IE 8000

Human Factors Engineering

Workload and time sharing
Fall 2024

1
 Time sharing
• Based on divided attention

• Ability to successfully time share depends on:

– Resource demands
– Switching costs and allocation skills
– Structural limits (e.g., bottlenecks)
– Confusion
 Resource demands
• Mental and perceptual resources are limited
• Can use the Performance Resource Function to
discuss resources
High Easy

• Easy tasks demand

Performance
less resources than hard Difficult

tasks for the same Low

level of performance Low High

Resources
 Performance resource functions

• Two time shared tasks

 Use of heuristics
• Heuristics are seen as mental shortcuts
• Improve the performance with lower resource utilization
 Switching costs and allocation skills
• Each time you switch between tasks, there is some switch-cost
in loading a new task
• The more switches that you make, the more times you will
incur a task-switching cost
• Trying to juggle two tasks at once is less efficient than finishing
one task before starting another.
 Structural limits
• Bottleneck Theory
– In the psychological refractory period (RPR), two tasks interfere with each
other only if they attempt to use the same stage of processing

• Multiple Resource
– The mind is made up of several independent units, each with a resource
pool
– If two tasks need to access the same units, then they will have to share
resources
– Tasks that use separate units do not interfere with each other
– Two verbal tasks are difficult to do at the same time.
 Confusion
• Cooperation
– When displayed information is similarly defined and presented in
proximity performance can be improved.
• Confusion
– However, when tasks are too similar, they can be confused.
• Trade off between confusion and cooperation. There is a fine
line between the benefits and the tradeoff
 Workload

"Workload is not an inherent property, but

rather it emerges from the interaction
between the requirements of a task, the
circumstances under which it is performed,
and the skills, behaviors, and perceptions of
the operator."

Hart & Staveland, 1988

Additional material from: Farmer & Brownson (2003) Review of Workload: Measurement, Analysis and Interpretation Methods
 Workload
• Ratio of time required (TR) to do a task to the time
available (TA)
Workload =TR/TA
• Can be used to predict
– the amount of subjective workload
– the performance decrement caused by workload overload
• Time stress and task scheduling
– People focus on tasks believed to be important and available
– Train to speed processing
– Train to calibrate importance
Task A

Task B

Task C

Time
Workload
 Measurement of mental workload
• Primary task performance
– Higher workload makes performance worse
• Secondary task measures
– Measure of spare capacity
– Mental arithmetic, time estimation, memory, reaction time
• Subjective measures
– Questionnaires
• NASA TLX (Task load Index)
• DRAWS (Defense Research Agency Workload Scale)
 Criteria for workload assessments
• Sensitivityà Detect changes in task difficulty or resource
demand.
• Diagnosticityà Should identify the sources of changes in
workload.
• Selectivityà Should be sensitive to workload, but not to
changes in physical load or emotion.
• Obtrusivenessà Should not interfere or affect the
primary task(s)
• Bandwidth and Reliabilityà Should be reliable and
should estimate changes in workload quickly enough.
Physiological measures of mental workload
• Heart rate and heart rate variability
• Respiration
• Perspiration
• Muscle tension
• Pupil dilation
• Eye fixations
• Galvanic Skin Resistance
• Saliva (spit test) (salivary cortisol) van Eck, Berkhof,
Nicolson and Sulon (1996)
• Urine (catecholamines and cortisol) Fibiger et al.,
(1986)
 Mental workload measurement dissociations

Workload measures may not always agree

• May have higher motivation (hence perform better) the first
time around but not for subsequent tasks.
• Performance and arousal curve
• Subjective measures (how they think they perform) may not be
equal objective measure (how they really perform)
Multiple measures are often needed to understand particular
situation
 Recommendations for workload
• Task redesign to assign tasks to automation or other operators
• Display designs with important tasks as available, interpretable,
and salient.
• Training (two approaches)
– Training on one task
• to increase automaticity
– Task management skills
• calibrate operators management of emergency or high workload situations
 Effect of training on workload
• A skilled worker will require fewer resources
to do the same task. Skilled

Unskilled

Performance

Resources used
 Fatigue
• Result of excess mental/physical workload
– Can lead to fatigue buildup
– Problems with maintaining attention
• Maintaining sustained attention to vigilance
tasks in low-arousal environments
– Can also lead to fatigue buildup
• Causes of vigilance decrement
– Time (duration)
– Event salience
– Signal rate
– Arousal level
 Recommendations for vigilance enhancement
Enhance arousal and make detection easier
• Signal enhancement
• Frequent rest breaks
– Vigilance declines in as little as 20 minutes
• Modify operator criterion
– Signal detection theory (large pay offs for hits or correct rejections,
depending on task).
• Introduce false signals
• Stimulants
– Caffeine
– Music
– Conversation
• Avoid sleep disruption
 Fatigue and performance
• Highly affected
– Judgment
– Learning new material
– Planning
• Less affected
– Motor activity/physical tasks
– Tasks involving highly interesting material
 Factors affecting fatigue
• Time of day (circadian rhythm)
• Time since last sleep
• Time on task, hours of work
• Amount and quality of sleep (sleep debt)
• Sleep inertia
• Circadian desynchronization
• Lack of environmental stimulation
 Circadian rhythm
• Latin word coined by Franz Halberg for circa (about)
and dies (day).
• ~24-hour cycle in the physiological processes of living
beings
• Circadian rhythms may be defined by three criteria:
– persists in constant conditions (for example constant dark) with
a period of about 24 hours
– can be reset by exposure to a light or dark pulse
– is temperature compensated, meaning that it proceeds at the
same rate within a range of temperatures.

http://www.nigms.nih.gov/Education/Pages/Factsheet_CircadianRhythms.aspx
Why does alertness matter?
Error rates in reading meters
 Circadian desynchronization
= Circadian rhythm does not match day/night cycle
= Mismatch between when your body wants to sleep and when
you need to be awake
• Shift work
– Fast rotation worse than slow
– Delayed shift better than accelerated (get up late)
• An extended day allows the body to extend the circadian cycle in line with the
unconstrained 27-hour cycle
• Easier to stay up late than get up earlier
 Circadian desynchronization
• Jet lag
– Eastbound worse than westbound, similar to delayed shift
– Allow one day for each time zone for recovery

• Shift work
– Weekend effect
– Switching between days and nights
– Longer shifts
 Fatigue mitigation
• Get more and better sleep: Sleep hygiene/
management
– No caffeine before bed
– Consistent sleep periods, with 8 hours per day
in a single period
– Attitude, as in medical interns
– Quiet, dark place to sleep
• Napping
– More than 15 minutes, but no more than 90
minutes
– Stop at least 10 minutes before work
 Fatigue mitigation/
Endurance management
• Sleep credit: sleep more before sleep
deprivation period
• Less effective: Drugs
– Melatonin, to support circadian synchronization
– Caffeine and other stimulants, to boost alertness
• Less practical: Curtail operations
– Siesta
– Avoid nighttime operations, consistent time in bed
 Drowsy Driver Detection Systems
• Systems have been designed to monitor your eye closure
patterns (PERCLOSE)
• Now can monitor facial muscles

http://gizmodo.com/5865203/driver-drowsiness-checker-watches-your-facial-muscles-instead-of-your-eyes
http://techon.nikkeibp.co.jp/english/NEWS_EN/20111202/202107/?SS=imgview_e&FD=46728356