Attractive Things Work Better - or does it just seem that way?

Usability and Aesthetics - The Story So Far

If you look at HCI literature, it will soon become apparent that usability dominates as its focus.  Although System Concepts uses the International Standards definition which breaks usability down into effectiveness, efficiency and satisfaction, some experts such as Jakob Nielsen place usability within a framework of system acceptability.  Here he separates out such inter-related factors as cost, likeability, functionality and of course, usability and aesthetics which all contribute to a decision making process when determining system acceptability.  Surprisingly, little work has been carried out to observe how usability interplays with factors such as aesthetics.

Do “Attractive Things Work Better”?

Don Norman asserts that “attractive things work better” but how valid is such an assertion?  The purpose of my research was to explore the association, if any, between user’s perceptions of usability and aesthetics for developed systems.

In the testing sessions, we used three visualisation variations for representing the results of a search tool - Radial interactive visualisation, Sammon cluster visualisation and Dendro map visualisation as illustrated in the screen shots below.

Collecting Usability and Aesthetic data

Twelve participants took part in an experimental design consisting of two tasks and the above three visualisations. Each user participated in 2 of the 6 conditions.

An aesthetic measure was developed following the format of an established usability measure (System Usability Scale (SUS)).  Judgements for both aesthetics and usability were elicited before and after completion of each assigned visualisation and task.

We also measured performance for each test run in terms of errors and task duration.

We found that before the tool was used, there were strong relationships between judgements of aesthetics and usability for the tool as a whole and individually for sammon cluster and dendro map visualisation i.e. the users thought the more attractive ones would be more usable.  After use, the strong relationship remained or was present for the tool as a whole and individually for radial interactive and dendro map visualisation.  From this we concluded that attractive systems are perceived to work better.

Usability and Aesthetics – The Story Now

However, we did not find an association between aesthetics and performance, which suggests that attractive systems may not actually work any better than unattractive systems. 

Norman suggests that the reason why attractive things may seem to work better is that pleasing aesthetics result in positive affect (emotion) which in turn influences the cognitive system, opening the mind to creative thinking.  Hence systems are perceived to be easier to use.

However, our research suggests that this reaction only shapes our judgements and not our actual experiences.  Attractive systems may not actually be more usable in practice.

Making products or systems more attractive should increase the user’s anticipation of their usability and this enhanced satisfaction can make them believe their use of the system or product really was effective and efficient.  We would therefore argue that aesthetics should be acknowledged in the development of any system to ensure its success with the user.  After all, it is our judgments that guide our decisions, so ensure a good first impression is formed.  Good aesthetics are beneficial to a product.  Make products that are attractive and people will feel they are more usable. 

March 2005