Those who cannot tell what they desire or expect, still sigh and struggle with indefinite thoughts and vast wishes.
—Ralph Waldo Emerson
The Psychology of User Actions
Conscious and Subconscious Behavior
Transforming Difficult Tasks into Simple Ones
Now that you’ve learned about good user design and what it takes to build both a good user interface and good user documentation, you need to understand how users behave so you can build a software product, hardware product, or Web product to meet your users’ needs. When designers approach the design of a product or documentation unaware of their users’ mindset, a product can become unusable very quickly. Only users who have more experience with the way something works can figure it out.
There are plenty of stories about the trials of technology. In the olden days before TiVo and digital video recorders, the “gold standard” of poor usability was the VCR machine, where children often had a better grasp of how to use one than their parents. If you have a great deal of experience with technology, you also know that you’re the person who’s the live-in technology repair department, especially within the family. Recently I visited my grandmother, and she asked me to fix her radio. She unintentionally pushed a button and lost her preprogrammed station (programmed by someone else in the family) and the time on her clock. I got it working in a couple of minutes because I’m used to playing with electronic gadgets that have a lot of buttons.
To understand your users’ needs, you’re going to have to delve a bit into psychology. Users bring their experiences to a new task, and they bring those experiences packed into a conceptual model of how they think the world works. Users also bring their various personality types into every situation.
People have to manage their knowledge in their brain versus knowledge that’s already in the world. Most of us deal with imprecise knowledge in our brains, but we’re often reminded about knowledge that’s also in the world. Sometimes we look for that world knowledge in other places, such as on the Internet. Users’ perceptions of the world are reinforced every day as they interact with it, and context affects perceptions, attitudes, and solutions. There are trade-offs using knowledge in the brain versus knowledge in the world.
According to Norman (2002), when people are presented with a new task, they adhere to one of three different types of task structures. They also adhere to previous information they’ve encountered and make their choices based on this information on both a conscious and subconscious level.
You’ll see how to develop a user interface design by using several steps to transform difficult tasks into simple ones and create a conceptual model of what you’re creating.
Everyone who has ever used anything or has tried to perform a task and failed has felt helpless. Indeed, many people find reasons they can’t perform a task using a product or object (Norman, 2002). These reasons include the following:
You may be familiar with the Myers-Briggs Type Indicator (MBTI) method of identifying and understanding personality type preferences. It was developed during World War II by Katherine Briggs and her daughter, Isabelle Myers (Eisenberg and Eisenberg, 2006). The MBTI has been adopted by many companies over the years as part of human relations training to show employees what types of personalities are in the workplace and how to best get along with people of each type. I remember going through the same type of mandatory MBTI training years ago.
The MBTI was based on psychological types pioneered by Carl Jung, and Hippocrates before him (Eisenberg and Eisenberg, 2006). These psychological types were split into four areas that Jung called feeler, thinker, sensor, and intuitor.
Briggs and Myers expanded each of these four areas into two different dichotomies, because people use their left and right brains to make decisions. The use of these personality dichotomies reflects the dichotomy created by the left brain, which is more logical and rational, and the right brain, which is more creative and emotional. The current version of the test asks a person 93 forced-choice questions, which means there are only two possible answers. The tester then scores the test and places the information into one of four dichotomies (Wikipedia, http://en.wikipedia.org/wiki/Myers-Briggs_Type_Indicator):
Myers and Briggs combined these four dichotomies into 16 different personality types because not everyone adheres to every type of personality characteristic associated with the left brain and the right brain. For example, a person can be an introvert but rely on her feelings when dealing with people and tasks. Indeed, with eight different personality characteristics, there are people on the extremes of the personality scale—those people who have completely left-brained characteristics and those who have completely right-brained characteristics—but most people are somewhere between both extremes.
These personality types are not always predictive of how people will behave in different situations, because context also drives behavior. For example, people who are participating in a usability test may react differently because they know they’re participating in a test. Those people may be more interested in providing feedback about the product’s usability than they would be in the real world, where they may just ask for a refund if they find the product’s usability lacking.
Bryan and Jeffrey Eisenberg (2006) also discuss the development of temperament and character types by David Keirsey and Marilyn Bates, who were more interested in long-term behavior patterns. These studies were later merged with the Myers-Briggs personality tests to show how people put their personality types to work when they make decisions. In Chapter 6, “Analyzing Your Users,” you will learn how to create groups of user models called personas that will include one or more of these personality types. From these models, you will learn who your primary users are and determine what interface suits those users best. For example, if your primary users are sensing people and also the judging type, you may want to create an interface that is consistent with interfaces that don’t introduce anything unfamiliar and provide visual cues for completing tasks.
Keirsey and Bates identified the sensing and intuition types as the first type of preference that people apply when they approach a task or situation, because people want to know how to process information from the brain and the world. (You’ll learn more about knowledge in the brain versus knowledge in the world later in this chapter.) Keirsey and Bates then applied this preference criterion against the four left-brain personality types to create four primary temperaments (Eisenberg and Eisenberg, 2006):
Eisenberg and Eisenberg (2006) give each of these four primary temperaments intuitive labels that provide a good idea of the behavior that each type of user temperament exhibits when the user is faced with a new task or situation. These four types are as follows:
If you’ve been through personality training, you may recognize these personality types as being associated with different types of animals. When I went through personality training as a contractor for a large high-tech company, these four personality types were associated with different types of birds. A methodical personality was associated with an owl, a spontaneous personality with a peacock, a humanistic personality with a dove, and a competitive personality with an eagle.
Methodical people are logical, and they approach a new situation slowly. They exhibit the following characteristics (Eisenberg and Eisenberg, 2006):
Methodical people want you to answer the how questions:
Spontaneous types live in the moment, which makes them impulsive when they take action. They wish to make an immediate impact—they skip many of the details and make a gut decision quickly. Spontaneous types exhibit the following characteristics (Eisenberg and Eisenberg, 2006):
Spontaneous types want you to answer the why and when questions:
Humanistic types put others before themselves, and they’re uncomfortable with allowing anyone else to do work (or anything else) for them. Humanistic types highly value relationships and enjoy helping others. They fear separation, are creative and entertaining, and are good listeners, so it’s not surprising that they have a wide circle of friends and acquaintances. Humanistic types prefer to look at the big picture. They greatly value human development in themselves and others.
Humanistic types exhibit the following characteristics (Eisenberg and Eisenberg, 2006):
Humanistic types want you to answer the who questions:
Competitive types seek competence in themselves and others and desire not only to understand life, but to control it. They enjoy overcoming challenges and learning new things and look for methods to achieve their goals. Competitive types are highly motivated and persuasive. They usually come to decisions quickly after they feel they have all the information at hand.
Competitive types exhibit the following characteristics (Eisenberg and Eisenberg, 2006):
Competitive types want you to answer the what questions:
Many everyday tasks aren’t planned, but they’re opportunistic—on most days, people simply decide to use something when they think about it. No matter if you’ve used a product before or not, you may experience difficulties with that product because of simple misunderstandings and misinterpretations.
These misunderstandings or misinterpretations can occur anywhere along what Norman (2002) described as the seven stages of human action when performing a task:
People can use one, some, or all of these stages when performing a task, and there is a continuous feedback loop. Misunderstandings and misinterpretations can come from people who are not using all of the stages for some reason, such as confidence that they’ve used a similar product in the past and belief that their actions will work for a new product just as well. Other misunderstandings and interpretations can come from poor design that fails to account for how the user perceives the state of the world.
People rely on a dichotomy of two different preferences: sensing and intuition (Eisenberg and Eisenberg, 2006). Sensing is based on knowledge that people bring with them from information they gather through the world, whereas intuition is based more on information they’ve stored in their brains, such as patterns they’ve noticed from similar tasks.
Most people rely on imprecise knowledge to get around in the world. For example, people know what a 10 dollar bill looks like because they see the number 10 in the corners of the bill. However, if you ask a group of people who the person is on the front of the bill or what the structure is on the back of the bill, you won’t get many correct answers. People don’t need precise knowledge in many cases. For example, if I see a $10 bill in my wallet to pay for lunch, as shown in Figure 5.1, I don’t care who the person is on the front of the bill because that information isn’t necessary to complete the transaction. However, I need to know that the number 10 is in the corners of the bill so that I use the right currency to pay for the transaction and get the food I need to survive.
Figure 5.1. Can you name the person on the 10 dollar bill?
Our brains can only accept so much information in our memories. Because of that, we require constraints that break information up so our brains can process it. For example, there are seven digits in phone numbers because our human brains can process only seven digits at a time. As we perform more complex tasks, we break up each task into smaller ones so we can keep track of everything, such as when we’re assembling a new computer desk that has a large number of parts and tasks.
People are also reminded about knowledge from the world—that is, from external sources. If you’re like me, you use a calendar program like Microsoft Outlook to remind you to perform tasks at a certain time. (And, like me, you may have your calendar synchronized with your Palm Treo or other handheld PC calendar.) There may also be instructions for a product that tell you how to do something, such as the controls on your stove showing you which knob controls the appropriate burner.
How much knowledge you use in your brain versus the world depends on the context of the situation. For example, if you’re presented with a new Windows program, you know how Windows works, so you’re likely to use more of what’s stored in your brain about the use of Windows products to guide you in the use of the program. There are trade-offs to using knowledge in your brain versus knowledge available in the world. For example, knowledge available in the world is easily retrievable as long as it’s visible or audible and within your visual and auditory range. However, if you rely on your memory, you may have a difficult time remembering or become distracted by something else. Yet, after you do remember something, you’re efficient because you know how to do the task. If you’re relying on instruction from the world, such as from an assembly diagram that comes with a product, you’re learning; therefore, you’re inefficient.
When you create personas in Chapter 6, you’ll learn more about how your users see knowledge in their brains and in the world. Then you’ll use that information to design an interface that helps your users complete tasks quickly and efficiently.
So how do people manage others’ knowledge in their brain versus knowledge in the world to communicate as effectively as possible? They structure tasks, which is something you learned in school. There are three types of task structures that people face when performing tasks (Norman, 2002):
In general, people find wide and deep structures the most challenging and the most time consuming when completing a task because there are so many options available. The reason for that lies in how people behave consciously and subconsciously.
We handle a lot of our everyday behavior subconsciously (Norman, 2002). After all, we’ve brushed our teeth, showered, dressed, and driven to work and back many times. Sometimes we can’t even remember driving through an intersection to get to work, even though we realize that we drove through that intersection because we’re obviously at work! Subconscious work depends on patterns and regularity, and our subconscious mind completes tasks every day based on these patterns.
Sometimes our subconscious mind takes over unexpectedly. When I was in grade school, I was so comfortable taking a spelling test that my conscious mind decided to take a vacation for a couple of minutes, and I didn’t remember those two minutes. However, when I looked down at my paper, I found I had written two more words, and both of them were spelled correctly.
Subconscious behavior can be fast, relies on longer-term memories, and can make it seem as though solutions pop into your head instantaneously. Now that I’m an adult, I find that my subconscious comes up with answers I thought about days earlier but had forgotten about, but then an answer suddenly popped into my head and I wrote it down. Subconscious behavior can also be the source of hunches about behavior or how something is supposed to work, as opposed to the conscious method of acquiring information.
By contrast, conscious behavior is slow, serial, and relies on short-term memories, which are limited and can be unreliable. When you perform tasks using your conscious mind, you have to be deliberate and go through the seven stages of human action to perform a task, as you learned about earlier in this chapter. The conscious mind also relies on the subconscious to see if there are patterns it can apply to the task. This assumption of patterns can lead to mistakes.
So how does a designer create something that’s as usable as possible before talking with users? Suggest to the designer that he follow Norman’s (2002) principles for transforming difficult tasks into simple ones:
When you’re exposed to a task for the first time, you bring your life experiences, your beliefs, and your own set way of doing things to try to complete that task. The same is true of other people, and as this book has mentioned before, it can be hard for designers to realize that fact.
These life experiences, beliefs, and other methods that a user has built up over the years—even if it’s asking another family member for help—build a conceptual model for her. A conceptual model is a person’s idea of how she should perform a certain task.
When a person tries to perform a task for the first time, she goes step by step through creating this conceptual model by asking three questions (Norman, 2002):
Contrast the example of using a cup with using a cell phone, which is far more complicated. The small LG Electronics phone I previously used has several buttons on top of the numeric keypad, and I don’t have any idea what they’re used for without looking at the phone documentation. What’s more, the star, zero, and pound buttons at the bottom of the keypad contain different symbols, and I don’t know what they mean just by looking at them.
I traded in my LG Electronics phone for a Palm Treo 650. This handheld PC is much more powerful and versatile than my old cell phone, but it’s also much more complicated. For starters, there is a screen at the top that’s blank, and there is no indication about how to turn on the unit until you look further down the front of the unit. The front of the unit has six buttons plus a center button ringed with four arrow buttons. There are also a couple of buttons on the side that don’t have labels, so that’s even more confusing. The buttons on the side are small and don’t give much of an indication about what to do, although some buttons do provide clues. One small button has a picture of a home on it, and another has an envelope on it, so that likely means these are buttons to go to the home page and the e-mail program.
Yet these buttons don’t tell you how to turn on the device. There are two buttons that have pictures of a telephone handset on them. However, if you’re unfamiliar with the on-off symbol, which is a letter O with a line going outward from the center beyond the top of the O, you wouldn’t know that the right phone handset button, which is red, is also the button that turns on the Treo.
My guess is that few people would consider the Palm Treo a good match for the user’s conceptual model. Given that the Treo is an example of what not to do when you want to create a good conceptual model, how do you design a product that does create a good model? Norman (2002) suggests four features that designers should consider as they create a product:
Now that you’ve put together a template and organizational model for your paper prototype test, you need to interview team members about the conceptual model they construct when they use the database application. This conceptual model will help you and Evan determine what tasks to add to your paper prototype list.
Your team’s model may also result in changing the existing user interface application to make things work more efficiently. You and Evan noted in your initial interview in Chapter 3 that there were several suggestions to change the existing application to integrate the new features and make the interface as a whole more usable.
The two of you decide to begin by scheduling a meeting in which all 10 team members take the MBTI test, which you find on an MBTI site on the Web. You give the test to the team members because you want to create personality types for the primary user persona that you’ll develop in Chapter 6. You also ask the team members about their conceptual model and correlate their answers to their psychological model.
After you and Evan give the MBTI test to all 10 members and analyze the results, you find that you have a good cross section of personality types:
As shown in Table 5.1, these 10 members fall into eight different categories (with each category denoted by an acronym that describes each employee’s personality types).
Table 5.1. Team Members by Psychological Type
Now that you and Evan know what your users’ psychological types are, you can ask the following questions of each user in the project team:
When you’re finished asking questions, group the answers into different areas. For example, from your questioning, you discovered that Laura, Michelle, and Traci want to make it easier to call product companies directly from within the application because they’re the three people who are most likely to call these companies for information. You can then alter the tasks and add more features in your paper prototype to fit.
For example, you may want to add a link to the phone number in the contact information that will call the company via the stores’ voice over IP (VoIP) phone system when the user clicks the link so the user doesn’t have to dial the number. You can then add a mockup of the phone in the paper prototype test so that the project team members can see that if they click the phone number in the database application, the VoIP phone system will activate and dial the number.
In the next chapter, you’ll ask more questions of the team members so that you can understand their needs and goals. You’ll learn how to take all the information you’ve gleaned from these interviews and create a primary persona for your application.
This chapter began with a discussion about the psychology of user actions and user misunderstandings and misinterpretations and how and why they happen. It also covered how personality types identified in the Myers-Briggs Type Indicator can affect users’ actions and how those temperaments can affect the type of questions they ask when being persuaded to do something.
Knowledge in the brain versus knowledge in the world was covered next. You learned how most people rely on imprecise knowledge to get through a situation. This is largely because our brains can accept only so much information, and we break up more complex tasks into smaller ones so we can keep track of everything. You also learned how people are reminded of knowledge by the world, and what the trade-off is using knowledge from the brain and from the world.
A discussion of task structures followed the topic of breaking up tasks. It talked about the different types of task structures: wide and deep structures that provide a large number of choices, shallow structures that offer a top-level choice and a few subchoices, and deep and narrow structures that provide step-by-step instructions for completing a task. Our conscious and subconscious behavior determine the type of task we use.
Next, this chapter covered transforming difficult tasks into simple ones to allow digestion of a user interface. It discussed seven task simplification principles that you can use in any design situation. The seventh and final principle is the most important one: When all else fails, standardize.
The chapter included a discussion on how computer users bring a conceptual model with them in their brains when they approach a new user interface. That conceptual model is based on the users’ past experiences, beliefs, and ways of doing things. The chapter concluded with the step-by-step process that users go through when they create a conceptual model.
Now it’s time to review what you’ve learned in this chapter before you move on to Chapter 6. Ask yourself the following questions, and refer to Appendix A to double-check your answers.
1. What are affordances?
2. What are constraints?
3. Why do people consider themselves helpless when they fail at a task?
4. What does the MBTI test do?
5. What are the seven stages of human action?
6. What are the trade-offs between knowledge in the brain and in the world?
7. What task structure is the most challenging for people?
8. Why must you be deliberate when you’re using your conscious mind?
9. Why do you transform difficult tasks into simpler ones?
10. What makes up a person’s conceptual model?