11 4a Human Intro-Sensory Dix1

8/6/2019 11 4a Human Intro-Sensory Dix1

http://slidepdf.com/reader/full/11-4a-human-intro-sensory-dix1 1/63

The Human

Introduction, Modeling, Vision

Human Computer Interaction, 2 nd Ed.

Dix, Finlay, Abowd, and BealeChapter 1



Introduction

� About HCI and humans «

� Interaction captured in figurebelow with ³system, task,user´ ± So far, have talked about

³interface design´ of system� Guidelines, principles,

evaluation, etc.

� Now will look at user

± Will consider ³model´ of human

± Models capture elements of whatever is being modeledthat are relevant to someendeavor

� E.g., ³cognitive engineering´Task

System

User ³feedback´and its representation

(possibly manipulable)



Introduction

� On describing humans ± To understate, a question of long standing «

� One of the big questions «

± Has and can be done in various ways and at different levels of analysis� Moral, spiritual, «., psychological, « physiological, « chemical

� Reductionist

� Psychology focuses on the individual

� Many orientations through psychology¶s short history ± Note: Clinical/personality orientations, e.g., Freud, Jung, have their own utility in

explanation, but are rarely subject to scientific investigation

± Structural, turn of 20th

century, relations among elements ± Gestalt, 1930¶s and on, general, especially of perception, still useful

± Behaviorist, all S->R, which is fine for some things, but fall short for explanationof mental phenomena

± Information processing/Cognitive, current orientation� Influences from Shannon¶s ideas on information, shortcomings of behaviorism,

successes in codifying information in computing

� ³Human information processor´



Modeling Humans

� Any theory or model is an abstraction

� For HCI, goals are primarily in ³Computer´ and ³Interaction´ ± Utility of human model lies in how well it helps with interfaces

� Card, Moran, and Newell (1983) M odel Human Processor ± ³Classic´ example of cognitive architecture with focus on humans interacting with

computers

± Perceptual system, motor system, cognitive system

± Each has own processor and memory

± Principles of operation dictate system behavior under

certain conditions

± A very simple model

� Dix et. al use similar information processing

division of elements in chapter



Overview

� Dix et al. model: ± Information i/o «

� visual, auditory, haptic, movement

± Information stored in memory

� sensory, short-term, long-term

± Information processed and applied� reasoning, problem solving, skill, error

� Idea tonight ± Give broad overview of elements of human

± Show that for some elements of user interface design, detailed knowledge is atleast useful and perhaps critical

� For design - sensory, perceptual, and cognitive guidelines andprinciples



Some Terminology

� Transduction of

energy, etc. by

sensory receptors ± Light to neural

impulses by retinalcells

Sensation Perception Cognition

� Latin ±� cognitio ± knowledge

� ³Higher level´ mental

representations and

procedures

� Process of thought

� Forming a ³mental

image´ or

awareness and

representation� To ³see a window´

� Top down andbottom up



Demo



Demo



Model Human Processor + Attention

� A ³useful big picture - Card et al. ¶83 plus attention ± Senses/inputp f(attention, processing)p motor/output

± Notion of ³processors´� Purely an engineering abstraction

� Detail next slide ± And in lectures!





Model Human Processor - Original

� Card et al. ¶83

� An architecturewithparameters for cognitiveengineering «

± Will see visual image store, etc.tonight

� Memory properties ± Decay time: how long memory lasts

± Size: number of things stored

± Encoding: type of things stored



Model Human Processor - Original



Dix: Information Input and Output

� Again, human consideredas information processor

� Input channels are thefive senses ± With some more important

than others

± Vision primarily

� Output channels arehuman effectors ± E.g., limbs, fingers, head,

vocal system



Vision

� Vision and visual perception studied across arange of disciplines

� Points tonight meant to highlight usefulness for HCI in knowing about vision

� For vision consider (Dix): ± Physical reception of stimulus

± Processing and interpretation of stimulus



The Eye - Physical Reception

� Mechanism for receiving light andtransforming it into nervetransmissions

± A transducer: light -> nerves

� Light reflects from objects ± Some strikes retina

± Images are focused upside-down onretina

� Ganglion cells (brain!) detect patternand movement

� As camera has ± equivalent of lens, aperture (pupil)

± and film (retina)

� Note that image is upside down onretina!

± « perception



Environment: Visible Light

� Generally, the body¶s sensorysystem is the way it is because ithad survival value ± Led to success

� survival and reproduction

� Focus on human vision (because of computer), but all share basicnotions

� Humans have receptors for (asmall part of) theelectromagnetic spectrum

± Have receptors sensitive to (firewhen excited by) energy 400-700nm

± Snakes ³see´ infrared, someinsects ultraviolet

� i.e., have receptors that fire

± What would life be like if humanscould see other parts of

electromagnetic spectrum???



Human Eye: Retinal Receptors

� Two types of (photo) receptors on retina: rods and cones ± Rods look like, well, rods «

± Will later look at color blindness « when cones fail

� Rods: ± spread all over the retinal surface (75 - 150 million)

± low resolution, no color vision, but very sensitive to low light (scotopic or dim-light vision)

� Cones: ± dense array around central portion of retina - fovea centralis (6 - 7 million)

± high-resolution, color vision, but require brighter light ( photopic or bright-light vision)



Human Eye: Lens

� Eye has compound lens:

± cornea (power) and lens (adjust focal length)f = focal length of lens

d = distance to object

r = distance to image that is formed

± Flexibility of lens changes with age, approaching 0 at 60 years



Human Eye: Depth of Field and Focus

� Depth of focus ±will see interesting affect «

± Distance over which objects are in focus without change in focus

� Note: Different colors have different depths of focus ± chromatic aberration

± Varies with size of pupil ± Range of focus:

Distance Near Far Depth of focus50 cm 43 cm 60 cm 17 cm

1 m 75 cm 1.5 m 75 cm

2 m 1.2m 6.0m 1.8 m

3 m 1.5m Infinity Large

± Rarely do computer systems model



Human Eye: Depth of Field - Example

� Photographic Images:

± Depth of field longer with small aperture(f stop)

± Range of focus:

Distance Near Far Depth of focus50 cm 43 cm 60 cm 17 cm

1 m 75 cm 1.5 m 75 cm

2 m 1.2m 6.0m 1.8 m

3 m 1.5m Infinity Large



Human Eye: Chromatic Aberration



Human Eye: Chromatic Aberration

� Different wavelengths of light focus

at different distances within eye

± Short-wavelength blue light refracted

more than long-wavelength red light

± Focusing on a red patch, an adjacent

blue patch will be significantly out of focus

� Strong illusory depth effects

�H

uman eye has no correction for chromatic aberration!

± Inadvisable: fine blue patterns in

visualizations!

± Visual effects in soap bubbles, crystal

sculptures, etc.



Using Physiology for Design

� Chromatic Aberration

� Different wavelengths focus differently

± Highly separated wavelengths (red & blue) can¶t be focused simultaneously

� Guideline: Don¶t use red-on-blue text

± It looks fuzzy and hurts to read



FYI: Human Eye: Receptors

� Lens focuses image onmosaic of photoreceptor cells lining retina

� From 1 ± 100 receptorsfeed into 1 ganglion cell

� Fovea

± Small area in center of retina densely packed withcones

± Vision sharpest

± ½o

- 2o

of arc

Receptor mosaic in fovea



Interpreting the Visual Signal

� Size and depth

± V isual angle indicates how much of view object occupies(relates to size and distance from eye)

� V isual acuity is ability to perceive detail (limited)

± familiar objects perceived as constant size(in spite of changes in visual angle when far away)

± cues like overlapping help perception of size and depth

� Brightness

± Human system not map directly to physical

� C olor

± Different sensory elements responsible for perception of color

± Implications for color use, e.g., color blindness



Human Eye: Visual Angle

� Visual angle - angle subtended by object at eye of viewer

± In degrees, minutes, seconds of arc

± Thumbnail at arm¶s length subtends ~1o

of visual angle

± 1cm (2/5´) object at 57 cm (20´), monitor distance, ~1o

of visual angle



FYI:

Human Eye: Acuities - Grating Acuity

� Acuities ± Measurements of abilities to

see detail

± Provide ultimate limits of information densities canperceive

� Resolve to about 1 min. of arc ± Roughly corresponds to with

receptor spacing in fovea

± E.g., to see 2 lines as distinctblank space between should lie onreceptor

± So, should be able to perceive

lines separated by twice receptor spacing

� Superacuities ± Resolution above what expected

by receptor density due tointegration of signals

GratingAcuity- 1-2 minutes of arc

- Ability to distinguish a

pattern of bright and dark

bars from a uniform gray

background



FYI:

Acuities: Point, Letter, Stereo, Vernier

� Point acuity ± 1 minute of arc

± Ability to resolve two distinct pointtargets

� Letter acuity

± 5 minute of arc ± Ability to resolve letters

± Snellen eye chart� 20/20 means a 5-minute letter target

can be seen 90% of time

� Stereo acuity

± 10 seconds of arc ± Ability to resolve objects in depth

± Measured as difference between 2angles (a and b) for a just-detectabledifference

� Vernier acuity

± 10 seconds of arc ± Ability to see if two lines are collinear



FYI:

Acuity Distribution and Visual Field� Again, receptors densely

packed at fovea

� Binocular overlap ± Region of visual field viewed by

both eyes

± Only here, stereopsis

� Visual acuity non-uniformlydistributed over visual field ± E.g., Can resolve only about 1/5

detail at 10o

� Next slide (tries) demonstrate³equi-resolvability´ of charactersas a function of distance fromfovea ± r = f(dist. fovea)

± (stare at center and see smaller characters at center as well or better than those at periphery)



Human Eye: Resolution



FYI:

Brightness, Luminance, Lightness� Brightness

± subjective reaction to levels of light

± affected by luminance of object

± measured by just noticeable difference

± Ecologically, need to be able to manipulate objects in environment

� Information about quantity of light, of relatively little use

± Rather, what need to know about its use

� Human visual system evolved to extract surface properties

± Loose information about quantity and quality of light ± E.g., experience colored objects, not color light

� Color constancy

± Similarly, overall reflectance of a surface

� Lightness constancy



FYI:

Luminance, Brightness, Lightness

� Luminance ± Amount of light (energy) coming from region of space,

� Measured as units energy / unit area

� E.g., foot-candles / square ft, candelas / square m

� Physical

� Brightness

± Perceived amount of light coming from a source ± Here, will refer to things perceived as self-luminous

� Lightness ± Perceived reflectance of a surface

± E.g., white surface is light, black surface is dark

� Again, ± Physical - Luminance

± Number of photons coming from a region of space

± Perceptual - Brightness ± Amount of light coming from a glowing source

� Lightness ± Reflectance of a surface, paint shade





FYI:

Brightness� Perceived amount of light coming from a glowing (self-luminous) object

± E.g., instruments

� Perceived brightness very non-linear function of the amount of light ± Shine a light of some intensity on a surface, and ask an observer, ³How bright?´

Intensity = How bright is the point?´

1 1

4 2

16 4

� Stevens power law ± Perceived sensation, S, is proportional to the stimulus intensity, I, raised to a power, n

± S = I n

± Here, Brightness = Luminancen

± With n = 0.333 for patches of light, 0.5 for points ± Applies only to lights in relative isolation in dark, so application more complicated

� Applies to many other perceptual channels ± Loudness (dB), smell, taste, heaviness, force, friction, touch, etc.

� Enables high sensitivity at low levels without saturation at high levels

� Just-noticeable difference depends on value



Dix: Interpreting the Signal - Color

� Color ± made up of hue, intensity, saturation

± cones sensitive to colour wavelengths ± blue acuity is lowest

± ~8% males and ~1% females color blind



Trichromacy Theory

� Recall, that there are 2 types

of retinal receptors:

± Rods, low light, monochrome

� So overstimulated at all but low

levels contribute little

� So only consider cones for color

vision

± Cones, high light, color

± Not evenly distributed on retina

Distribution of receptors across the retina, left eyeshown; the cones are concentrated in the fovea,

which is ringed by a dense concentration of rods

http://www.handprint.com/HP/WCL/color1.html#oppmodel



Trichromacy Theory

� Cones (3 types) differentially sensitive towavelengths ± ³trichromacy´

� Each type cone has different peak sensitivity: ± S: 450 nm ³blue´

± M: 540 nm ³green ± L: 580 nm ³red´

± More later

± No accident 3 colors in monitor

� Color space: ± An arrangement of colors in a 3-dimensional

space� Monitor: R,G,B� Primary paint colors: R,Y,B� Printer: cyan, magenta, yellow

± There are many, each designed for differentpurposes

± Will consider several

± Can match all colors perceived with 3 colors� Does not matter that spectral composition of that

patch of light may be completely different

± But, chickens have 12 « ± Different gamut, more later

Cone sensitivity functions

Cone response space, defined by

response of each of the three cone

types. Becomes 2d with color deficiency



FYI:

Color Space Ex.: RGB Color Cube

� Again, can specify color with 3 ± Will see other way

� RGB Color Cube ± Neutral Gradient Line

± Edge of Saturated Hues

± ppt example

http://graphics.csail.mit.edu/classes/6.837/F01/Lecture02/

http://www.photo.net/photo/edscott/vis00020.htm



Cone Sensitivity Functions

� Cone receptors least sensitive to(least output for) to blue

Relative sensitivity curves for the three types of

cones, log vertical scale, cone spectral curves from

Vos & Walraven, 1974

Relative sensitivity curves for the three

types of cones, the Vos & Walraven

curves on a normal vertical scale



Color Blindness

� ~9% male, and ~1% females havesome form of color visiondeficiency!

� Most common:

± Lack of long wave length sensitivereceptors (red, protanopia)� See figure at right bottome

± Lack of mid wave length receptors(green, deuteranopia)

� Results in inability to distinguish redand green

� E.g., cherries in 1st

figure hard to see

� Trichromatic vs. dichromaticvision







FYI: Perceived Color

� Color perceived relative to context ± Are the ³X´s in the figure below the same color?

± Easy implications for use in maps

� Contrast illusion

± An illusion is an extreme case

� Somewhat ³surprising´ because it leads to error

± Appears to be different color X!



FYI: Perceived Color

� With color of x touching «



FYI: Afterimage

� Occurs due to bleaching of photopigments

± (demo next slide)

� Implications for misperceiving(especially contiguous colors ±and black and white)

± ³I thought I saw «´

� To illustrate:

± Stare at + sign on left� May see colors around circle

± Move gaze to right

± See yellow and desaturatedred



Afterimage Example



Dix: Interpreting the signal (cont)

� The visual system compensates for:

± movement

± changes in luminance.

� Context is used to resolve ambiguity

� Optical illusions sometimes occur due to

over compensation



FYI: Simultaneous Brightness Contrast

� Gray patch on a dark background looks lighter than the same patchon a light background ± Predicted by DOG model of concentric opponent receptive fields



FYI: Mach Bands

� At pointwhere uniform area meets a luminance ramp, brightband is perceived ± Said another way, appear where abrupt change in first derivative of

brightness profile

± Simulated by DOG model

± Particularly a problem for uniformly shaded polygons in computer graphics� Hence, various methods of smoothing are applied

Ernst Mach



Psychology and Design of

Interactive Systems� Some direct applications

± e.g. blue acuity is poor blue should not be used for important detail

� However, correct application generally requiresunderstanding of context in psychology, and anunderstanding of particular experimental conditions

� A lot of knowledge has been distilled in ± guidelines

± cognitive models

± experimental and analytic evaluation techniques



End ?



Optical Illusions

the Ponzo illusion the Mueller-Lyer illusion



Optical Illusions



Preattentive Processing and Illusions

� What is wrong with this

triangle?

± Impossible (or at least

difficult) to build

� Cues for perception

misleading

± Must rely on conscious

(rational) processesintelligence to figure it out,

± Conscious/rational

processes much slower



The Other Senses «

� But not smell and taste ± Long latency, practical challenges

� Hearing

� Touch

� Kinesthetic ± Movement ± Fitt¶s law



Hearing

� Provides information about environment:distances, directions, objects etc.

� Physical apparatus:

± outer ear ± protects inner and amplifies sound ± middle ear ± transmits sound waves asvibrations to inner ear

± inner ear ± chemical transmitters are releasedand cause impulses in auditory nerve

� Sound ± pitch ± sound frequency

± loudness ±amplitude

± timbre ± type or quality



Hearing (cont)

� Humans can hear frequencies from 20Hz to15kHz ± less accurate distinguishing high frequencies than

low.

� Auditory system filters sounds ± can attend to sounds over background noise.

± for example, the cocktail party phenomenon.



Touch

� Provides important feedback about environment

� May be key sense for someone who is visually impaired

� Stimulus received via receptors in the skin

± thermoreceptors ± heat and cold

± nociceptors ± pain

± mechanoreceptors ± pressure

(some instant, some continuous)

� Some areas more sensitive than others, e.g., fingers

� Kinethesis - awareness of body position

± affects comfort and performance.



Movement

� Time taken to respond to stimulus:reaction time + movement time

� Movement time dependent on age, fitness etc.

� Reaction time - dependent on stimulus type: ± visual ~ 200ms

± auditory ~ 150 ms

± pain ~ 700ms

� Increasing reaction time decreases accuracy in theunskilled operator but not in the skilled operator.



Movement ± Fitt¶s Law

� Fitts' Law, 1954 ±Fundamental law of human sensory-motor system

±Practical application in interface design

±Describes the time taken to hit a screen target

±Time, Mt, to move hand to a target of size, S, at distance, D, away

Mt = a + b log2(D/S + 1)

where: a and b are empirically determined constants

Mt is movement time

D is Distance

S is Size of target

± ³index of difficulty´: log2(D/S + 1)� Same performance at greater distance with greater size

targets as large as possible, distances as small as possible



Fitt¶s Law ± Example Implication

� Hierarchical menus are hard to hit ± Especially when it takes two actions «

P h l d D i f



Psychology and Design of

Interactive Systems� Some direct applications

± e.g. blue acuity is poor blue should not be used for important detail

� However, correct application generally requiresunderstanding of context in psychology, and anunderstanding of particular experimental conditions

� A lot of knowledge has been distilled in ± guidelines

± cognitive models

± experimental and analytic evaluation techniques



End

Documents

11 4a Human Intro-Sensory Dix1