1. Sensing information
ISE

2. Sensory register, part I: the visual sensory system (see pp
Sensory register, part I: the visual sensory system (see pp of Buck & Lehto)
How we see: "Light (electromagnetic radiation) is collected by the cornea and passes through the anterior chamber, then through the pupil, which is an aperture in the iris diaphragm, to the crystalline lens. The lens focuses the light by changing shape in a process called accommodation. The image passes through the vitreous humor in the posterior chamber and is projected onto the retina. The retina contains two kinds of photo receptors: rods, which are sensitive to dim light, and cones, which are used in bright light and convey color information. The central region of the retina, which provides the best visual acuity, is the fovea, which contains only cones. Information from the photoreceptors passes through the bipolar cells and ganglion cells (cells in the retina) before passing out of the eye via the optic nerve."[1]
[1] Coren, S., Ward, L., & Enns, J. (1999). Sensation and perception (5th ed.). Fort Worth :Harcourt Brace, p. 83.
ISE

3. Example: driving at night
Specific hazards caused by:
Glare
Reduced contrast sensitivity
Loss of color vision
Particularly bad for older drivers due to:
Loss of contrast sensitivity due to age
Loss of accomodation
Glare is irrelevant light of high intensity. It has the effect of destroying the rod's sensitivity to low spatial frequencies.
ISE

4. What affects visibility?
Of print…
Good
Bad
BAD
item 1: too small (font size)
item 2: too light (contrast)
item 3: difficult fonts (e.g., spatial frequency)
item 4: polarity (light on dark) – what is “good” depends on environment, lighting, etc.
Bad
Bad
ISE

5. Characterizing and measuring light
Color
hue
pure wavelength
visible spectrum ~400 – 700 nm
saturation
amount of achromatic light mixed in
brightness
amplitude
Design considerations
use color as a secondary source of information
design for monochrome first
consider
simultaneous contrast
negative afterimage
. Know how to use color in displays. It should be used as a secondary source of information. Design for monochrome first. Simultaneous contrast and negative afterimage affect our perception of color.
ISE

6. Characterizing and measuring light (cont.)
Brightness measures
luminous intensity, luminous flux
energy at the source, candela
illuminance
amount of energy striking an object, lux or foot-candles (fc)
luminance
amount of energy reflected from an object, foot-lambert (FL)
reflectance
ratio of the amount of light striking the object to the light reflected from the object,
illuminance
luminance
ISE

7. Characteristics of visual displays
Visual angle,
VA = tan-1(H/D)
VA = 3438H/D min
Example: reading ‘Good’ text from your notes page,
H = ______
D = ______
VA = ___________________________
For example, H = 0.25 in
D = 18 in
VA = tan-1(H/D) = tan-1(0.25/18) = ° = min.
ISE

8. Characteristics of visual displays (cont.)
Contrast,
Luminance of light areas, LL
Luminance of dark areas, LD
Contrast =
Spatial frequency
cycles of light and dark per degree of VA
Polarity
dark on light vs light on dark
(LL – LD)
(LL + LD)
ISE

9. What else affects visibility?
Characteristics of observers
Visual Acuity
Age
Contrast sensitivity
Night vision
Characteristics of environment
Ambient light
e.g., daylight vs night, glare, etc.
Movement
Distractions
ISE

10. Your turn …
Define system requirements for a new information kiosk for the engineering school based on this understanding of the visual sensory system.
Typical statements could include:
The system must accommodate ____________________.
The system should allow for _______________________.
The system should include mechanisms that will _____________________.
Be careful not to start designing yet!
ISE

11. Signal Detection Theory (SDT) (see pp. 582-586 of Buck & Lehto)
A situation is described in terms of two states of the world:
a signal is present ("Signal")
a signal is absent ("Noise")
You have two possible responses:
the signal is present ("Yes")
the signal is absent ("No")
ISE

12. If you decide "Yes" and the true state of the world is "Signal," that's called a Hit.
If you decide "Yes" and the true state of the world is a "Noise," that's called a False Alarm.
If you decide "No" and the true state of the world is "Signal," that's called a Miss.
If you decide "No" and the true state of the world is "Noise," that's called a Correct Rejection.
ISE

13. The theory assumes that what you are doing is:
First, you collect sensory evidence concerning the presence or absence of the signal.
Next, you decide whether this evidence constitutes a signal. This means that you must have some criterion C that you use as a "cutoff": if the evidence is less than C, you decide "No"; if the evidence exceeds C, you decide "Yes".
ISE

14. Measures of Performance in SDT: 1. Sensitivity (d’)
A function of the keenness or sensitivity of the human's detection mechanisms and the relative strength of the signal in noise.
This value may be calculated from the probabilities of a hit and a false alarm.
Example: Radiologists evaluating a candidate MRI machine are given 1000 sample slides, 500 of which are from patients with a tumor. The results of the readings are as follows:
Tumor (“signal”)
No tumor (“noise”)
Positive (“yes”)
400
100
Negative (“no”)
ISE

15. Calculating d’ P(FA) = __________ zFA = ____________
P(M) = ___________ zM = _____________
d’ = _______________
P(FA) = __0.2________ zFA = ______0.8415______ (table D.1A from B&L)
P(M) = ___0.2________ zM = _____ ________
d’ = = 1.683
ISE

16. Measures of performance in SDT: 2. Response bias (b)
Another way to describe performance is in terms of response bias: you may be prone to say "yes" (which is "risky") or you may be prone to say "no" (which is "conservative").
Response bias the ratio of the heights of the two curves at the cutoff point and is measured by the quantity:
where Xi = "evidence variable”
S = signal
N = noise
ISE

17. Going back to our example …
Tumor (“signal”)
No tumor (“noise”)
Positive (“yes”)
400
100
Negative (“no”)
P(FA) = __________ OrdFA = __________
P(M) = ___________ OrdM = __________
β = _______________
P(FA) = __0.2_______ OrdFA = ____0.28______ from table D.1B
P(M) = ___0.2________ OrdM = _____0.28_____
β = __0.28/0.28 = 1.0 _____________
ISE

18. radiologists reading x-rays for signs of tumors
Studies of human performance show that humans do change beta in response to changes in probabilities and payoffs ‑‑ but not as much as they should! This phenomenon is called sluggish beta.
Note: the terms “risky” and “conservative”refer only to a person’s propensity to say “yes (signal)” or “no (noise).”
Examples:
radiologists reading x-rays for signs of tumors
radar operators on a battle ship looking for incoming enemy aircraft
scanning a parking lot for a parking space
ISE

19. What is an “optimal setting” for ß?
The cutoff (C) for determining the presence of a signal is not the same as the response bias parameter (ß). However, they are correlated. For example, if your strategy becomes more "risky" (in other words, your ß goes down), then your cutoff also goes down. As you get more conservative, both your cutoff and your ß go up.
What is an “optimal setting” for ß?
Note that as terms in the denominator increase, ß decreases and thus response becomes riskier. As terms in the numerator increase, ß increases and thus responses become more conservative. Also, the first fraction shows the effect of signal likelihood and the second fraction shows the effect of payoffs.
ISE

20. The receiver Operating Characteristic (ROC) curve plots the probability of a hit against the probability of a false alarm. Each curve represents the same sensitivity at different levels of response bias.
ISE