Lecture 8. Human Information Processing (1) CENG 412-Human Factors in Engineering May

Transcription

1 Lecture 8. Human Information Processing (1) CENG 412-Human Factors in Engineering May

2 Outline Visual Sensory systems Reading Wickens pp

3 Today s story: Textbook page 61. List the vision-related problems that our traveler encounters: 3

4 The stimulus: Light All visual stimuli that the human can perceive may be described as a wave of electromagnetic energy. The visible spectrum is about nanometers (wavelength). 4

5 How sensation relates to perception Sensation ( physical entity) wavelength amplitude (intensity) purity of wavelength Perception (human experience) hue brightness saturation Adapted from Rebecca Boren, Introduction to Human Factors in Engineering 5

6 How do we see? The key building blocks in the human eye: Lenses: focus the light emitted by objects in the world onto the retina at the back of the eye Retina: contains light-sensitive cells which convert light to electrical activity A network of neurons which collects visual information and transmits it down the optic nerve to the brain 6

7 Accommodation Light is focused by two lenses: a fixed lens and a flexible lens. The shape of the flexible lens is changed by the donut-shaped ciliary muscles. Which muscle state is used for focusing a distant/close object onto the retina? 7

8 Human ability to clearly focus on an image Depends on two factors: Shape of the eyeball Shape of the lens Near-sighted need concave lens Far-sighted need convex lens 8

9 The Iris Primary role: controls the amount of light entering the eye through the pupil (aperture) bright light: iris constricts low light: iris relaxes 9

10 The Iris (cont d) Can also improve the focus of the image on the retina Why? The simplest model of the eye is a pinhole camera. 10

11 11

12 Retina What do cells in the retina do? 1. the input layer: light sensitive receptors: rods (R) and cones (C) 2. Ganglion (G) cells: the only output from the eye 3. Bipolar (B) cells connect the receptors to the ganglion cells 4. Horizontal (H) cells converge signals from several cones. They determine how many receptors each ganglion cell sees 5. Amacrine (A) cells converge signals from peripheral rods. 12

13 What is the effect of light on rods and cones? Light hyperpolarizes these cells (i.e. the voltage inside drops). Darkness depolarizes them (i.e. the voltage inside rises). 13

14 HF issue Why is reading difficult in low illumination? 1. The retina is not uniform Peripheral retina contains primarily rods Fovea (center of the eye) contains only cones 2. Rods and cones are not equally sensitive to low light levels. Cones are less sensitive to light. 3. The periphery has poor acuity 14

15 Why the rod system has poor acuity 1. Ganglion cells integrate information from a large area of retina (3 deg) 2. Large spacing and large convergence results in low acuity 1. Ganglion cells integrate information from a small area of retina (.03 deg) 2. Small spacing and low convergence results in high acuity. 15

16 What the eye sees Daylight Nightlight 16

17 What is the advantage of colour vision? 17

18 Colour: an intrinsic feature? We usually think of colour as an intrinsic property of objects. Yet, colour is associated with the wavelength of light. The illuminating energy interacts with the molecules of the object surface. As a result of this interaction, light energy (radiance) is emitted from the surface. 18

19 Colour perception The perception of an object s color depends on 3 factors: the spectrum of energy in various wavelengths illuminating the object surface the spectral reflectance of the object s surface the spectral sensitivity of the sensor irradiated by the light energy from the object s surface. 19

20 Colour perception: examples A blue object has a surface material that appears blue when illuminated with white light. Why? The same object will appear violet if illuminated only with red light. A blue car under intense sunlight (white) will become hot and radiate energy in the IR spectrum. 20

21 Colour perception: examples (2) (Left) a warm light source: it enhances reds and oranges while dulling blues and greens; (Middle) a neutral light source; (Right) a cool source: it enhances blues and greens while dulling reds and oranges. 21

22 Cones - Cones respond better to a particular wavelength of light than to others - blue cone is mildly sensitive to blue light nm. - green cone is very sensitive to green light, but also sensitive to blue and red. - red cone 22

23 Why do we need three types of cones? What are the problems of a one cone system? real world green cones see red cones see 23

24 The three-cone theory suppose that light in the yellow range of wavelengths strikes the retina. it activates both the green and the red cones of the retina. electrical messages are sent by both the red and the green cones to the brain. The brain recognizes that the light has activated both the red and the green cones and somehow interprets this to mean that the object is yellow. In this sense, the yellow appearance of objects is simply the result of yellow light and stimulating the red and the green cones simultaneously. 24

25 The three cone theory (cont d) 25

26 Colour blindness Do you see a pattern in this image? 26

27 Various types of colour blindness Missing one cone type Missing two cone types Missing all three cone types (vision limited to rods) 27

28 original image Simulated image of vision in a person with deuteranopia (no green sensitive cone). This person is unable to distinguish between red and green. Simulated image of vision in a person with triteranopia (no blue sensitive cone). This person is unable to distinguish between blue and yellow. Simulated image of vision in a person with protanopia (no red sensitive cone). This person is unable to distinguish between red and green. from 28

29 Color Vision Deficiency What Colorblind People See Normal Vision Color-Deficient 29

30 Color Vision Deficiency What Colorblind People See Normal Vision Color-Deficient 30

31 Maps Color Vision Deficiency When to be Concerned with Color Graphs & Charts Color-coded navigation Using color to highlight text (especially red) 31

32 Color Vision Deficiency What You Can Do Simplify or eliminate any color coding as much as possible. The fewer colors, the better. Use label text along with color. Include a key or legend. Don t simply refer to colors by name. Avoid purples when there is meaning to the color. Substitute orange for red when highlighting text. If you can, get someone who is colorblind to review your text and graphics in context. Use a filter such as vischeck ( to preview your web pages and files. 32

33 Color Vision Deficiency What You Can Do Avoid the following combinations of colors, especially when it is important that they be easily distinguished: Reds and Browns Bright Greens and Yellows Purples and Blues or Pinks Dark Greens and Browns Light Grays and Light Pinks Reds and Greens 33

34 Depth perception -useful for navigation and interaction with the real 3D world -Depth cues: -Accommodation (speed of the process) -Binocular disparity -Textural gradient -Convergence/perspective -Occlusion 34

35 Binocular (retinal) disparity results from eyes being in slightly different locations; each eye builds a slightly different image of the world. 35

36 Depth: Textural Gradient Surfaces appear to have a finer texture as they recede into the distance. from Vaughan Bell, Lecture notes on perception and perceptual distortions, Oct

37 Depth: Convergence / Perspective Perspective projection: parallel lines meet 37

38 Depth: Occlusion Objects located closer to the sensor partially occlude more distant objects. 38

39 Bottom up versus top down processing A tentative flat box of vision -Signals from the eyes and other senses are bottom-up -Both top-down and sideways are knowledge; - top-down: specific (such as faces being convex), - sideways: general rules applied to all objects and scenes (perceptual grouping and perspective). -Perceptual learning: feedback from experience 39

40 When top down and bottom up information collide (a) and (b) (black hat) show the front and side truly convex view; (d) (white hat) shows the inside of the mask; it appears convex although it is truly hollow; (c) is confusing as part of the hollow inside is seen as convex, combined with the truly convex face. Top-down knowledge of faces is pitted against bottom-up visual information. 40

41 Human Factor Issues: Visual search and detection A critical aspect of human performance: the process of visual search as it is linked with object/event detection. Today we ll talk about visual search only. The task: try to detect a target among a field of distracters Eye movements: we move our eyes to see better. 41

42 Eye movements Saccades Abrupt, discrete movements from one location to the next saccades rotate both eyes so that the current location now falls on the fovea. Ex: in reading, saccades are used to point the fovea at each word in this sentence Each saccadic movement can be characterized by initiation latency, a destination, a movement time (or speed), a dwell duration, and a useful field of view Perception occurs during dwells (94% of total duration of the saccade). The dwell duration depends on: Information content (i.e. longer words require longer dwells) The ease of information extraction (i.e. blurred target takes longer to interpret) -tracking/demo 42

43 Eye movements (cont d) Pursuit Constant velocity designed to track moving targets (smooth pursuit) When an object that we are looking at moves, its image is kept still on the retina by means of a pursuit eye movement (e.g. tracking a ball or your moving finger). 43

44 The serial search model In the search process, we distinguish between targets and nontargets (distracters or noise). Many searches are serial: each item is inspected in turn The predicted time to detect a target can be calculated by multiplying the number of items to be searched by the time to search each item. Divide by 2. On average, the target will be encountered after half the targets have been inspected. This is a useful model for predicting search times with computer menus or listings in phone books. In more critical situations, it is useful to predict how long it will take a driver to find a sign by taking his eyes off the highway and searching. 44

45 Factors affecting visual search Conspicuity : how much the target looks like non-targets related to bottom-up processing The search for conspicuous targets is parallel and thus a lot faster Ex: flashing warning signal, highlighted item on a check list etc. 45

46 Factors affecting visual search (cont d) Expectancies Top-down implications of searcher expectancies of where the target might be likely to lie. Based upon a priori knowledge from our experiences, we expect road signs to be on the right side Our familiarity with the alphabet helps us search phone book listings 46

47 You know now: how the human visual system works Limits of the visual system influence the nature of the visual information that enters the brain Some aspects of interpretation Top-down influences: expectancy, learning 47