Session Overview: 10:30 – 11:45, SES  New York 2010

Eye Tracking Research Update
Eye-tracking provides valuable information about a Website visitor’s focus of attention. This information is crucial in improving the design of Web pages and products to enhance the visitor’s experience. Effective search engine marketing relies on understanding user interactions with continually evolving search engines. Google’s universal search, which now includes media results, is one such evolution. This panel will discuss a neuromarketing study performed by One to One Interactive which analyzed user behavior through physiological, emotional, and qualitative data. Also covered will be the newest information in eye tracking of SERPs and the latest on heatmaps.

Speakers:
Shari Thurow, Founder & SEO Director, Omni Marketing Interactive
Jeremi Karnell, Co-Founder & President, One to One Interactive
Susan Weinschenk, Chief of User Experience Strategy, Human Factors International 

OTOinsights’ recent efforts to measure the user impact of a new generation of search engine result pages which include multimedia elements such as images and videos has caught the eye of two prestigious upcoming conferences.  Both the Search Engine Strategies (SES) (Chicago, IL- December of 2009), and the Usability Professional’s Association (UPA) International Conference (Munich, Germany – May 2010) has asked OTOinsights to present their findings from its report titled “Implications of User Engagement with Search Result Pages“.

This paper, which may be downloaded for free on OTOinsights website, was most recently presented at the UPA Boston Conference held at Bentley University on May 26th, 2009.  That presentation is provided below:

Aspects of the above presentation will be discussed on a panel titled “Eye Tracking Research Update” at 4pm December 9th, 2009 at the SES Conference and Expo in Chicago, IL. 

Probably even more exciting to us is the opportunity to present our findings at next year’s UPA International Conference in Munich, Germany.  For the first time in its history, the UPA conference will travel to the heart of Europe. Their key theme will focus on how UX professionals can create great user experiences across different cultures, meeting the challenges of delivering to diverse audiences. This prestigious event includes a rigorous process to review both general presentations and peer reviewed papers.  To that end, we  received very high marks for Dan Berlin’s excellent submission.  Indeed, the conference reviewers expressed their excitement about the inclusion of new Neuromarketing research at the event.  

Screen shot of the detailed reviewer scores and comments below:

(Click to Enlarge)

(Click to Enlarge)

I urge everyone to stay tuned as we are about to publish the second installment to our Search Engine research, titled “User Engagement with Yahoo! Rich Ads in Search”.  We also plan to incorporate this new research into our upcoming presentations.

Brain Measurement for Usability Testing and Adaptive Interfaces: An Example of Uncovering Syntactic Workload with Functional Near Infrared Spectroscopy – Tufts U.

Wow, that title is a mouthful and the presentation was absolutely intriguing.  The authors used functional near infrared spectroscopy (fNIR) to determine users’ cognitive workload while interacting with an interface.  fNIR measures the amount of blood that is flowing through areas of the frontal cortex.  The relative change in blood flow in the frontal cortex indicates a working mental load.  While using an interface, a person has two sources of workload: using the interface itself and performing the task.  They designed a special interface to separate these workloads which are then compared to the workloads during a usability study.  There is plenty of work to be done before this type of methodology is mainstream, but I think it will be very important to usability practitioners and neuromarketers everywhere.

O’Game, Can You Feel My Frustration?: Improving User’s Gaming Experience via StressCam – U. of Houston

This presentation was interesting on two different levels.  First, the goal of the study was to make a game that adapts to the user’s frustration by measuring their physiological state.  But their method of measuring frustration was equally intriguing: an infrared camera to detect heat changes on the forehead.  Their literature review showed that mental engagement increases not only the frequency of upper facial expressions, but also increases the temperature on the forehead, between the eyes (the supraorbital region).  They went on to explain how the game adapted, based on the input from the camera.  Too bad the camera they used costs around $300k.

Discriminating the Relevance of Web Search Results with Measures of Pupil Size – Google & UC Berkeley

I’m always up for hearing about how people are using eye tracking data and this was another great one.  The goal here was to determine if relevant (Google) search results affect pupil size.  They chose this method because they want to find a way to determine the best results without the filter of the user’s cognitive bias.  Using pupil size is tricky though.  The pupil changes size for many different reasons: luminance, affective response, cognitive work load, and overall interest.  They presented the participants with a question that could be answered with a Google search.  They then showed them three image and three text results of varying relevance.  The first experiment found that the relevant text results had a significant change in pupil size, but the image results did not.  So, for a second experiment they showed VERY irrelevant images, along with a relevant one.  This time, there was a difference in pupil size.  It is exciting to see that people are exploring new ways of using eye tracking data to measure the user experience.

Interaction Criticism and Aesthetics – Indiana U.

This excellent presentation, which received a Best of CHI honorable mention, was from our friend Jeff Bardzell at IU.  His main idea was that “criticism and aesthetics are on the threshold of HCI and should be used in an intellectual and rigorous way” (again, loosely using quotes).  Frankly, there is no way that I can accurately convey the details of his intriguing presentation, as the field of aesthetics and criticism is a little above my head.  Nonetheless, it was captivating to hear the definition of these terms, a bit about their history, and the implications they should have on HCI.

Lab Tours

This was not an official activity of CHI, but wrapped up the week nicely.  Today, Google, IBM, and Microsoft offered tours of their Cambridge facilities.  The day started at Google and included a tour of their offices.  All the stories are true: their offices are completely off the hook – toys, food, musical instruments, whiteboards EVERYWHERE, whimsy, and smart people.  Google then gave a series of short presentations on things they are working on that pertain to the HCI community.  We then walked to IBM, which really wasn’t worth the walk other than it was a nice day out.  They setup a hallway with posters and demos of various things they are working on, many of which I had previously seen or heard about.  Ah well, can’t win them all.  Microsoft fed us and gave a series of presentations on the projects that MS Research is working on.  This included surface computing, Windows 7 Usability testing, and some other projects.  It was a great day to walk around Cambridge and I feel bad for IBM and MS, because Google was a very tough act to follow.

And thus ends the CHI updates.  It has been quite a week and I look forward to next year’s CHI in Hotlanta.

Opening Plenary

Even Small Distance Matters: Social Ergonomics in Collocated and Remote Teams – Judith S. Olson, UC Irvine

Social ergonomics is something that all of us deal with daily, but rarely think about.  The most common example of this is ‘personal space.’  When having a conversation, a person will feel different levels of comfort depending on his relationship with the other person and the distance between them (among other factors).  For example, a stranger who is conversing very close to you will make you feel uncomfortable, but not if he is further away.

Research shows that when two people are sitting and chatting in the same room, a 90 degree angle is the most comfortable position for the participants.  Additionally, in conversations, the taller person typically has dominance.  Both of these also hold true for telepresence.  By placing the camera at eye-level and arranging the room so that there is a 90 degree angle between the participants, users will have a much more relaxed experience.

Another example which intrigued me was a study that put 6 coworkers in the same room, no cube walls, just space to get work done.  Evidently, these coworkers were extremely productive because productive ad hoc meetings popped up and everyone knew what everyone else was working on, in real time.

Presentations

Correlating Low Level Image Statistics with Users’ Rapid Aesthetic and Affective Judgement of Web Pages – Siemens

One day, computers will churn out designs and music that humans will find aesthetically pleasing.  This work from Siemens takes us one step closer.  They are developing an algorithm that can examine the screen pixels and determine if the page is aesthetically pleasing.  A study was presented which showed that they are getting quite close to perfecting this technology.  I, for one, welcome our new algorithmic designers (I kid, I kid).

What Do You See When You’re Surfing?  Using Eye Tracking to Predict Salient Regions of Web Pages - DFKI (Germany) & Microsoft

This study presented an interesting eye tracking method to predict salient areas of a Web site.  Three eye tracking measures were used to determine these areas: view frequency, time to first fixation, and fixation impact.   Fixation impact is different from fixation duration, which is typically used in eye tracking studies.  Eye trackers typically track a single point of fixation and don’t account for the fact that we process items in the periphery.  The authors increased the size of the fixation area by 2 degrees to account for the rest of the fovea (the part of the retina with the clearest vision).  The fixation impact is the area of a page element that is covered by this increased fixation area.  They then examined the Document Object Model (DOM) of the page and assigned different areas of interest (AOIs) scores based on the calculated fixation impact.   In turn, they can use this information to essentially predict if code will produce salient areas of a Web page.

Some additional learnings from this presentation:

• Size, position, and level in the DOM tree are the most expressive features
• Users had low expectations for the right side of the page (across different types of Web pages)
• The results align with the notion of a ‘Golden Triangle’ for both information foraging and page recognition tasks
• Women look at each each region of a Web page longer
• Younger participants look at each region of a Web page shorter

Predicting Tie Strength with Social Media – U. of Illinois at Urbana-Champaign

The authors constructed an algorithm that can determine how close-knit two people are on Facebook.  The algorithm is based on the number of connections the individuals have, the number of wall posts (reciprocated and non), messages, and so forth.  Why is this important?  Individuals are not the only entities on Facebook anymore and a company would love to have close tie with its customers.  For instance, have you seen the Skittles homepage lately?

What Can User Experience Learn From Food Design – Panel Session

While not the most utilitarian session, this was certainly an interesting one.  On the panel were the head Chefs of Rialto in Cambridge and Gargoyles in Somerville, an HCI practitioner, and an HCI academic.  Though I didn’t get to see the whole session (ducked out to see the Tie Strength session) the major takeaway was: design holistically – it’s more than the interface, it’s the entire (dining) experience.

Invited Talk: Jan Chipchase, Nokia

Jan spoke at length about his various experiences traveling around the world and watching users in their element.  I don’t have much to say here other than world wide ethnography seems like quite the adventure.

That’s it for now.  Time to rest up for the next four eventful days.