercial in nature. We expected to see a significant shift in scanning, moving quickly down the page to the algorithmic results. But no matter how we sliced the data, we didn’t see what we thought we’d see. Gaze time (how long users looked at the listings) interaction with the sponsored results varied little between the two groups. In fact, it was a little higher in the information gathering group. No matter the user intent, it seemed that users scanned the results in the same way.
The difference came when we looked at where they clicked. The commercial intent group split almost evenly, with half clicking in the algorithmic results and half clicking in the sponsored results. But in the information gathering group, everyone clicked in the algorithmic results. They scanned the page in the same way, but they certainly didn’t respond to it in the same way.
As I’ve thought about this, I’ve adjusted my “preslicing” hypothesis to accommodate habitual behavior. I believe we don’t really need to think about which listings to look at first. Google has done a pretty good job of always presenting relevant listings at the top, whether they be sponsored or algorithmic results. In terms analogous to Graybiel’s work with the mice and the maze, Google doesn’t give us a reason to start thinking our way through the maze here. They allow us to scan the page by habit. As long as Google keeps true to its goal of showing the most relevant results at the top, they keep the environment stable and we don’t have to think.
But we don’t click by habit. Google isn’t good enough at determining intent to ensure that the top result will always be the one we click on. There are too many factors, many of them locked away in our brains, at play here to provide this degree of stability. So while Google keeps us scanning the page by habit, most of us click deliberately. We at least scan the top 4 results and then make our selection based on perceived relevancy from them. The habit “chunk” stops playing out when it comes time to click and the cognitive mind kicks in.
If my “habitual” hypothesis of search is correct, it bodes well for Google’s future. We have made Google a habit and that means that it will be very difficult for competitors to disrupt that habit. And if we scan the page by habit as well, it means that Google will keep interaction with the top sponsored results at a high level. Both factors are good news for Google’s bottom line.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.