AA #2: The Inference of Perceived Usability From Beauty. Marc Hassenzahl and Andrew Monk, Human-Computer Interaction Vol 25.

07 Oct

The second article I am reviewing is entitled Inference of Perceived Usability from Beauty. The article takes a very rigorous approach at establishing why it is, or what it is, that makes a user correlate good usability with beauty: for example, it investigates why when we (users) go to a well designed website, we sometimes automatically assume that the usability is sufficient, or alternatively, are much more willing to tolerate poor usability. It conducts several quantitative studies, exploring the covariance between products and participants in an attempt to address the problem. This study was brought about when 15 papers reporting 25 independent correlations of perceived beauty with perceived usability showed high variability in reported correlation coefficients (Hassenzahl and Monk, p.238). They theorized that this is the result of methodological inconsistencies and limitations due to lack of further specification of judgmental processes (Hassenzahl, p. 240). Ultimately, the tests found that the relationship between beauty and usability has been ‘overplayed’; the correlation that is commonly known to exist (pragmatic quality vs. beauty) has a mediator: “goodness”.

The article thoroughly discusses how HCI, over time, has incorporated a heavy dose of aesthetics into its testing models. ‘Beauty’ has become an intricate part of the designing process, and aesthetics as a result have become a part of user experience research, helping researchers define ‘good’ and ‘bad’ graphics, but ultimately-what is and is not working. The primary problem this presents is many have assumed that the user’s perception could be easily quantified. It has been established that a correlation does exist between aesthetics and usability; however, that does not automatically assume that aesthetically pleasing is synonymous with good usability. This article strives to define a challenge: where many have assumed the previous correlation is true, Hassenzahl has conducted studies prior to this one that points to a different conclusion. Thus, the study has been conducted to strengthen and clarify the correlation that resides within. The article criticizes how, regardless of how unintentional, different researchers utilize different terminology that can be similar in context. When the time for testing occurs, the researchers understand one another, but in many cases the population tested does not. Different motivators also contribute to different results. The article mentions as well the need to quantify emotion associated with content-not because it actually is needed, but rather it has been a catalyst to propel the ongoing correlation: “what is beautiful is usable.” The article itself tests four separate groups, varied in ages, with surveys and websites on information and product that ranged from very poor to very high aesthetics, and varied levels of usability. Each group had a specified rubric to follow on grading websites, and each group was unaware of the other groups’ tasks, as each had a different set. The results generated concluded that when addressed in a particular fashion, results were extremely varied and skewed. However, once variables were declared as more concrete-as in hedonic quality vs. pragmatic-the correlation held true. It shows that an explicit model can lead to better understanding of what the relationship between beauty and usability signifies. Ultimately, the article proposes a new method to aesthetics research and leaves the door open for new variables to be added, such as trust.

This article in particular intrigues me for two reasons: 1) in the world of computer Graphics, we have thrived on the idea that we can manage with ‘just enough’-just enough shading, just enough graphics, textures, etc. Unfortunately, at this point we have reached a problem area; we teach our students the ‘just enough’ theory, yet some students want more than just enough (to which the obvious solution is to go out on your own and explore-the internet can be a somewhat safe place), and most importantly hiring professionals are tired of tolerating just enough-they want more. Beauty is becoming something such that-well, as my friends and I say about a lot of games-while it looks nice, that doesn’t mean it works for squat. the other reason: our work with nanoHUB is somewhat of a victim to this ideology. The nanoHUB site itself doesn’t looks bad-it actually looks very clean and professional. However, feedback and retention rates of its use is deplorable. The ongoing complaint is that it is ineffective in its use, and the population it caters to generally doesn’t like it. It makes me curious to see how much further this type of work can go, and what other factors will be taken into account. I support this article very strongly-HCI, CGT, Entertainment, etc. are functioning very heavily on the face value. The ending result: varied, unclear data; animators that don’t get it, games that are gorgeous but lack true value; movies that make a killing because it looked good, and then you hear from across the cubicle how it’s actually pretty terrible. I find that going back to what needs to be focused on (is this usable? can it be learned easily? does it fulfill its purpose?) rather than how good it looks is key to turning a lot of problems (in this sense) around. This is part of why I brought into question the need for content oriented design over user design-it’s more difficult to discern what is needed when you’ve covered it up with something that looks nice. Aesthetics should be at the end of the to-do list, and I think that worrying about the appearance before you take care of the usability is a sign of such.

-Vanessa B!

Reference: Hassenzahl and Monk. The Inference of Perceived Usability From Beauty. Human-Computer Interaction Vol 25. p. 238-253.


Posted by on October 7, 2010 in Uncategorized


4 responses to “AA #2: The Inference of Perceived Usability From Beauty. Marc Hassenzahl and Andrew Monk, Human-Computer Interaction Vol 25.

  1. Mihaela

    October 7, 2010 at 7:42 pm

    I’m glad you read this, I saw it in the table of contents, and was hoping someone would read it for me 😉

    Your summary is sufficiently descriptive, and your analysis very thoughtful. I’d like to ask you to proofread or spell check in the future, and to use APA style for the citation at the bottom.

  2. msvanessab

    October 8, 2010 at 2:17 am

    I fixed it!
    I’ll be less dumb and more awake next time. 🙂

  3. Rebecca Ivic

    October 19, 2010 at 5:15 pm

    Where does ‘goodness’ fit in as a mediator? I’m unclear on this, but intrigued at the concept.

    That being said, I believe the points you brought up about the link between aesthetics and usability is an important one. Both concepts need to work together, but one can’t necessarily overpower another.

  4. msvanessab

    October 20, 2010 at 1:53 am

    “Goodness” is what the authors were referring to as a means of measurement. However, it was an accidental mean of measurement: “goodness” in their eyes, was what the user was believing to be rated. In turn, the assumption was that having a higher rating in “goodness” meant a user was more likely to pass a product off with good usability. For example: a website that initially is extremely aesthetically pleasing; in this case “goodness” would be how nice it was from a visual perspective. If it registered ‘high enough’ with a user in terms of a rating, they were willing to forgo the fact that it was difficult to accomplish a task, or that some content was missing, because ‘it looked good.’ Thus, it ties back to their theory of ‘what is beautiful is usable’, and how it is in some cases, not true.

    Does that help? 🙂


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: