Design of an eTextbook

This project was completed as a requirement for cognitive engineering course in my Ph.D. program.  It was a team project focused on developing early design requirements for an eTextbook (i.e., a digital solution to physical textbooks).  Because of the many tasks that current physical textbooks support (e.g., annotation, bookmarking, highlighting, etc.) and the many different ways students (and faculty) interact with them, our team adopted a work domain analysis to develop an abstraction hierarchy.

Work domain analysis is part of the cognitive work analysis (CWA) methodology which is unique compared to more traditional engineering methods as it focuses on identifying the requirements, constraints, and affordances for cognitive work in a domain.  CWA was originally proposed by Rasmussen et al. (1994) and further refined by Vicente (1999) as a means to support the design of complex systems that would be more adaptable under unanticipated operating conditions.   Since its original application to the design of process control systems, CWA has been adapted and successfully applied to other complex domains and systems.  Across these domains and systems, CWA has effectively been shown to support the design of novel interface displays through its ability to produce models of work objectives combined with constraints and affordances that inhibit or support those objectives.  Additionally, as Bisantz and Roth (2003) illustrated, CWA is particularly effective for design of novel systems requiring adaptability based on unanticipated domain conditions.

The final report generated by my team, along with a design process overview and final recommendations, can be viewed below.

Eureka! Human Factors made clear.

I’m a Human Factors specialist (a.k.a. cognitive engineer, usability designer, information architect, etc.) by trade.  Unfortunately for me though, despite the growing prevalence of  human factors I still have one major problem with the field.  Every time I tell someone “I’m  a human factors engineer” or “I’m getting my Ph.D. in human factors engineering” I am faced with a momentary stare while the person escapes into rapid activation of their neural network starting from somewhere around genetics.  This momentary look, usually accompanied by a smile or nod, is then followed by the question “What does that mean?” (or the dismissing “oh, sounds interesting”).

Well, no more I declare!  Thanks to the 2010 Inter-University Workshop I have a 5 second, visual way to quickly explain Human Factors (I hope).  I shall grab a napkin, an arm, or even a dollar from my wallet and quickly draw two circles and explain that one is a job or task that a human will need to perform (ex., using a website, playing a game, driving a car, etc.).  The other represents human capabilities- how fast can we run? how much can we lift? how much information can we handle at once?  What if we’re under stress or time pressure? etc..  And then in a moment of pure enlightenment… I say “and Human Factors is the science of understanding human capabilities and designing jobs/tasks to ensure that they do not exceed human limitations.”

Alright, i’m still working on the last quote, but I really think the diagram helps.  It’s better than nothing.

you can’t argue with behavioral science

Amazing article on Cracked.com which covers 5 ways video games are DESIGNED to be addicting.  The amazing part?  It’s based on cognitive science (which is a major component of what I study in school).  Two of the experiments the article refers to I have covered in class.

First is variable ratio rewards… I learned in the context of how to get kids to do their homework, but of course it applies to every MMORPG or other game that offers random rewards for completing a task.  Basically, rather than give someone a reward every time they do something good, you give rewards on random occurrences of the good ‘thing.’  This way they don’t know when to expect it so they just keep doing it trying to get that reward.  If you gave them the reward every time they’d only do it when they wanted it.

Second, was the B.F. Skinner research (which actually covers 3 of the 5 things listed in the article). Skinner studied behavioral tendencies but his research’s application to gaming is summed up nicely by John Hopson who is a dev that worked on Halo and Age of Empires:

“Each contingency is an arrangement of time, activity, and reward, and there are an infinite number of ways these elements can be combined to produce the pattern of activity you want from your players.”

Emotiv EPOC, gaming for the true couch potatoe

Last December I finally received my long awaited Emotiv EPOC EEG headset.  However, after trying it on initially and quickly trying the demo content that I downloaded off their website I have had no significant timeframe to test out the device. <FAIL>

I’ve heard good things about it’s capabilities from a fellow researcher who has one and has spent a significant amount of time looking at EEG and other measurements for evaluating user engagement.  My own goal is to use the device in order measure a users excitement and/or frustration levels as they interact with (theoretically) any interface.  For the freelance work I do, this will let me add a quantitative and scientifically founded argument to design recommendations that would be qualitatively based using more traditional HF methodologies.

My goal is to spend a week over the summer ramping up on the device, getting a baseline for myself (testing it with video games), and then run a pilot study with 2-3 people to see how hard the process would be to incorporate in a traditional usability test.

Paintball Gun Fault Tree Analysis

Just a sample Fault Tree Analysis for a paintball gun that was created for a class project during the Spring 2009 semester at UB.  Project goal was to recommend safety improvements for paintball guns (and paintball speedball fields to a lesser extent) to the ASTM sub-committee on paintball.

Complete version of the fault tree available below (flash required).

Interesting results from mini-study on listening to music while exercising

Ran a quick study for a course to test my hypothesis that listening to music makes it easer to run.  I always think that when I listen to music I can run forever without tiring.  The results (although only n=4) were pretty interesting.  The experimental setup used a treadmill with runners wearing Bose on-ear noise-cancelling headphones that played either slow tempo (Zamfir’s “Lonely Shepard”), fast tempo (Hans Zimmers’ “Up is Down”), or no music running for 5 minutes at a pace of approximately 4mph.

The results (summary below) showed that running with slow music actually lead to an increased heart rate and (as hypothesized) running with fast music lead to a decreased heart rate.  Unfortunately the findings weren’t significant, however, i’d be curious to see if with a more precisely matched music tempo to running pace and increased sample size if the effects would be amplified.  Interestingly though, the second figure shows that perceived physical demand was greatest for fast music; so while it was physiologically easier to run with the fast music, individuals perceived it to be much harder than running without or with slower tempo music.