In When Bias Trumps Design Part 1 we used a photo of M&Ms® to show how we tend to overlook anything that is not right in front of our eyes. In the photo, the missing items were brown M&Ms®.
While overlooking missing M&Ms® is hardly important, the implications for interface design are serious. Just as we overlook something that is not in our direct line of sight, we often fail to consider information that we cannot easily recall sometimes resulting in faulty assumptions. For example, rather than conduct user research, product owners, subject matter experts (SMEs), and even some designers rely on what they think they know about their target audience.
What drives these assumptions is the availability bias, the tendency to make decisions based on information we can easily recall, even if this information is irrelevant. This bias is unconscious meaning that the assumptions emerge from the brain’s fast and unconscious processing.
As if these assumptions were not bad enough, the availability bias leads to a more insidious dynamic. It causes us to take a complex question and unconsciously replace it with an easier question. Economist Connel Fullenkamp illustrates the point with the following anecdote.
Imagine that you have just finished a delicious sandwich at a new restaurant chain. Based on this delightful experience you become convinced that the company is surely on the rise, and you rush home to buy the company’s stock.
The problem? You’ve substituted an easy-to-answer-question (was the sandwich delicious?) for a much more difficult question (do the company’s financials and market share make it a good candidate for revenue growth?). As Fullenkamp explains, “The really amazing thing about system 1 is that it seems to be really good at translating the experiences on one dimension and expressing this intensity on another dimension.”* System 1 refers to the brain’s fast, automatic, and largely unconscious processing, a concept that behavioral economist Daniel Kahneman explains in depth in Thinking Fast and Slow.**
Basically, the availability bias unconsciously tricks us into thinking we have understood a complex situation. The most common examples of this bias emerge when product managers and SMEs substitute their knowledge and opinions for end user needs as happened on a web application design project for a telecommunications company.
The supervisors and product owners assured us that the critical need was a single screen crammed with as much information as possible: “I was thinking more in terms of clicks,” said the product owner referring to the misguided goal of reducing the number of clicks. Specifically, she objected to our proposed design with all critical information on the landing page and clear cues to details. Our goal was to meet the users’ needs by presenting access to critical information up front with links to details. In short, we aimed for a dense but not cluttered design.
The point of this anecdote is not to engage in endless debate about clicks. Rather, it illustrates the risk of substituting an easy question, how many clicks?, for the more difficult and more important question, how to best meet the users’ needs.
The key word here is “substitute.” Replacing a complex question about how end users can find and act on important information with an easier question about clicks is equivalent to substituting our opinion about a tasty sandwich for the more nuanced questions about a company’s prospects and financial footing. The result is often poor design based on faulty assumptions.
In our next post, we will offer suggestions for circumventing brain quirks like the availability bias.
*The Economics of Uncertainty (The Great Courses) by Professor Connel Fullenkamp.
**Kahneman and Tversky are properly credited with developing the concept of a brain with system 1 (fast, automatic) and system 2 (more deliberate) processing. Other scholars, however, have contributed to the literature on multiple processing systems and, in some cases, have expressed a different view. For example, in Rationality and the Reflective Mind, Keith Stanovich argues that the human brain engages in three types of processing.