Human-computer interaction (HCI) has its own models to describe the way people and computers interact together. Among many others, one of the most notable HCI models is Interaction Model (HM). The Interaction Model describes that people use computers by having a computer system perform tasks on their behalf and provide information in response to their requests and behavior.
This model introduces what kind of behaviors are expected from human users when interacting with computers, and also defines how computer systems should react to these behaviors through user interface elements including input/output devices and display screens.
What is Interaction model in HCI?
- The definition of an interaction model
- The characteristics of an interaction model
- Pros and cons of interaction model in HCI
The definition of an interaction model
An interaction model, sometimes called a task model, represents an abstract description of one user’s perception of system functions and purpose from his or her point of view. Task models describe what users do (activities) with systems and how they do it (tasks). It’s about what is being done at any time.
This definition makes interaction models synonymous with task models and differentiates them from interaction scenarios. Interaction scenarios are more like stories that show users performing tasks within their environment using technology as a medium.
The activity-centered approach to defining tasks and designing interfaces has been used since at least as early as 1977. It got a big boost when it was adopted by GOMS (goals, operators, methods, and selection rules) in 1983.
The reason for adopting an activity-centered approach to user interface design lies with one of its tenets—you should provide users with an interface that supports activities they already do naturally.
Interaction models are always specific: for instance, if you’re writing a paper on hiking equipment, your task model would describe your audience’s perception of what it means to buy or rent hiking equipment and how they go about doing it. The primary advantage of using an activity-centered approach is that it helps you identify tasks and goals early on. An alternative to task modeling called goal modeling attempts to achieve a similar end.
The basic difference between task models and goal models boils down to what you use them for: a task model describes tasks people can do naturally, while a goal model describes goals they want or need to accomplish. However, both give designers insight into what users do with their products. One major difference between task and goal modeling is that goals change frequently, whereas tasks are more stable. So, it’s better to use a task-centered approach when you’re designing something like an application or service since your users will have specific ways of doing things that you can code into your interface.
The characteristics of an interaction model
Interaction models characterize not only what you have to do, but how. And, ideally, they also tell you whether or not it’s possible for your users to achieve a goal that was previously considered impossible. They capture and communicate assumptions about a product’s design so that all team members can work from a common language and vision. For example, if our interaction model specifies that users are able to log in by keying letters on their mobile phone keypad (which most of them can), we could then plan our sign-up process based on that assumption.
Interaction models make sure every team member understands how every other team member thinks and feels about design—and helps us avoid potential miscommunications, misinterpretations, and iterations around wrong assumptions. Characteristics of a good interaction model include: Validity – The user can perform what’s required with available technology, infrastructure, and time. Consistency – Any variations from one step to another are justified and consistent with usability standards for your type of product. Flexibility – The process accommodates a wide range of users, devices, technologies, tasks, environmental conditions, and constraints.
Usability tests are especially helpful here to identify weaknesses that need to be addressed. Simplicity – All information should be front-loaded upfront; users shouldn’t have to click back and forth between pages or try different approaches to get through tasks. Relevance – If users can’t understand what they need to do, they won’t complete a task. Some information may not be necessary (especially if there are better alternatives), but don’t assume anything.
Don’t hide necessary information just because it would look ugly or be hard to implement, and try to keep text and graphics as concise as possible. Consensus – Every team member needs to agree on a user interface approach that supports your interaction model—and how each person feels about it. Design by committee results in confusing designs and too many alternatives presented at one time; unanimity usually occurs with very little discussion.
Pros and cons of interaction model in HCI
A few notes on both pros and cons: A huge benefit of having an interaction model is being able to thoroughly evaluate and test your interface before you commit to it. You’ll be able to figure out how effective your interface will be at guiding users’ attention, behaviors, and emotions. This knowledge can save your team countless headaches later down the line as you encounter unexpected behaviors or problems with certain elements of your system.
No matter how much effort you put into creating an awesome user experience, there will always be a learning curve for users; no interface exists that immediately allows new users to start using it without any guidance or training whatsoever. With an interaction model, you’ll be able to determine how much training your users will need.
Even with a well-crafted interface, it’s likely that your users will still have to learn some new behaviors and techniques in order to successfully navigate your UI. A benefit of using an interaction model is that you can reduce learning time by distilling out repetitive tasks that don’t add value to your product and creating shortcuts or prioritizing critical tasks first. You can also break down complex interactions into smaller, less intimidating chunks.
On that note, there are definitely some disadvantages to using an interaction model, especially if you or your team isn’t familiar with interaction design.
Building an interface without one can allow for a lot of freedom and creativity when it comes to laying out screens and making decisions about how your product should work. However, you’ll likely have to deal with more unexpected behaviors and problems as you get deeper into development. Ultimately, creating an interaction model will save you time later on as you’ll be able to avoid wasting resources on developing functions that users don’t need or won’t use often enough.
