How to evaluate interaction design using GOMS models

The quality of human-computer interaction of an interface design should be evaluated before it is finalized. Such testing is usually done using a prototype or mockup design, which can sometimes be costly. This can be avoided by using model-based evaluation to test the interaction design, reducing time and cost. One methodology for this is to use the GOMS models.

GOMS models

GOMS stands for Goals, Operators, Methods, and Selection rules. GOMS models are a family of model-based evaluation techniques. These are used to objectively evaluate human performance on an interaction design without actual users' involvement. Therefore, when time and budget are constrained, GOMS models are used to make predictions about the interface design instead of user testing.

There are four variants of GOMS models, which include the following:

• KLM (Keystroke Level Model)

• CMN-GOMS (Card, Moran, Newell model)

• NGOMSL (Natural GOMS Language)

• CPM-GOMS (Critical Path Model or Cognitive Perceptual Model)

How to evaluate interaction design using KLM

The evaluation technique varies for each of the models mentioned above. For example, let us look at how to evaluate interaction design using KLM.

KLM calculates the task execution time for an interface. Therefore, it compares two interfaces in terms of execution time. KLM focuses on both physical and cognitive activities performed to complete a task. Following is a step-by-step breakdown of the method of calculating the KLM value of a task.

Step 1: Choose the task

Suppose that the task is to save an MS Word document using the "File" menu. The first step we need to do is to identify the sequence of steps that need to be performed. For this task, the following steps will be performed:

1. Grab the mouse (assuming the user was typing on the document using the keyboard).

2. Move the cursor to the "File" menu.

3. Click the left mouse button.

4. Move the cursor to the "Save" option (various options will be displayed after clicking on the File menu).

5. Click the left mouse button (assuming the prerequisites of saving, such as naming the document, have been done).

Besides these physical actions, we must also identify the cognitive activities required to complete the task. An example of a cognitive activity for this scenario is to locate the mouse and decide to grab it. All such decisions are categorized as mental activities and are also considered in the final task execution time.

Step 2: Identify the KLM operators

KLM uses five operators to evaluate the total time. Each operator takes a specific time to execute, which is visualized in the table below.

Next, identify the sequence of operators used in the task under consideration. For this example, the following operators are used:

1. M: Taking the cognitive decision to move to the "File" menu.

2. H: Switching from the keyboard to the mouse.

3. P: Moving the cursor to the "File" menu.

4. B: Clicking the left mouse button on the "File" menu.

5. B: Releasing the left mouse button.

6. M: Deciding to select the "Save" option from the menu.

7. P: Moving the cursor to the Save option.

8. B: Clicking the left mouse button on the "Save" option.

9. B: Releasing the left mouse button.

Therefore, the sequence of KLM operators for this task will be MHPBBMPBB.

Step 3: Compute the execution time

Now, we'll compute the final execution time using the respective time of each operator given in the table above. The total execution time will be:

$1.35 + 0.40 + 1.10 + 0.10 + 0.10 + 1.35 + 1.10 + 0.10 + 0.10 = 5.7 seconds$