Basic Course Information
- Sarah Schachman: firstname.lastname@example.org
- Yujie Hong: email@example.com
- Ofra Amir: firstname.lastname@example.org
- Bernd Huber: email@example.com
- Winston Boucher: firstname.lastname@example.org
- Thursday @ 5pm: Maxwell Dworkin 123 (Yujie)
- Thursday @ 6pm: Maxwell Dworkin 123 (Winston)
- Friday @ 10am: Maxwell Dworkin 223 (Ofra)
- Friday @ 1pm: Maxwell Dworkin 123 (Bernd)
- Friday @ 2pm: Pierce 100F (Sarah)
Over the past few months, I have shared with you the foundations of design and human-computer interaction. In this class, I will share with you some of the most exciting things that are happening at the cutting edge. I will also take some time to talk about research more broadly and highlight the ways in which you can get involved (in any area of CS).
No prep required.
- Attention capture is involuntary. If a stimulus appears in your visual field, it will capture your attention regardless of whether you want it to or not.
- Two ways to capture attention
- Rapid luminescence captures attention - something turns abruptly bright
- Rapid appearance of a novel perceptual object captures attention - an object suddenly becomes visually distinct from its previous scene
- Example: like a tiger moving behind the reeds. Tiger becomes a novel perceptual object when it emerges from the background
Banner ads didn’t work (see lecture reading) because people can completely tune these ads out; if we know we’re not interested in these things, they no longer work.
- Visual “pop-out” occurs when a unique visual target (e.g. a feature singleton) is present among a set of homogeneous distractors (source: Hsieh et al., 2011)
- Example: Adaptive tool-bar highlights things that they used before
- Assuming the system is pretty good at predicting what you need next, this is a very effective feature
- The problem is that when things are highlighted to help people find things they had used before, the next time people looked for a tool they wanted to use they skipped over highlighted tools because before the tool they were looking for was not highlighted
Summary: Visual Pop-Outs only work if you know what to look for! If you are primed to look for it, you will find it, if not you will ignore it
- The Hick-Hyman law demonstrates that the time it takes to make a decision is proportional to the logarithm of the number of options
- Given n equally probable choices, the average time T required to choose among them is approximately T = a + b log2(n)
- Applies only to cognitively simple decisions
- Carries important implications for novel vs. familiar interfaces.
- Fitt’s Law declares that performance is logarithmic. It utilizes an UI element's ratio of distance (to it) to (its) width.
- A + B * log2(D/W)
- log2(D/W) is the index of difficulty in bits.
- 1/B is the index of performance in bits per second.
- Fitt’s Law can be used to calculate the difference in index of difficulty between Mac and PC menu designs. When using indirect pointing (a mouse), the barrier behind the Mac’s menu button turns its target size from 20 to infinity -- the W used in the equation is the button’s width, 60. The Mac’s index of difficulty is log(600/60) = log(10), compared to PC’s log(20). The Mac menu button has a lower index of difficulty.
A paper published in Nature in 1998 (http://bit.ly/1q8L83D) showed that neural control signals are corrupted by noise whose variance increases with the size of the control signal. Although that may seem like gibberish, our conversation in class boiled it down in a simple example: the greater the acceleration of your movement, the less accurate it will be. For example, pretend you have to extend your arm as if to throw a punch. If you do it slowly, you have exceptional control over this action. If done quickly, the accuracy lowers as the noise increases. This has to do with user mechanics because it helps the designer understand the nature of a user’s movements and thus shape the screen layout accordingly. We then spoke to the idea of planning for the optimal movement for the user and learned that the observed results come pretty near to what was predicted.