HF Core Knowledge

Question 1

Sluggish Beta

Answer 1

People are slow to adjust their response criterion. We overestimate the probability of rare events and underestimate the probability of frequent events

Question 2

Change blindness + Factors that make a change less likely to be detected

Answer 2

Failure to notice change. Factors that make a change less likely to be detected include: being engaged in a different task, change based on memory is harder than change based on perception, change is unexpected, N-SEEV

Question 3

Inattentional blindness

Answer 3

Failure to notice something even when looking directly at it (gorilla basketball example). Related to difficulty of primary task and the degree of visual similarity

Question 4

SSTS Model: ST = a_p + bN/2

Answer 4

ST = search time, a_p = residual non-search components, b = time to inspect each non-target item, N = set size

Question 5

Guiding attention: central vs peripheral cues

Answer 5

Central cues are more cognitively driven, take longer to process, and aren’t valuable when unreliable. Peripheral cues are perceptually driven and always beneficial even when unreliable.

Question 6

Proximity compatibility principle

Answer 6

Information that is compatible should be close in proximity. Close in proximity can include: close in space, color, connections, and features. Applies to information integration: elements mapped to a single task. There is a free lunch as close proximity (> 2 deg) increases integration without decreasing focused attention.

Question 7

Ecological compatibility

Answer 7

Properties of an interface reflect the dynamics of the physical system. This will help the operator’s mental model to correspond better to the physical system dynamics.

Question 8

Display compatibility

Answer 8

Achieved by display representations whose structure and organization are compatible with the user’s mental model

Question 9

Principle of pictorial realism

Answer 9

The direction and shape of the display should be compatible with physics and mental model (high is up, low is down). If a physical system is analog, then the display should be analog (e.g., flight altitude). Still don’t want it to be too real and have irrelevant information (like a super realistic map)

Question 10

5 Color coding considerations

Answer 10

o A unique color stands out (allows for more rapid parallel search)
o Color hue is useful for coding categorical information (red=bad)
o Don’t use more than 7 hues (less if in bad lighting like a cockpit)
o Certain colors have societal meanings, and depends on country (green=go)
o Color hue does not generate a natural ordering (green is not “more” than red) so it’s bad for relative judgment/comparison tasks

Question 11

Principle of the moving part

Answer 11

Direction of movement of an indicator on a display should be the same as the direction of movement in a mental model (needle moving up should represent altitude getting higher). Can conflict with the principle of pictorial realism (e.g., moving scales) but this conflict can be mitigated with hybrid displays that move in multiple ways.

Question 12

Direct perception vs Indirect perception

Answer 12

Direct perception is automatic, egomotion, ambient vision, ecological, optical flow. Indirect perception is cognitive inference, object perception, focal vision, information processing

Question 13

Keyhole cost and Line of sight ambiguity

Answer 13

Keyhole cost occurs in 3D egocentric views in which only viewing what’s in front of you hinders understanding. LOS ambiguity occurs in 3D exocentric views in which location and movement of objects in 3D space is ambiguous on a flat surface.

Question 14

4 basic guidelines to create visual momentum

Answer 14

o Use consistent representations
o Use graceful transitions
o Highlight anchors (invariant features e.g., North)
o Display continuous world maps (e.g., map of a county has the state on it)

Question 15

Factors that affect icon usability: concreteness, visual complexity, semantic distance, familiarity

Answer 15

o Concreteness: the extent to which the icon depicts a real-life object. Matters initially but we can get used to abstract icons.
o Visual complexity: the intricacy of an icon. The more visually complex, the higher the search time
o Semantic distance: the degree of closeness of the relationship between the icon itself and its meaning (picture of printer for ‘print’ vs picture of turtle for ‘slow’). It takes effort to interpret an icon like a turtle.
o Familiarity: the user’s experience with the icon and the object in the icon

Question 16

Spatial Contiguity

Answer 16

For instructions, text should be near pictures to reduce cognitive load (words and pictures should be linked)

Question 17

Temporal Contiguity

Answer 17

For instructions, voice/verbal information can also be effective with pictures, but the voice should be presented at the same time as their correlated images

Question 18

4 benefits of face to face communication over voice-only communication

Answer 18

o Visualizing the mouth - provides redundancy
o Nonverbal cues - from facial expressions and gestures provide info
o Disambiguity – nonverbal cues allow for more flexible conversations, clears ambiguity
o Shared knowledge of action – seeing a team member act is informative

Question 19

4 roles for the central executive

Answer 19

o To temporarily hold and manipulate information stored in long term memory
o To change retrieval strategies from long term memory
o To coordinate performance on multiple tasks
o To attend selectively to stimuli

Question 20

Proactive interference and Retroactive interference

Answer 20

Proactive interference occurs when the previous activity disrupts retrieval (e.g., looking at a mugshot before a lineup). Retroactive interference occurs when new learning interferes backwards in time (witnesses have a hard time remembering suspects if other people are shown)

Question 21

Cognitive load theory

Answer 21

The mental workload of a learner can be partitioned into 3 elements
o Intrinsic load: mental resources required to perform a task
o Germane load: demand for mental resources needed to learn a task
o Extraneous load: mental demand from unrelated sources like bad user interface

Question 22

Dual coding principle

Answer 22

Material is better retained if it has multiple representations in the brain. It’s why you should have multimedia instructions to decrease extraneous load

Question 23

Mental model

Answer 23

A mental structure that reflects the user’s understanding of a system. It’s the source of expectancies about how a system will respond to changes.

Question 24

Three properties of cues that influence selective attention (cue seeking and hypothesis formation):
Cue diagnosticity, reliability, Physical features

Answer 24

o Cue diagnosticity: how much evidence a cue offers (e.g., dark clouds for rain)
o Cue reliability/credibility: the likelihood that a cue can be believed (e.g., a witness)
o Information value = cue diagnosticity x cue reliability
o The physical features of the cue: conspicuity and saliency affects selective attention

Question 25

Representativeness heuristic

Answer 25

People tend to make diagnoses by matching how many of the cues represent the “ideal” prototype (I have chills and a fever which are two cues that match the “common cold” prototype)

Question 26

Availability heuristic

Answer 26

People entertain hypotheses that are easier to recall. Recency makes hypotheses easier to recall (when you learn something new you see it everywhere). Simplicity makes hypotheses easier to recall (occams razor)

Question 27

Anchoring heuristic

Answer 27

We favor initial hypotheses and do not easily shift views. There is both primacy and recency bias

Question 28

Confirmation bias and cognitive tunnel vision

Answer 28

People look for information that supports their initial hypothesis, people ignore information that contradicts their initial hypothesis

Question 29

Decision fatigue

Answer 29

Repeated decisions leads to decreased effort in accuracy and analysis (the probability of granting parole was 75% in the morning and 25% in the evening because the default strategy of not giving parole dominates with fatigue)

Question 30

Heuristics and biases in uncertain choice:

• Direct retrieval
• Loss Aversion
• Endowment effect
• Framing effect

Answer 30

o Direct retrieval: people choose options that have worked for them previously
o Loss aversion: people care more about not-losing than they do about winning
o Endowment effect: we overvalue our own things (why I overvalue my own fantasy football players)
o Framing effect: peoples views change based on an object’s description (e.g., 80% lean vs 20% fat)

Question 31

Stimulus-Response compatibility

Answer 31

Compatibility between a display location/movement and the location/movement of the associated operator response. Consists of both static (location) and dynamic (movement) compatibility

Question 32

Location compatibility, colocation principle, congruence principle

Answer 32

o Colocation principle: controls should be next to relevant displays
o Congruence principle: spatial array of controls should be congruent with the spatial array of displays

Question 33

Movement compatibility

Answer 33

Controls should move in the same direction as the object in the world moves

Question 34

Mistakes

Answer 34

Failing to formulate the right intentions. There are knowledge-based mistakes and rule-based mistakes. Knowledge-based mistakes result from biases and insufficient knowledge, the operator is likely unsure of whether they made the right call. Rule-based mistakes result from a rule being misapplied, the operator is likely confident (incorrectly so) that they made the right call.

Question 35

Slips

Answer 35

The right intention is wrongly carried out, likely because the intended action is close to another routine action, or you aren’t paying attention

Question 36

Lapses

Answer 36

Failures to carry out any action

Question 37

Mode errors

Answer 37

Forgetting that a certain setting is on (e.g., using rads instead of degrees on a calculator)

Question 38

Six properties of the Ongoing Task that affects switching

Answer 38

o More engaging OTs are harder to interrupt (cognitive tunneling)
o Auditory OTs resist interruption because they need to be rehearsed
o Unstable dynamic systems (e.g., merging lanes while driving) resist interruption
o Higher priority OTs resist interruption
o Interruptions are less bad if they occur after a subgoal (e.g., finishing reading a paragraph)
o Waiting for a moment before switching tasks helps you remember where you left off

Question 39

4 Properties of the Interrupting Task that affects switching:

Answer 39

o More salient ITs are more likely to be switched to
o Tactile and auditory disruptions are more salient (sensory/auditory preemption)
o ITs that are similar to OTs degrade the return to the OT
o Long and difficult ITs degrade the return to the OT

Question 40

Cognitive Appraisal Theory

Answer 40

Stress reactions reflect a person’s cognitive appraisal of the environmental event and their competence in coping with it

Question 41

Complacency (in automation)

Answer 41

When over-trusting automation, humans are less likely to detect when automation fails and less likely to know the state of automation. There is a loss of situation awareness.

Question 42

The Generation Effect

Answer 42

It’s easier to remember an action you have chosen for yourself than an action that you watched someone else do

Question 43

Automation Bias

Answer 43

We tend to default to automation’s choices rather than perform our own information seeking and processing (i.e. we fail to check the raw data). It can also create attentional tunneling

Question 44

S-C-R Compatibility

Answer 44

The principle of stimulus/central-processing/response compatibility prescribes the best association of display formats to the codes of working memory used by a task

Question 45

HTA, steps to perform it

Answer 45

Hierarchical task analysis: Used to explain and represent the activities performed in complex systems. Steps include:

1. Identifying the task
2. Collecting task data
3. Analyzing the data
4. Representing the analysis

Question 46

GOMS

Answer 46

Goals, Operators, Methods and Selection Rules: Part of a family of HCI based techniques that is used to provide a description of human performance. Provides a predictive, descriptive, and prescriptive model.

Question 47

FMEA and HFACS

Answer 47

Failure Mode Effects Analysis (FMEA) and Human Factors Analysis and Classification System (HFACS) two methods for accident/error investigation.