Human Factors and

Ergonomics
Information Processing
Feb 4, 2019
 Objectives of This Course

• Learn the basic concepts of human factors engineering.

• Learn the importance of considering human capabilities

and limitations in the design of systems.

• Develop skills to apply human factors principles to the

analysis, design, and evaluation of systems.

• Develop clear and cohesive communication

through analysis of needs and constraints of
contexts, construction of supported arguments,
organization at all levels and effective use of multiple
media.
 What is Human Factors?

• Definition: The engineering discipline that deals with all

aspects of human characteristics in the design of things
people use in their work and everyday lives.

• Mission: Improve the interaction between people and

technology
– design systems that take advantage of people’s
capabilities and compensate for their limitations
 What is technology?

• Any tool – physical, virtual, conceptual, or

cultural – that helps people make
decisions, act, or achieve their goals
• Two types of technology:
– Hard: wheels, power plants, electronics,
snowboards
– Soft: work schedules, teams, policies &
procedures
 Who are ‘people’?
• A surprisingly difficult question
• Stakeholders:
– Designers
– Users
– Owners
– Third parties
• Demographics:
– Age
– Culture
– Education & expertise
• Discussion: Concept of a course as
technology
 Main Fields of HF Research
• Physical - human anatomy, and some of the anthropometric, physiological
and bio mechanical characteristics as they relate to physical activity
– Physical workload, manual material handling, musculoskeletal disorders, product design, etc.
• Cognitive -mental processes, such as perception, memory, reasoning, and
motor response, as they affect interactions among humans and other
elements of a system
– mental workload, decision-making, skilled performance, human-computer interaction, human
reliability, work stress and training as these may relate to human-system and Human-Computer
Interaction design.
• Organizational - optimization of socio-technical systems, including their
organizational structures, policies, and processes
– Relevant topics include communication, crew resource management, work design, work systems,
design of working times, teamwork, participatory design, community ergonomics, cooperative
work, new work programs, virtual organizations, telework, and quality management.
• Environmental is concerned with human interaction with the environment
– The physical environment is characterized by climate, temperature, pressure, vibration, light
What knowledge/skills are required to
practice HF engineering?

• People
– How they take up space, move, think,
learn, remember, pay attention, work in
teams, etc.
– How they interact with technology
• Design
– How environments, tasks, equipment,
procedures, and training come to be
• Domains
– Varieties of engineering and industries
 The Systems Perspective

• System: An entity composed of humans, machines,

and other things (natural or technological) that
interact to accomplish a purpose that could not be
met independently
– Many components are interconnected
– People adapt to changes
– Environment guides behaviour

• HF is inherently multi-disciplinary
Human Factors is multidisciplinary
• Chapter 6: Information Processing
 Learning Objective

• Learn about the basics of cognitive

psychology
• Form a USEFUL model of cognition
• Apply that model to inform DESIGN
decisions
Wickens et al Model of Human Information
Processing (Fig. 6.2)
https://www.youtube.com/watch?v=pMMRE4
Q2FGk
 Cognition

• Perception
• Working memory
• Long-term memory
• Attention and mental resources
 Perception
Compares incoming sensory
data to stored memory:

• feature analysis
• unitization/automaticity
• top down processing/ability
of familiarity to offset
degraded bottom-up
processing
 Sensory systems

• Visual
• Auditory
• Somatosensory (touch, proprioception: sense of
relative position of body parts)
• Gustatory (taste)
• Olfactory (smell)
• Vestibular (balance/coordinating movement with
balance)
 Visual sensory system

• Perceptually relevant characteristics of light

• The receptor system: The eyeball
• Basic sensory limits of the receptor cells (rods and cones)
• Perceptual processing
– Depth perception
– Visual search
– Detection
– Discrimination
– Absolute judgment
• Bottom-up and top-down processes combine to guide visual
perception
 The Visual Environment: Perceptually
relevant characteristics of light

• Humans sensitive to wavelengths between

400 and 700 nanometers
• Hue=wavelength (composed of Red, Green,
Blue)
• Brightness=amplitude (number of photons)
• Saturation=purity of color, undiluted by
achromatic gray or white light

https://en.wikipedia.org/wiki/Color
Wavelength
 Light in the world

• Illuminance-light hitting the surface (lux)

– Declines with square of distance
• Luminance-light reflecting (leaves “il” on
surface)
• Reflectance-how much light sent back
=luminance/Illuminance

• Eye detects 10-6 cd/m2, injured by106

cd/m2, direct viewing of the sun 109 cd/m2
, bright day 105 cd/m2
 Bringing the image into focus

• Cornea bends the light

• Lens does finer adjustment
(accommodation – lens changes its shape)
• Controlled through the ciliary muscles
• Pupil: adjusts the amount of light
• Vergence: The degree of rotation of the eyes
inward or outward
Photoreceptor cells (rods,
cones) and luminance

Photopic

Mesopic

Scotopic
Distribution of rods and cones
 Sensitivity to light

– convergence (funneling) of
the signals into the bipolar
layer
– convergence 120:1 for rods
• specialized to amplify
weak luminous signals
– convergence 6:1 for cones
• less convergence
means more ability to
resolve detail
• foveal cones almost no
convergence
https://en.wikipedia.org/wiki/Photoreceptor_ce
ll#/media/File:1414_Rods_and_Cones.jpg
Visual acuity and distance from fovea
 Visual angle

• VA = arctan (Height/Distance)
– For small angles ~ H/D

<θ
Area of fovea shown by dot in
center
Implications of eyeball characteristics

• Foveal vision of 2 degrees (cones) surrounded by

peripheral vision (rods)
• Fovea (cones) sensitive to detail, greater acuity/
Periphery (rods) relatively more sensitive to motion
• Rods sensitive to light provide night vision, but lose
sensitivity when exposed to bright light
• Cones sensitive to color
• Differential sensitivity to red light makes it possible
to engage cones and preserve sensitivity of rods
 Basic sensory limits

• Contrast ratio (ratio of light to dark) and

spatial frequency (level of detail)
determine visibility
• Color deficiency (red==green) affects
7% of males therefore design for
monochrome
• Sensitivity of rods to peripheral motion
and cones’ reduced sensitivity to detail
in low light at night
Application of sensory limits to design

• Contrast ratio and text conspicuity (not gray on black for

warnings)
• Optimal size/level of detail 3 cycles/degree (11 point font)
I can not see!
• Use of mixed case in print conveys information through
word shape
• Lower Icase
can see!
letters have more variety and are easier to
read THAN IF THE SENTENCE WAS ALL
UPPERCASE.
• Avoid excessive detail in icons, traffic signs

30
Perceptual processing: Bottom up/Top
down

• Bottom up: Interpretation of stimulus

driven by data in the world
• Top down: Interpretation of stimulus
driven by knowledge in the head
Bottom-up and Top-down processing
 Perception: Feature Analysis

• Role: bottom-up process of

extracting information from sensory
stimuli
• Process (3 stages):
– Break stimulus into features (geons)
– Map features onto (i.e., compared with)
stored patterns (Long-term memory)
– Select the best match
 Perception: Feature analysis

• Design implications (text and objects):

– Gaps between strings enhance perception
(thismeanstoleavespacebetweenwords)
– Abbreviations – minimize, be consistent,
truncate
(if you must abb. trunc. don’t abvt.)
– Line drawings may be superior to photo
renderings for shape perception
– Color and detail are often not needed
 Perception: Top Down Processing

• You see what you are prepared to

see (expectation)
• Uses context to resolve ambiguities
• Occurs simultaneously with bottom
up processing
• https://www.youtube.com/watch?v=aJy5_p_LAhQ
 From letters to words (unitization;
automaticity) to sentences (expectancy)

Aoccdrnig to rscheearch at Cmabrigde Uinervtisy,

it deosn’t mttaer in what oredr the ltteers in a word
are, the olny iprmoetnt tihing is that the frist and
lsat ltteer be at the rghit pclae. The rset can be a
total mses and you can still raed it wouthit
porbelm. This is bcuseae the huamn mnid deos
not raed ervey lteter by istlef, bu the word as a
wlohe.
 Perceptual processing: Depth perception

• Bottom up
– Accommodation: Activation of ciliary muscles
– Convergence: “cross-eyedness
– Binocular disparity: Different image in each eye
• Top down
– Linear perspective
– Relative size
– Interposition
– Light and shading
– Texture gradients (e.g., groups of objects appearing denser
as they move farther away)
– Relative motion
Traffic circle application of
depth cues
Perceptual processing: Visual search

• Unstructured visual search time= (N*I)/2

– N=number of targets
– I=search time per target

Design applications:
Cost of clutter, eliminate unnecessary items (BU)
Make critical items more salient (color, contrast) (BU)
Structure visual field (alphabetical, consistent location) (TD)
A visual search challenge
Perceptual processing: Visual
search
• Bottom up factors affecting search
time:
– Conspicuity (color, size, brightness,
motion, simplicity)
• Top down factors affecting search
– Expectancies (knowledge of
organization, knowledge of likely
location)
 Perceptual processing: Detection

• Described in terms of signal detection

theory (SDT)
• SDT defined by two parameters
– Sensitivity (d’): Ability to differentiate signal
from noise
– Response style (Beta): Risky vs.
conservative
Signal detection theory
https://www.youtube.com/watch?v=BHkw9_GWPJA
Baggage screening

https://bitrefine.group/aboutcompany/news/251-computer-system-has-
learned-to-recognize-x-ray-images-and-alerts-if-it-sees-illegal-items
 Affecting detection bias and sensitivity

• Sensitivity (d’): Ability to differentiate signal from noise

– Experience
– Visual template
– Redundant representations
– Alertness, frequent breaks
– Slow down rate of decisions
• Response style (Beta): Risky vs. conservative
– Value—indicate costs of hits, false alarms (motivation)
– Expectancy—Introduce false signals
Perceptual processing:
Discrimination and Absolute judgment

• Going beyond detection to categorize or to judge

magnitude
• Discrimination: judgement of differences
between two sources of information actually (or
potentially) present
– Norflox (antibiotic) vs. Norflex (muscle relaxant)
(typed and handwritten)
• Absolute judgment: Attaching labels to levels
(e.g., traffic lights)
– Limited to 5 levels
49 at most; no more than 7 colors
 Object and pattern perception

• Bottom-up and top-down processing

– Context affects both
– Design to enhance each (e.g., optimize
features, reduce noise; use redundancy,
enhance context)
• Icons vs. Text
– Can be universal, space efficient, effective
– Design for:
• Legibility, discrimination, and interpretation
• Standardization
 The Auditory Environment

• Humans sensitive to frequencies between 20 and 20,000 Hz,

0dB-140dB
• Dogs up to 45 kHz, Cats 65 kHz, Bats 110 kHz
• Psychophysical scaling translates physical stimulus to
perception
– Intensity or amplitude => loudness (dB)
– Frequency => pitch (Hz)
– Envelope (temporal characteristic) => quality
– Location => perceived location
– Higher harmonic frequencies => timbre

52
 Basic sensory characteristics

• Masking: Sound obscured by other

sound
– Decreases d’
– Minimum difference to ensure sound is
heard = 15 dB
– Greatest masking when frequencies are
similar
– Low-pitched sounds mask high-pitch more
than converse
• Consonants more susceptible than vowels
 Hearing loss

• Age-related decline in high frequency,

particularly for males
• Noise induced hearing impairment
– Loss of sensitivity when noise is present
(masking)
– Temporary threshold shift (concert effect)
– Permanent threshold shift, selective loss
 Perceptual processing
• Alarms
– Guarantee to be heard (30 dB over noise level)
– Not exceed danger to hearing (adjust
frequency)
– Not startle (at least 20 msec rise time)
– Not interfere with communication
– Should be informative (no more than 5 based
on limits of absolute judgment)
– Differentiate based on sound characteristics:
e.g., rhythm, timbre, pitch
 Perceptual processing

• Voice alarms
– More informative
– More confusable with voice communication
– Susceptible to masking
– Meaning may not be universal

56
The Cognitive Environment
Wickens et al Model of Human Information
Processing (Fig. 6.2)
 Attention

• Attention is the process of selecting which information to

process.
• Meaning can be extracted directly from perception without
conscious attention (cocktail party effect)
• Attention is required to support most cognitive activities (not
LTM or sensory registers)
• Analogies include:
– Shining a flashlight in a dark attic
– Channel switching
 Types of Attention
• Selective/Focused Attention
– Maintain attention on one or a few sources of
information and not be distracted by other
sources
• Divided Attention
– Required to do more than one task at the
same time
• Sustained Attention
– The ability of observers to maintain attention
and remain alert to stimuli over prolonged
periods of time (i.e., vigilance)
 Selective attention

• Role: Selective attention does not guarantee

perception, but is a necessary condition
• Process:
– Salience—pay attention to abrupt visual onsets,
loud noises, tactile cues (bottom-up)
– Expectancy—pay attention to what might change
or may be a source of useful information (top-
down)
– Effort—in attending is minimized (e.g., blind spot
check)
– Value—pay attention to what matters (top-down)
 Selective attention

• Design implications:
– High value information sources should
require minimal effort to attend
– Important, unexpected or rare events
should be conveyed in a salient manner
 Working memory (WM)
• Role: Keeps
information active
while in use
• Process:
– Central executive acts
as attentional control
system for 3 storage
units
 Working memory (WM)

• Chunk is the unit of working memory space defined by

physical and cognitive properties that bind items together
• Chunking is analogous to unitization for perception
• Limited capacity (4 chunks)
– USA, FBI, CIA (3 chunks not 9)
• Rehearsal is needed to retain
– attention demanding; disruption more likely if attention
is diverted to similar information
• Limited by time: half life=7s for 3 chunks, 70s for 1 chunk;
similar information will decay faster
 Working memory
• Design implications
– Minimize WM load (time*chunks)
• Side by side windows for MS
– Use redundant sources (e.g., text+speech)
– Provide placeholders for sequential tasks
– Exploit chunking (1-800-GET-HELP;
2376100 vs. 237-6100)
– Minimize confusability
– Order text/instructions carefully
– Avoid the negative
 Long-term memory

• “Memories are films about ghosts” (C. Crows)

??

• “We paint the remote past, as it were, upon a

canvas in our memory, yet often imagine that
we have direct vision of its depths” (W.
James) ??

66
 Long-term memory (LTM)

• Semantic or declarative memory –

knowledge and facts
• Episodic or event memory
• Procedural memory – memory of how to do
things (e.g., how to tie your shoe)
• Prospective memory – to remember to do
things in the future
67
 Long-term memory (LTM)
• Role: Keeps event information available for recall over
long periods of time
• Process:
– Item strength = f{frequency, recency, importance}
– Number of associations (associations between items
form chunks)
– Networks of associations (e.g., schemas, mental
models)
– Forgetting (weak strength, few associations,
interfering associations)
– Recognition is easier than recall
– Habit formation (environment, rewards)
 Organization of information in LTM

• Semantic network: Organization of

knowledge where sections of the network
have related information
• Schema: Knowledge structure about a
particular topic (MIE240)
– Script: Schema that is a sequence of activities
(turn off computer)
– Mental models: schemas of dynamic systems
• Cognitive maps: Mental representations of
spatial information
 LTM
• Design implications
– Encourage regular use (frequency, recency)
– Encourage active reproduction (e.g., taking notes)
– Standardize (make schemas and mental models
transferrable)
– Memory aids (knowledge in the world) – Support
recognition rather than depending on recall
– Carefully designed information (limit LTM dependence,
support associations, provide stories)
– Develop habits and correct mental models (e.g.,
information processing model for designers)
Knowledge trade-offs

-- Norman’s The Design of Everyday Things

Ali Baba and the Forty Thieves
 Divided Attention and Time-Sharing:
Multiple Attentional Resources

• Process: Multiple Resources

– Input modality (auditory, visual, tactile,
olfactory)
– Code (spatial, verbal)
– Stages (perceptual, cognitive, response)
• Each resource can be tapped for an
information processing task
• Conflicts when same resource is
demanded in parallel
Multiple attentional resources
Stroop Effect (response conflict)
 Stroop demonstration

• On the next three slides look at the

items down the columns and name
the color the item appears in
• Ignore what the item is - just focus
on the color
• See how long it takes to go through
each list
BLUE YELLOW RED
GREEN BLUE YELLOW
BLUE GREEN GREEN
RED YELLOW RED
YELLOW RED YELLOW
GREEN GREEN BLUE
RED YELLOW GREEN
YELLOW RED BLUE
GREEN BLUE RED
BLUE GREEN YELLOW
RED BLUE GREEN
YELLOW RED
BLUE
BLUE YELLOW RED
GREEN BLUE YELLOW
BLUE GREEN GREEN
RED YELLOW RED
YELLOW RED YELLOW
GREEN GREEN BLUE
RED YELLOW GREEN
YELLOW RED BLUE
GREEN BLUE RED
BLUE GREEN YELLOW
RED BLUE GREEN
YELLOW RED BLUE
 Stroop effect
• People decide faster when available
information is unambiguous
• With Stroop stimuli there is a response
conflict (e.g., colour word is displayed in
versus colour described by the word)
• Response conflict can be quantified in terms
of how much it slows people down and how
many errors they make
• Interventions like the Stroop task allow us to
probe “automated” processes and see how
they work
 Automaticity and Practice

• Criteria for a cognitive process to be

automatic
– It must occur without intention (e.g.,
reading the color names in the Stroop
task)
– It must occur without conscious
awareness
– It must not interfere with other mental
activity
 Task Switching

• When two tasks cannot be performed

concurrently
• Voluntary switching:
– Cognitive tunneling (when one task is
neglected)
– Salience, priority, interest, task difficulty,
time on task
• Interruptions
– Interruption/resumption lag and accuracy
 Design implications
• Reduce attention demands of task
• Provide information regarding importance to guide people in
allocating resources to tasks
– (e.g., design tone urgency to match message urgency)
• Design to engage different attentional resources
– (e.g, voice dial rather than manual on car phone)
• Facilitate timesharing by making tasks automatic

Action

Stop
Wickens et al Model of Human Information
Processing (Fig. 6.1)
 Cognition: Key concepts

• Cognition can be divided into processes,

each of which has implications for design
• Effective perception depends on bottom-
up and top-down processing
• Working memory is limited and volatile
• Chunking makes working memory
effective
• Long-term memory depends on
associations
• Attention is limited