Sasha Homework Models Homework

From Cmsc434_f09
Jump to: navigation, search

Sasha Slutsker

1

  • Perceive the first symbol (tp)
  • Here, the user must perceive the first symbol.
  • Perceive the second symbol (tp)
  • Here, we assume that the user will independently perceive the second symbol, being otherwise unfamiliar with it.
  • Match the first and second symbol (tc)
  • It takes one cognitive step to match the first and second symbol. Assuming they match, he will need to proceed with the additional steps. Note that I am not including a "recognize" step since he does not need to recognize the symbols, merely say if they look the same.
  • Perceive the third symbol (tp)
  • Match the second and third (tc)
  • Initiate response (tc)
  • Click button (tm)
  • Clicking the button is a separate motor action once done matching.
  • 3 * tp + 3 * tc + tm = 580ms

2

  • In order to perceive an echo, we must be at a distance from the wall that is great enough so as to ensure that we fully perceive the first sound before the second sound reaches us
  • Perceiving takes 100ms or .1s (based on the HIP model)
  • This means that the sound must travel to the wall and back in at least .1 s
  • ((330 m/s * .1s) / 2 = 16.5 meters away (the last division by two is because the sound must go both ways - rest establishes that the sound can travel 33 meters in .1 seconds)

3a

  • 1.35 seconds = Mentally prepare on 27
  • 1.1 seconds = Pointing at 2
  • .2 seconds = Pressing 2
  • 1.1 seconds = Pointing at 7
  • .2 seconds = Pressing 7
  • 1.35 seconds = Mentally prepare on +
  • 1.1 seconds = Pointing at +
  • .2 seconds = Pressing +
  • 1.35 seconds = Mentally prepare on 5 = (I think the "=" operator is fully anticipated)
  • 1.1 seconds = Pointing at 5
  • .2 seconds = Pressing 5
  • 1.1 seconds = Pointing at =
  • .2 seconds = Pressing =
  • 1.35 seconds = Mentally prepare on *
  • 1.1 seconds = Pointing at *
  • .2 seconds = Pressing *
  • 1.35 seconds = Mentally prepare on 2 =
  • 1.1 seconds = Pointing at 2
  • .2 seconds = Pressing 2
  • 1.1 seconds = Pointing at =
  • .2 seconds = Pressing =
  • 1.35 seconds = Mentally prepare on SQRT
  • 1.1 seconds = Pointing at SQRT
  • .2 seconds = Pressing SQRT
  • TOTAL: 1.35 * 6 + .2 * 9 + 1.1 * 9 = 19.8 seconds

3b

  • Width of each button: 32 pixels
  • Height of each button: 26 pixels
  • Height of = sign: 65 pixels
  • Distance from center of button to edge of another button (moving right or left): 24 pixels
  • Distance from center of button to edge of another button (moving up or down): 19 pixels
  • Distance from 2 to 7: width is 24 pixels, height is (19 + 26 + 6) = 51 pixels, making the hypotenuse = 56 pixels
  • Distance from 7 to =: 65 pixels right, 103 pixels down, making it 121.8 pixels
  • Distance from + to 5: 78 pixels left, 64 pixels down, making it 100.9 pixels
  • Distance from 5 to =: 65 pixels right, 19 pixels down, making it 67.72 pixels
  • Distance from = to *: 35 pixels up, 24 pixels left, making it 42.44 pixels
  • Distance from * to 2: 62 pixels left, 19 pixels down, making it 64.85 pixels
  • Distance from 2 to =: 100 pixels right
  • Distance from = to *: 105 pixels up
  • 2 to 7: .1 * (log(56/26 + .5)/log(2)) = .1408 seconds
  • 7 to =: .1 * (lg(121.8/32 + .5)) = .211 seconds
  • + to 5: .1 * (lg(100.9/26 + .5)) = .213 seconds
  • 5 to =: .1 * (lg(67.72/32 + .5)) = .138 seconds
  • = to *: .1 * (lg(42.44/26 + .5)) = .109 seconds
  • * to 2: .1 * (lg(64.85/32 + .5)) = .134 seconds
  • 2 to =: .1 * (lg(100/32 + .5)) = .186 seconds
  • = to SQRT: .1 * (lg(105/26 + .5)) = .218 seconds
  • Total: 1.35 seconds

Comparing 3a and 3b

The Fitt's analysis showed a much faster time than the KLM analysis did. I think this was partly due to the fact that the k was so low (it's hard to imagine someone can actually click the buttons that fast.) However, KLM analysis was probably not too accurate either, as it probably has too many "mentally prepares" that inflated the final answer. Also, the Fitt's analysis only deals with the time to actually click the button, and not the time to think about where the button is and what must be done, etc. This means that even if they had been similar in how long it took to click the buttons (and they're not - our Fitt's analysis is much faster even for just the mechanical tasks), the KLM analysis would still likely report a slower time for taking into account the cognitive steps.

I did some tests and I usually completed the task in about 8-12 seconds, and that seems more accurate (but of course I could not help being influenced by the task, so my results should be taken with a grain of salt.)