Experiment One | Some Results
February 28th, 2007 by Kevin Capota
We’ve just finished writing the paper for the Pleasurable Products and Interfaces’ conference, so it’s time to present some results. For now, I will post three scoring cards for the visual appeal of websites. The three websites clearly differ as the first is experienced as positive and the second as negative. The top 10 of emotion words related to the websites indicate a strong relation with the position on the circular structure. The last website elicits mixed feelings and has both positive and negative evaluations.
The website that was rated both negative and positive:
[…] Nevertheless, when online retailers are going to see the significance of focusing on customer experiences, just like they do in the normal store, they need tools to help them out. Monito is developing the LEMTool, where emotional impact and experiences are measured during online interaction with, for example, an online store. Some results can already be seen between the visual appeal of websites (see LEMTool blog). That could be a good start. […]
hi Kevin
first of all let me say I really love this site- you’re a very talented designer.
second, I participated in the online experiment (guess I didn’t win the book after all…), and I thought it was very creative- these are actual websites, right?
I could definitely feel the distinct emotions they evoked, but I wonder, how can you tell which element in the website is responsible for these reactions? Also, I would like to know how are the results you wrote about going to effect the developement of LEM?
keep up the good work (-:
Hi Sonya,
thanks for your comment, and indeed: you didn’t win the book (sorry). I guess I should have mailed all participants to make that clear, but I didn’t want to spam too much. And you’re right; they are all actual websites.
I can’t tell which elements in the website are responsible for the different reactions. That’s a very interesting step, which I won’t cover in my graduation thesis. But, the LEMTool will try to explain certain elements by pairing the ratings to heat maps from mouse movements. I think the most high-lighted elements will be the ones with a big influence. But I can’t prove that in the limited time I have for this project.
The findings of this experiment and the second (that currently runs in a Dutch version) are meant to validate the working principle of the tool. It proved to work for measuring distinct reactions for first impressions of websites. Next, I’ll try to measure how interaction with a website changes user experience. Do the evoked user emotions change? And can I use the same set of emotion words for first impression and the impression after interacting with the interface? These are some questions that I hope to answer…
I’ll write much more about this when the 2nd experiment is finished.
Hello
I would like to know about the Lemtool. Could you help me sending some paper or some article, please?
Best regards
Guilherme
Hi Guilherme,
What exactly do you want to know? We’re still busy developing the tool and for now, I only have information about the basic of the tool.
The basis is the emotion radar, of which the results are shown in a draft version above.
The LEMTool will be built around this emotion radar. It will trigger the radar at specific points while interacting with a website/interface. It may be triggered by server data (like exit-points), mouse behaviour, or by set points in time.
The results of the emotion radar will be backed up by (again) mouse behaviour (as heat maps), server data, and by feedback from the user/tester.
Eventuelly, it should give a clear image of user experience with an advice how to improve the experience.
Hi again, thanks for the reply
yes, limited time is always a problem..
How do you expect the mouse movements to corellate to the emotional reaction?
I was also wondering ,in the first experiment did the websites always appear in the same order to all participants (if the answer is yes, it could be interesting to see if the results are the same in a randomized order of appearance)
Looking forward to the final results,
Sonya
Hi, I don’t think in terms of a correlation between mouse movements and emotional reaction. Not for the LEMtool I mean.
You should see it more like an ‘extra’ to the other results. It’s a subjective interpretation rather than a scientifically grounded interpratation of the mouse movements. The purpose is to identify elements that may influence the emotional reaction. For example, the LEMtool may ask a user to comment about certain spots on the website after comparing it with the real-time heat map of that user. The most-highlighted elements would seem obvious to be influential, but.. that is something to be proven later.
Perhaps there are correlations, and it would be very interesting to study them. I haven’t really focused on this subject, so I can’t say much more about it.
The websites were all shown in a randomized order of appearance. And the first results of experiment 2 are coming up monday or tuesday!