HUMAN INFORMATION
PROCESSING
IE 341 Human Factors Engineering
1 Alabdulkarim, 2022
Content
• Human Information Processing Model
• Signal Detection Theory
• Memory
• Attention
• Multiple Resource Theory
• Validate the theory in an experiment!
IE 341 Human Factors Engineering
2 Alabdulkarim, 2022
Model of Human Information Processing
Wickens’ HIP Model
1. Represented as a series of stages whose function is to transform”
(or carry out other operation on the) information
2. There is no fixed starting point in the sequence of operations
IE 341 Human Factors Engineering
3 Alabdulkarim, 2022
Model of Human Information Processing
Sensory Processing
• Visual and audio receptors & environment can affect signal
STSS – Short Term Sensory Store
• Automatic
• Unlimited capacity
• Time limit
•0.5 seconds (visual)
•2-4 seconds (audio)
Perception
• Where raw sensory data is interpreted and given meaning
• Automatic and rapid (requiring little attention)
• Driven by sensory input (bottom up), and by inputs from
long-term memory storage (top down)
IE 341 Human Factors Engineering
4 Alabdulkarim, 2022
YouTube!
• https://www.youtube.com/watch?v=pMMRE4Q2FGk
• https://www.youtube.com/watch?v=piDEgCefgKI
IE 341 Human Factors Engineering
5 Alabdulkarim, 2022
Signal Detection Theory
• Threshold to detect signal
• Presence of Noise and Signal
r
• There is NOT an absolute Threshold
• It is a continuum…
• Detection thresholds (broken bone on xray, buzzer in noisy factory)
• Discrimination thresholds
• Pain
IE 341 Human Factors Engineering
6 Alabdulkarim, 2022
Distributions response
criterion
noise signal normal probability
distributions
Signal
+
Noise
X = evidence variable (e.g., rate)
response
criterion
noise
Noise
signal
X = evidence variable (e.g., neural activity)
IE 341 Human Factors Engineering
7 Alabdulkarim, 2022
Explanation of d’
• d’ : the sensitivity of the signal detection system
• Larger d implies higher sensitivity, easy to detect the signal
• Small d implies lower sensitivity, more difficult to detect the signal
• Measured in #SD’s that separate both curves
d d
IE 341 Human Factors Engineering
8 Alabdulkarim, 2022
Low Sensitivity (d’)
IE 341 Human Factors Engineering
9 Alabdulkarim, 2022
High Sensitivity (d’)
IE 341 Human Factors Engineering
10 Alabdulkarim, 2022
Shifting Response Criterion
• Personal differences
• Probabilities & Circumstances (e.g., xray w/ context)
• Payoffs – Penalties and rewards
says more yes => more hit but says no more => more
11 correct rejection but
more false alarms! more misses !
IE 341 Human Factors Engineering
Alabdulkarim, 2022
Bias (Response Criterion)
P (X | S)
• ß measured at the criterion point ß= P (X | N)
ß<1 ß>1
• If signal is more likely, shift criteria to the left for optimal
performance
• E.g., driving on busy road, more liberal (says more yes) braking
• ß will shift according to payoffs and individual differences
IE 341 Human Factors Engineering
12 Alabdulkarim, 2022
Shifting Response Criterion
• Depending on the culture, instructions, and experience, a
given operator might have a different response bias.
• They might be:
• Prone to say “Yes, there is a signal”
• Less probability of miss, higher probability of Hit
• But higher probability of false alarm, lower probability of correct rejection
• Prone to say “No, there is no signal”
• Higher probability of miss, lower probability of hit
• Lower probability of false alarm, higher probability of correct rejection
• When would you want to instruct operators to be more
liberal or more conservative?
IE 341 Human Factors Engineering
13 Alabdulkarim, 2022
SD Performance and Response Bias Can
Change with Time
• Change not only between individuals, but within a
single individual over time
• Vigilance decrement – SD performance
decreases over time
• “Alertness, ability to pay attention over a period of time”
• Prolonged monitoring leads to poorer performance after
about 30 min
• Performance also decreases if signal does not happen
very often
• Fire alarm drills
• Evacuation simulations
IE 341 Human Factors Engineering
14 Alabdulkarim, 2022
Ways to avoid Vigilance Decrement
• Shorter shifts
• Training and experience
• Memory Aids
• Signal transformations
• Feedback with alarms, signals, controls
• Use of Redundant coding (increase signal relative to
noise)
• Remove noise and extraneous stimuli
(increase signal relative to noise)
IE 341 Human Factors Engineering
15 Alabdulkarim, 2022
Videos!
• Signal detection theory
• https://www.youtube.com/watch?v=UHEiWws8t-g
• https://study.com/academy/lesson/signal-detection-theory-
definition-examples.html
IE 341 Human Factors Engineering
16 Alabdulkarim, 2022
Memory
IE 341 Human Factors Engineering
17 Alabdulkarim, 2022
Memory
• STSS
• Working Memory
• Immediate information recall
• Limited capacity, 7 +- 2 chunks of information
• Decreases with age!
• Stimuli are encoded
• Limited in time
• Facilitate storage – lists, reminders, feedback, “ditties”, repeat
• Long-term Memory (information data base)
• Unlimited in capacity
• Information easily lost
• Episodic memories (Spatial, temporal), Semantic memories (rules,
relationships), Mental Models (purposes of systems), Schemas
(integrations of information)
IE 341 Human Factors Engineering
18 Alabdulkarim, 2022
Limits to WM
• How much information can be held active?
• Miller (1956): 7±2 chunks
• More like 5±2 chunks (Cowan, 2001)
• X, F, D, U - 4 chunks
• DOOR - 1 chunk
• Lets see…
IE 341 Human Factors Engineering
19 Alabdulkarim, 2022
Limits to WM
• Slowly and steadily read the digits in a single row, (out
loud).
• At the end of each row close your eyes and try to repeat
the sequence of digits back, in the same order.
• If you make a mistake go onto the next row.
• The point at which you cannot correctly remember the
digits in any two rows, of a given length indicates your
capacity limit; the number of digits in the previous row
IE 341 Human Factors Engineering
20 Alabdulkarim, 2022
Limits to WM
• 8704
IE 341 Human Factors Engineering
21 Alabdulkarim, 2022
Limits to WM
• 2193
IE 341 Human Factors Engineering
22 Alabdulkarim, 2022
Limits to WM
• 3172
IE 341 Human Factors Engineering
23 Alabdulkarim, 2022
Limits to WM
• 57301
IE 341 Human Factors Engineering
24 Alabdulkarim, 2022
Limits to WM
• 02943
IE 341 Human Factors Engineering
25 Alabdulkarim, 2022
Limits to WM
• 73619
IE 341 Human Factors Engineering
26 Alabdulkarim, 2022
Limits to WM
• 659420
IE 341 Human Factors Engineering
27 Alabdulkarim, 2022
Limits to WM
• 402586
IE 341 Human Factors Engineering
28 Alabdulkarim, 2022
Limits to WM
• 542173
IE 341 Human Factors Engineering
29 Alabdulkarim, 2022
Limits to WM
• 6849173
IE 341 Human Factors Engineering
30 Alabdulkarim, 2022
Limits to WM
• 7931684
IE 341 Human Factors Engineering
31 Alabdulkarim, 2022
Limits to WM
• 3617458
IE 341 Human Factors Engineering
32 Alabdulkarim, 2022
Limits to WM
• 81042963
IE 341 Human Factors Engineering
33 Alabdulkarim, 2022
Limits to WM
• 07239861
IE 341 Human Factors Engineering
34 Alabdulkarim, 2022
Limits to WM
• 578149306
IE 341 Human Factors Engineering
35 Alabdulkarim, 2022
Limits to WM
• 293486701
IE 341 Human Factors Engineering
36 Alabdulkarim, 2022
Limits to WM
• 721540683
IE 341 Human Factors Engineering
37 Alabdulkarim, 2022
Limits to WM
• 5762083941
IE 341 Human Factors Engineering
38 Alabdulkarim, 2022
Limits to WM
• 4093067215
IE 341 Human Factors Engineering
39 Alabdulkarim, 2022
Limits to WM
• 9261835740
IE 341 Human Factors Engineering
40 Alabdulkarim, 2022
Chunking
• Chunk: set of adjacent stimulus
units tied together by associations
in the long-term memory
• a f g c j r w
• cat dog bat …
• “KSU is in Riyadh KSA”
• Chunking: recoding information by
semantically associating low-level elements
• Chunking can be helped or hindered by design
• E.g., license plates “YAG 3069” & “ZFL 8445”
• Letters afford better chunking
• Chunking is facilitated by parsing
4149283141865 > 4 1492 8 314 1865 > for Columbus ate pie @ Appomattox
https://www.youtube.com/watch?v=hydCdGLAh00
IE 341 Human Factors Engineering
41 Alabdulkarim, 2022
HF Implications of WM limits
• Minimize working memory load
• “Lazy” user
• Provide visual echoes
• Redundancy in displays
• Provide placeholders for sequential tasks
• Reminder of what steps have been accomplished
• Exploit chunking
• Physical chunk size: 3 to 4 letters or numbers per chunk
• Meaningful sequence and Spacing: IB MJF KTV → IBM JFK TV
• Superiority of letters over numbers: 1 800 GET HELP
• Keeping numbers separate from letters
• Minimize confusion
• Use physical distinctions such as spatial distance
• Avoid unnecessary zeroes in codes
• Consider WM limits in instructions
• Avoid long sentences
• Avoid unfamiliar words or codes
• Before doing X and Y, do A vs Do A. Then do X and Y.
IE 341 Human Factors Engineering
42 Alabdulkarim, 2022
Attention
IE 341 Human Factors Engineering
43 Alabdulkarim, 2022
Defining Attention (broadly)
• Attention is the process of allocating cognitive resources
to a task
IE 341 Human Factors Engineering
44 Alabdulkarim, 2022
Attention
• Focus of your consciousness
• Limited resources: How many tasks can you do at
once?
• Focused attention: Concentrating on one specific
signal
• Talking to your friend in a gym
• Sustained attention or vigilance: Prolonged period of
time monitoring infrequent signal
• Security guard monitoring TV monitors
• Selective attention: Monitoring several channels of
information to perform one task
• Flying a plane, driving a car
• Divided attention: Performing two or more separate
tasks
• Driving and talking on the phone
IE 341 Human Factors Engineering
45 Alabdulkarim, 2022
Attention
divided attention
focused attention direction = selective attention
IE 341 Human Factors Engineering
46 Alabdulkarim, 2022
Attention
http://www.youtube.com/watch?v=IGQmdoK_ZfY&feature=player_embedde
d#t=0
IE 341 Human Factors Engineering
47 Alabdulkarim, 2022
HF Implications of Attention
• Selective
- several channels, single task (drivers, pilots)
• As # of channels increases, performance decreases
• If signals are present in a few channels more often, we attend to those
• Guidelines for Design of Tasks
• Use few channels
• Identify important channels for effective resource allocation
• Reduce stress
• Preview where signals will occur in the future
• Train on effective scanning patterns
• Place channels close together
• Make auditory channels distinctly different from background
noise
• If responses are required, ensure there is ample time to
respond
IE 341 Human Factors Engineering
48 Alabdulkarim, 2022
HF Implications of Attention
• Focused attention task design
• Largest obstacle is noise
• Make each channel distinct
• Separate competing channels from the
channel of interest
• Decrease the number of channels
• Distinguish the channel of interest
• Sustained attention task design
• Work rest schedules are important
• Increase conspicuity of signal
• Decrease the uncertainty to when signal
will occur (provide a precursor)
• Use artificial signals and provide feedback
on detection performance
IE 341 Human Factors Engineering
49 Alabdulkarim, 2022
HF Implications of Attention
• Divided Attention
• Guidelines for Task Design
• Minimize the number of sources of information
• Where stress is inevitable, then provide rules for
prioritization
• Keep the difficulty low
• Keep tasks dissimilar to utilize all possible attentional
resources
• Greater learning of manual tasks reduces drain
• Use different modalities to allow more efficient time
sharing
• Different modalities for perceptual, processing, response
• Resource Pool Theories
IE 341 Human Factors Engineering
50 Alabdulkarim, 2022
Mental Workload
IE 341 Human Factors Engineering
51 Alabdulkarim, 2022
The architecture of multiple resource theory
Executive Control
Resource Demand Multiplicity
IE 341 Human Factors Engineering
52 Alabdulkarim, 2022
Resource Allocation Model (Kahneman, ‘73)
“Reserve Capacity” “Overload”
Spare
capacity
reserved
for
secondary Resources supplied
by user
task MAX
(task 2)
Attention Resource Supply
Primary task
performance
Capacity consumed by primary task (task 1)
Red line of workload
Attention Resourced Demanded (Difficulty)
IE 341 Human Factors Engineering
53 Alabdulkarim, 2022
Multiple Resource Theory
driving
Listening to
talk show
• EX: draw two time-shared tasks within in the multiple resource cube
• driving and listening to talk show
• No common resources needed for the two tasks shown, there should be minimal
competition of resources
• https://www.youtube.com/watch?v=cKjZMU5QAqM
IE 341 Human Factors Engineering
54 Alabdulkarim, 2022
Multiple Resource Theory
• Different tasks will require different resources on any of the three
dimensions. This leads to three phenomenon:
• more efficient time-sharing
• changes in the difficulty of one task not leading to a decrease in
performance on the other task
• the performance operating curve between the tasks being more
boxlike as they require different resources
• Information processing stages: Individuals are able to effectively
time-share between tasks that have a perceptive-cognitive demand
and a response demand.
• Example: A motorist is able to acknowledge vocally each
change happening in the motor car (a response demand)
without disrupting his/her ability to maintain an accurate
mental model of the environment (a perceptual-cognitive
demand) through which he/she is driving
IE 341 Human Factors Engineering
55 Alabdulkarim, 2022
Multiple Resource Theory
• Perceptual Modalities: Individuals are able to divide attention between
visual and auditory tasks better than between two auditory tasks or two visual
tasks. In other words, cross-modal sharing is better than intra-modal sharing.
• Example: An individual is able to listen to music and read a book at
the same time, as opposed to watching two television programs on
separate televisions simultaneously, or listening to two people talking
at the same time.
• Processing codes: Individuals are able to divide attention between spatial
and verbal processes better than between two tasks of the same processing
codes.
• Example: A secretary is able to type a letter while simultaneously
reading it off the paper that his/her boss has written it on.
• These three dimensions are not intended to account for all the structural
influences on the performance of two tasks and the ability to time-share
efficiently between these two tasks.
IE 341 Human Factors Engineering
56 Alabdulkarim, 2022
Mental Workload
•Mental workload is:
• The degree of mental effort required for
sufficient task performance
• Dependent upon operator experience (why?)
• Multidimensional
• Effort, frustration, feeling of time pressure, sense of performance
quality, result of competition between low-level subconscious
processing (ambient attention) and conscious task processing
•Mental workload is not:
• Stress
• A unitary dimension
• No single measure can completely measure WL (primarily because
we don’t know the interaction between mental and physical
workload and because we don’t fully understand the functioning of
brain)
IE 341 Human Factors Engineering
57 Alabdulkarim, 2022
Mental Workload
• Relationship between task demands and operator’s mental
resources (attention, working memory)
• Much like physical workload characterized energy demand of
muscles
Demands
Lower mental
workload
Resources
Higher mental
Demands
workload
Resources
IE 341 Human Factors Engineering
58 Alabdulkarim, 2022
Mental Workload
Physiological Measures
based on single resource theory
• Evoked brain potentials (voltage recorded from scalp,
smaller wave = less resources avail)
• Pupillary response (dilation or constriction)
• Respiration rate
• Body temperature
• Advantages of physiological measures
• Continuous data collection
• Don’t interfere with primary task performance
• No activity or performance from participant (totally objective)
• Disadvantages
• Bulky equipment
• Don’t isolate specific stages of information processing (is the WL in
detection, WM, or what?)
• Subject to changing due to other factors
IE 341 Human Factors Engineering
59 Alabdulkarim, 2022
Mental Workload
• Subjective Measures
self-reported levels of workload
• many say these are the closet representations of
WL (individual based)
• We can measure this stuff…kinda
• Questionnaires, surveys (most common NASA TLX,
SWAT, Cooper Harper Scale)
• What are some inherent issues with using subjective
measures to measure MWL?
• Performance Measures
primary & secondary task measures
• Response time & execution
IE 341 Human Factors Engineering
60 Alabdulkarim, 2022
IE 341 Human Factors Engineering
61 Alabdulkarim, 2022
Lets plan the lab!
IE 341 Human Factors Engineering
62 Alabdulkarim, 2022
Model of Human Information Processing
Cognition and Memory
• Cognition requires more time, mental effort, or attention
• Cognition is resource limited, and vulnerable to
disruption
IE 341 Human Factors Engineering
63 Alabdulkarim, 2022
Model of Human Information Processing
Response Selection and Execution
• Selection can be fast!
• Execution requires coordination of muscles
IE 341 Human Factors Engineering
64 Alabdulkarim, 2022
Decision Making
• Evaluate alternatives & select course of action
• Human decision making process (from research):
• Lazy
• Primacy
• Treat all information sources as equally reliable
• Personal preferences influence our decisions
• Limited capacity for analyzing options (5±2 chunks)
• Fear of mistakes
IE 341 Human Factors Engineering
65 Alabdulkarim, 2022
• HIP Model
• https://www.youtube.com/watch?v=pMMRE4Q2FGk
• Multitasking and attention
• https://vimeo.com/240556710
• https://www.youtube.com/watch?v=ubNF9QNEQLA
• https://www.youtube.com/watch?v=Lw-YPKR0grk
IE 341 Human Factors Engineering
66 Alabdulkarim, 2022
