« Back to Writing

Response to: Seductive Interaction Design excerpt

This post was written for UW's HCDE 517, Usability Testing, in response to one of the course readings

Summary

In this chapter from Seductive Interaction Design, author Stephen Anderson challenges the traditional view that a great product user experience can be attained by making it highly usable through usability testing and design iteration. While not discounting the benefits of making a product highly usable, Anderson argues that improving a product's usability merely clears the way for a product's true potential experience to be attained; usability removes the friction that users must face when trying to use the product. Instead, he argues that there's far more behind a successful, well-design product, and that designers need to focus on the underlying psychology of the product in order to make it truly great. A product that no one wants to use is not a good product, so product designers would do well to strive to understand what really motivates their potential users why they would be compelled to use it in the first place and continue to use it over time.

User Experience Hierarchy of Needs

In support of his argument, Anderson defines six stages of maturity in his User Experience Hierarchy of Needs, with the earlier stages providing a foundation and the later stages serving to differentiate:

Functional
Paramount to all else, a product must do what it is intended to do. If a product can't be used to do what the user wants, then it is useless.This quality is often sufficient for an emerging technology's success, but is a merely one of many requirements when competition exists.
Reliable
A product should be consistent, accurate, and precise so that it instills trust and confidence.
Usable
A product should function in a way that does not introduce speed bumps or obstacles which a user must overcome to use it fluidly and with minimal errors.
Convenient
A product should work in a way that a user would expect or fits within the user's mental model.
Pleasurable
A product should have a positive effect on the user's emotions, whether through pleasing aesthetics, friendly language, and being engaging.
Meaningful
A product should be meaningful to the user, such as through the beliefs it represents or the communities to which it connects its user.

Affect's Effects

Anderson challenges emotion's understated effects on a product's user experience through three examples, ultimately arguing that emotion and cognition are not separate, as a traditional view might suggest, but that "our experiences are based on perceptions."

First, he highlights the importance of a product's personality, which is often dependent upon branding and presentation style. Personality, he says, can strongly impact the expectations a potential user will have, which in turn influences the likelihood of the user having a positive or negative experience. Products with identical task flows will be experienced differently depending upon the personality.

Second, he describes the sense of trust that products convey. Noting that impressions are formed incredibly quickly from a product's attractiveness, he gives examples of poor or sloppy visual design contributing to a sense of distrust. Even if a product is highly usable, its experience could be tainted by distrust.

Lastly, he provides the example of time, stating that visual tweaks can make activities of identical duration be perceived as happening more slowly or quickly.

Overall, Andersen argues that we cannot truly separate affect and emotion from cognition and logic, and that our experiences therefore are tied to how we perceive a product. In fact, studies have shown that attractive products not only seem to work better, they actually can lead to shortened task time and reduced mistakes, even if all other aspects of the design are identical. In light of this, Andersen advises elevating aesthetic considerations to a level near those of task-based usability.

Critique

While no credible user experience design expert would argue that aesthetics and branding aren't an important component of a product's user experience (I think we were all confused by Jakob Nielsen's pre-2013 useit.com), fewer might be as adamant as Andersen, especially in his assertions that performance is measurably dependent upon perception to an extent that the experience is changed. Even having studied cognitive science, I still found Andersen's points about the interconnectedness of emotion and cognition to be revelatory.

Of course Andersen is right, but it's complicating, if not inspiring, as a designer to be tasked with this additional level of complexity in crafting a positive user experience. Not only does this require a shift in how we approach design problems, but it may also have impacts on the iteration process, especially as it pertains to presenting lower-fidelity designs to stakeholders, clients, or potential users. UX designers have been taught to present wireframe designs with minimal visual treatment to clients in order to begin to solidify the information architecture and interaction flow, but perhaps wireframes encourage designs that are too utilitarian.

This applies to usability testing, as well. Traditional usability teachings mostly focus on testing the ability of the users to perform assigned tasks, often encouraging designers to test unpolished or prototyped versions of the final design in order to avoid spending time perfecting something that may very well change because of the tests' outcome. However, given the inseparable nature of perception and experience, maybe researchers would benefit from approaching usability tests with more refined designs or with aesthetic variations.

Discussion Questions

  1. How does the message of this excerpt change the way you think about usability testing? How might it change your approach to testing?
  2. How can visual perception be properly tested in usability tests? Should it be tested along with the interaction? Or should (can?) the interaction be tested separately from the visual design and branding?
  3. What implications does this have outside of usability testing? For example, how can this be reconciled with the benefits of showing low-fidelity wireframes to stakeholders? Are the differences significant enough?
  4. What other types factors may be affected by perception (beyond time, trust, and personality)?

« Next: Static Visualization Critique: Cost-of-Living Index

Previous: Response to Personas: Practice and Theory »