What I Have Learned in Human-Computer Interaction. Part 1

This is part one of multiple posts. Attempting to summarize the experience I’ve gained on HCI is just too much to produce on a single post.

After 5-6 months of hard work, I have finished my first graduate course on my way to getting a Master’s in Computer Science. It was challenging to navigate papers and readings while attempting to manage life struggles, but I am 1 class closer to my degree. I must say that I did enjoy Dr. Joyner’s explanations. Each topic seemed very easy to understand, and I am astonished about how much I’ve learned.

Great Interfaces are “Seamless”

Exemplary Interfaces are like Ninjas: they are more effective the less they are seen. This invisibility can be achieved through practice, but that just translates as your interface has a steep learning curve. Ideally, the interface design doesn’t make the user “think” too much about their choices.

You Cross Two Gulfs when Interacting with an Interface

To turn on a light, users have to first think of what they need, understand their possible actions, and then act upon one of those actions. This separation between a user’s goal and the outside world is called the Gulf of Execution. The larger the divide, the more painful it is to begin to use an interface. As a designer, you want to minimize the cognitive load required to use your interface. Once the user has interacted with the interface, the user has to see the world’s state, interpret that change concerning their goal, and finally evaluate if the goal was achieved. This second divide is known as the Gulf of Execution. As the Gulf of Evaluation, it is better if a designer minimizes the distance between the outside world and the user’s goals. Now, if the goal isn’t achieved after the first interaction, the user loops around and starts on the previous gulf, creating a feedback cycle until they reach their final purpose. Knowing the user’s steps to interact with an interface allows us to minimize the user’s work to reach their goals. The gulfs are a fantastic way to note the sub-interactions made with an interface.

Design is a Cycle

Interface design takes 4 steps: needfinding, design alternatives, prototyping, evaluation. The needfinding stage is filled with data gathering. The design alternatives step is mostly planning. One needs to create and test interface mockups on the prototype step. The data gathered from the user feedback is evaluated on the evaluation step. Lastly, using the data from the evaluation step, the design lifecycle starts again until the interface’s goal is reached.

Never forget to look for the Inventory of Useful Data

There is nothing more important than knowing your user and their needs. To do so, we must look to answer the following questions:

  1. Who are the users? (age, gender, expertise)
  2. Where are they? What kind of environment are they in when using this interface?
  3. What else asks for their attention?
  4. What are they trying to accomplish?
  5. What do they need right now?
  6. What is the main task required to achieve their goal?
  7. What are their subtasks? What tasks can we derive from the primary task?

When Doing Research, Be Aware of Biases

Due to the importance of data, one must be careful to only add valid information to your findings. There are multiple biases to be aware of, but here is a list of the ones that I tend to run into more often:

  • Confirmation Bias: we only see what we want to see.
  • Observer Bias: one is more helpful during research when the research attempts to support our own ideas.
  • Social Desirability Bias: people are naturally more willing to help than to go against you.
  • Voluntary Response Bias: people with strong opinions about the research are more willing to participate.
  • Recall Bias: people have a hard time recollecting events.