Fil Salustri's Design Site

Site Tools


Human-Machine Interaction Loop

The Human-Machine Interaction Loop (HMIL) is a model of a single “interaction” between a user and a thing being used, and helps focus attention on all the HF aspects that must be attended to during design.

What is an interaction loop?

When a human interacts with any machine, the general set of activities that happen is always the same. These activities can be modelled in various ways.

A single HMIL represents a single task, like raising the volume of a radio or grasping a blender on an upper kitchen cabinet or setting the speed of a lathe. Every task can be represented with a HMIL.

Fig. 1: The HMIL per Chapanis (1976).

One classic model is that of Chapanis [Cha76], shown in figure 1.

  • In this model, both the human and the machine are represented, albeit only implicitly, by the same general three-part grouping: sensing, processing, and actuating.
    • Actuation by one produces a flow of mass, energy, or information that is sensed and ultimately acted upon by the other.
  • While Chapanis's model was intended to treat information flows only, we can through systems thinking see that the model works just fine for flows of mass and energy too.

A (slightly) better model for our purposes is shown in figure 2.

  • Colours help identify the three types of elements, high-lighting the similarities.
  • All inputs and outputs have to pass through some kind of interface, so the machine “inputs” and “outputs” from outside the human-machine system are now directed through the interface elements.
  • Each of the human and machine are more readily identified as of similar general structure (they're both just systems).
  • We clearly mark a boundary between not only the human and the machine, but between each of them and their environment as well.

Fig. 2: A more sophisticated model of HMIs.

The Human-Machine Interaction Loop (HMIL) shown in figure 2 is the one we will refer to subsequently.

Consider the following examples.

Example: Adjusting the volume on a radio

Assuming you know how to adjust the volume on the particular radio in question, think of the specific activities involved with respect to the six elements of the HMIL (figure 2).

  1. In this case, the feedback has two components: the music coming from the radio, and the current position of the volume control.
  2. You observe this feedback; you hear the music and see the volume control setting.
    • At this point, we can already begin to consider the human factors that can come into play.
      • How is your hearing?
      • How is your vision?
      • Even if your vision is good, is there enough light to see the volume control setting?
      • How much other/ambient noise is there that the radio has to compete with to be heard?
  3. Your brain does some cognition to process the observations.
    • You decide if the sound from the radio is too loud or too soft.
    • You determine (cognitively) how to use the control to change the volume.
  4. You actuate the volume control to change the volume.
    • More human factors apply here:
      • Are you able to operate the control?
      • Is it too far away, or too stiff to operate, or too close to other controls that block access by your hand?
      • How much attention do you have to apply to operate the control?
      • Is the “fineness” of the control too fine for your hand to make minute adjustments?
  5. The new volume control setting is used by the radio to change its loudness.
    • Now that there is a new volume level and a new volume control setting, you return to step 1 above. This loop repeats until you have found the most suitable radio volume.

Implications for designing

A well-designed product will promote safe, easy, even enjoyable interactions. To understand the nature of that those interactions via analysis of HMILs is very important.

Consider the radio volume example above. Clearly, the design of the volume control will be dictated in large part by the need of the user to participate safely and easily in the HMIL outlined above.

Exercise for the reader Enumerate as many design decisions as you can think of that will help to ensure the volume control is safe, functional, and usable. Justify each decision by referring back to the HMIL above.

As usual, HMILs are analytic tools, so they don't actually help you design something. However, they do help you understand the design situation better so that you improve the odds of designing a good product.

An interesting case study about how humans can misunderstand what they see in a product designed by other humans - and that results in disaster - is the slide that led to the Columbia Shuttle disaster.


The problem of granularity can cause you to go into too much detail with HMILs.

In the example of the radio volume control, it may not be relevant if your job is to design a whole car dashboard. You will have to depend on the scope of the design brief to guide you: you may not need to design all the details of a product.

Furthermore, in scholastic settings, you may simply not have time to study all the HMILs of all the possible components of a design. In this case, you are advised to find a few particularly important components and focus on them. If you can demonstrate that you can (1) justify your choice of important components, (2) perform good HMIL analyses of those components, and (3) admit that the remaining components simply cannot be done within the timeframe of your project, then you'll have demonstrated appropriate mastery.


Cha76. a A. Chapanis. 1976. Engineering psychology. Dunnette, M. D. (ed.), Handbook of industrial and organizational psychology. Rand McNally, Chicago
design/human-machine_interaction_loop.txt · Last modified: 2021.06.18 12:13 by Fil Salustri