Information vs Knowledge Flashcards

1
Q

working memory

A

handles relatively short-term (but volatile) storage and processing of memories- duration may be unlimited–as long as memory is actively rehearsed; is prone to interference- can hold 7 ± 2 items- codes information phonologically, visually, or semantically

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

working memory and implications for design

A

• minimize “cognitive load”: reduce number of things people are required to remember• provide visual “echoes”e.g., automated 411 services that speak a number could use phone’s display as well• provide placeholders for sequential tasks• exploit “chunking”: strategy that organizes several memory items into a larger whole - physical chunk size: optimal size is 3-4 numbers or letters per chunk- meaningful sequences:e.g., 7, 8, and 0 represent one chunk - letters superior to numbers:e.g., 1-800-GET-HELP- keep numbers separate from letters:e.g., licence plate of “458 GTS” better than “4G5 8TS”• minimize confusabilitye.g., A5433 easily confused with A5423; sound-alike letters confused with each other:B C P T and V• avoid unnecessary zerose.g., 002385 takes up unnecessary memory• ordering of text and instructionse.g., poor design: “before doing B, do A first” e.g., good design: “do A, then do B”• minimize distractions/interruptionse.g., compared to rotary phones, touch tone phones allow faster dialing, reducing riskof interruptions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

long-term storage

A

handles longer-term, more stable memories; potentially unlimited capacity and duration• declarative (or explicit) memory: “knowing what”- episodic memory: specific, “autobiographical” events - semantic memory: facts, general knowledge• procedural (or implicit) memory: “know-how”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

long-term memory/storage and implications for design

A

• encourage regular use of information- memory traces are strengthened by frequency and recency e.g., practice effects• standardizee.g., shift pattern for manual transmission is becoming standardized; same commandworks in different Windows® programs carefully design information to be remembered- should be meaningful/semantically related to other knowledge- avoid arbitrary memory, which makes it difficult to learn, and recover from errors:e.g., all prime ministers, Alt-F4 in Windows® - use concrete, not abstract words- distinctiveness reduces interference- organization/grouping- avoid jargonuse memory aids!-prompt user for correct operational sequencedesign to support correct mental models, using natural designknow that memory is not necessarily precise or infallible, is reconstructive!-introduced $1 coin to US, people confused with quarter, CHAOS ENSUES

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

the seven sins of memory

A

Sins of Omission/Forgetting1. Transience2. Absent-Mindedness3. BlockingSins of Commission/Distortion4. Misattribution5. Suggestability6. Bias7. Persistence

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Sins of Omission/Forgetting: transience

A

memory for facts and events becomes less accessible over time, both in short-term and long-term memory e.g., forgetting a phone number- causes: interference, retrieval failureGOOD it may be useful to forget old/outdated knowledge BAD but we may lose some that we need later

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Sins of Omission/Forgetting: absent-mindedness

A

forgetting caused by lapses of attention during encoding or attempted retrievale.g., watching The Amazing Race and burning the cookies- causes: low arousal/fatigue, interruptions/distractions, absence of cuesGOOD directing attention allows selective encoding of informationBAD but attention may be misdirected, precluding encoding

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Sins of Omission/Forgetting: blocking

A

temporary inaccessibility of well-encoded material e.g., tip-of-the-tongue phenomenon- causes: interference from similar itemsGOOD most memory contents should be inhibited or suppressedBAD recall should be of exact item, not similar one(good: Good, because when you are retrying to remember something, you don’t want a flood of a thousand related things that might be it, like a google search. You want THE answer.)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Sins of commission/distortions: misattribution

A

memory is retained, but an error is made in attributing the sourcea) source confusion: correctly remembering an item or fact, but misattributing it to an incorrect sourcee.g., thinking your friend told you about a great new restaurant, but you actually read about it in the newspaperb) cryptomnesia: misattributing an idea as one’s own, when it is really the product of previous experiences (STEVE JOBS SIGH)e.g., unintentional plagiarismc) false recall/recognition: false remembering of items or events that never happened e.g., thinking you got lost in the mall as a child- causes: reliance on semantic content (gist, not verbatim)(Memory is tied to meaning, not like a tape recorder. Evolutionarily, meaning tends to be the important thing)GOOD content can be encoded wellBAD but at the expense of (often irrelevant) source information

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Sins of commission/distortions: suggestability

A

tendency to incorporate (possibly incorrect) information provided by others into one’s memoriese.g., misleading questions asked of eyewitnesses- causes: memory is constructive, not reproductiveGOOD knowledge structures are changeableBAD plausible information may be incorporated into memory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Sins of commission/distortions: bias

A

current knowledge, beliefs, and expectations can influence and distort memories; memories of the past may be coloured by present mood/emotional statee.g., remembering bad behaviours of members of a disliked group, more than goodbehaviours- causes: schemas aid interpretation of incoming informationGOOD existing schemas are efficient, and organize knowledgeBAD at the expense of accuracy, when it comes to exceptions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

7th sin of memory, intrusive recollections that are difficult to forget: persistence

A

emembering a fact or event that one would prefer to forgete.g., recollection of traumatic events, chronic phobias- causes: amygdala (limbic system) and stress hormones tied to memory systemGOOD strongly emotional experiences are usually important to remember BAD no way to retain the knowledge without the memory PTSD :(

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

two types of memory aids

A

• Signal: don’t forget to remember! e.g., watch alarm, string on fingerGOOD cues rememberingBAD lacks content• Message: content; the info itselfe.g., shopping listGOOD provides contentBAD does not cue remembering- Is this a dichotomy? Is there any other possibility with no cons?- Natural Mappings need no extra reminders or labels; require no extra memory demandor interpretations of signs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

tradeoff between knowledge and infomation

A

(a) knowledge: you know what to doGOOD no external cues neededBAD time/effort is required to gain knowledge (learning) BAD memories are fallible(b) information: you follow instructions no memory demandGOOD external info may be more difficult to BAD use/understandBAD info may disappear (lost manuals)s there any other possibility with no cons?YES, cell phones

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Cockpit Checklist: origins

A

Army getting submissions for new aircraft (1935)-Boeing submitted the favorite, but after test flight, crashed and killed two crew members, injured 3!-cause was “pilot error,” forgot to disengage gust lock-Boeing lost contract, but 12 more aircraft purchased for further testing-pilots decided to make sure nothing would be overlooked again, and developed FOUR checklists: takeoff, flight, before landing, after landing-the 12 aircraft flew 1.8 million miles without a serious accident-army accepted aircraft in 1939, given the designation B-17, THE FLYING FORTRESS!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

recent cockpit checklist related crashes

A

Northwest Airlines flight 255- MD-82 aircraft crashed after takeoff from Detroit in 1987- flight crew failed to ensure flaps and slats configured for takeoff - flight crew distracted by weather and runway change- takeoff warning system had no power (disabled)Delta Air Lines flight 1141- B-727 aircraft crashed after takeoff from Dallas-Fort Worth in 1988- flight crew failed to ensure flaps and slats configured for takeoff- flight crew and flight attendants were discussing dating habits (despite FAA “sterilecockpit” regulation)- takeoff warning horn not operativeUSAir flight 5050-B737 crashed on takeoff, 1989-rudder incorrectly set, takeoff aborted, entered bowery bayALL AVOIDABLE IF CHECKLISTS WERE FOLLOWED

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

US Federal Aviation Regulation about checklists

A

a) airlines must have checklist for each aircraft typeb) all items necessary for engines/takeoff/landing/emergencies, designed so people dont need to rely on memoryc) checklist procedure must be “readily usable”checklists for:-proper aircraft configuration (not an option!)-for psychological reasons like team building and to facilitate communication-should NOT be considered a nuisance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Checklist procedure types

A

-“Do List” follow checklist like recipe (no redundancy)-Challenge-Response: configure aircraft from memory then use checklist as a backup (most common)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

engineering vs human factors POV in checklists. tradeoff?

A

the ITEM!-engineering POV, check EVERYTHING. human factors POV, check critical systems!-tradeoff was the glass cockpit

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

the glass cockpit

A

flight deck computers- automated monitoring of flight status- 6 displays: 2 in front of each pilot (flight info), 2 centre console (engine, systemsdata)Degani, Chappell, & Hayes (1991):- analyzed ASRS reports of altitude deviations- crews in glass cockpits detected more deviationsGOOD reduces workloadGOOD more reliableGOOD faster checklist; fewer chances for distractionsBAD poor displays can lead to mode errorsBAD computers can ___________ (NW 255, DAL 1141: takeoff warning systems failed)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Mode errors/confusion:

A

confusion or errors from being in a mode you don’t know you’re in, sometimes the electronic interface doesn’t signal the mode you’re in correctly, so you misread data

22
Q

Degani & Wiener Checklist Study (1993): methods

A
  • field study: observe flight crews in operation- interviews with flight crews- reviewed incident/crash reports implicating checklists (NTSB, ICAO, ASRS databases)(Partly a field study, but had to invent a cover story to allow them to go into the cockpit. Usually that’s forbidden! Especially nowadays. They said they were executives of the airline lol. So went to see if they did the cockpit checklists)
23
Q

Degani & Wiener Checklist Study (1993): observations

A

Observations:1.Checklist Initiation- no formal signal to begin- pilot decides when to initiate checklist, must remember if previous checklists werecompleted- may be affected by workload, stress WAFOS2. Calls & Responses- checklist cards ignored in favour of Memory - instruments not actually checked, didn’t even look at the object they were checking- cross-checking not always done3. Completion- not all airlines require final item “Is checklist complete?”

24
Q

Degani & Wiener Checklist Study (1993): Checklist performance factors

A

• Perception: misreading of controls or checklst-Degani, gave 19 recommendations including typeface, colors, glare, age of pilot even• Social factors: Cockpit or Crew Resource Management (CRM)-captain must give time to crew to perform duties-crew might have to challenge captain to do/finish checklist (AWKWARD)• interruptions/distractions-can preclude cross-checking-maintenance crews sometimes forget to reset things after changing configuration•ambiguity-some checklist items only ask for “SET”, is control really verified (what about ON, or OFF?)

25
Q

checklist design issues: longer checklist

A

-longer checklist -> greater chance overlooking an item.so subdivide long checklist into smaller ones…e.g. ground phase into (preflight, before start, after start, taxi, before takeoff), more manageable chunks, delay between phases allow for catching up.use CHUNKING, group items by…similar system controls…physical location: panel-by-panel…both!

26
Q

checklist design issues: item order

A

-item order.operational sequence (must do some tasks before others)…e.g. activate hydraulic pipes, THEN check hydraulix pressure.operational logic-denver flight, deferred checking for fuel, forgot to check, had to land immediately-aircraft not refueld prior to preflight checklist completion 75% of time!-move items back so memory need not be relied upon.sequencing of critical items…place important tasks FIRST, less prone to interruptions.item duplication…some items must be rechecked and reset…if so important, put it in there twice!…tradeoff: longer list

27
Q

systems approach to cockpit checklists: coupling, complex interactions, and management

A

coupling: integrate checklist into procedure of configuring aircraft- electronic checklists (ECLs) have many benefits:GOOD faster checklistsGOOD lower cognitive workloadGOOD decreased training needsGOOD prevents many errors resulting from paper checklists (e.g., skipping an item)(computer can remember where you left off or deferred items, won’t let you fly!)complex interactions.systems failure often caused by unpredictable interaction of several failed components.move critical tasks to earlier phase (some airlines require takeoff checklist on runway!!)management.lots of pressure to maintain on-time schedule, but checklists should NOT BE relegated to save time

28
Q

System: definition

A

a group of interrelated components forming a complex whole; all elements involved in a task or job

29
Q

systems: task/job

A

any behaviour with a goale.g., driving a car to get to work, using a word processor to write a term paper- helps change current state into goal state

30
Q

implications of the systems approach

A

-user/operator is part of system-system goals are superordinate.if system goals not achieved, system failed-systems are hierarchical-systems & components have inputs/outputs.error analysis considers relationships among components-a system has structure.system emerges from interacting parts, emergentdeficiencies in components –> deficiences in system.if a system fails, it is due to one or more system components.a system operates as part of a larger environment-the larger physical and social context must also be considered

31
Q

human-machine systems diagram (Kranz, 2001)

A

FUCKING LOOK AT IT OK, information knowledge world page 9

32
Q

different human-machine system interactions

A
  1. person supplies power and controle.g., shovel: dormant without human2. machine supplies power; person controlse.g., automobile: self-powered; requires human control3. machine supplies power and information; person controls & monitors- aircraft: require human pilot to fly; give feedback, so monitoring is needed4. machine supplies power, info and control; person monitorse.g., aircraft with Flight Management System, Flight Guidance System, and autopilot:humans monitor and detect errors
33
Q

Control Systems Theory: types of control

A

• open loop control: process cannot be modified or corrected once it is initiatede.g., irrigation sprinklers will not turn off if it rainse.g., ballistic motions: batters cannot change their swing once they start• closed loop control: process uses feedback to allow modification/error correction e.g., dishwashers may detect increased dirt in rinsewater and lengthen the wash cycle e.g., drivers can see where they are and maintain lane-holding behaviour

34
Q

Power of Feedback: Handwashing at Medical Centre

A
  • Cedars-Sinai Medical Center in LA had problems with hand-washing/disinfecting compliance- only 65% of doctors complied- problem: patients would become infected; doctors would not - several schemes were tried:• information campaign (e.g., posters, emails)• bottles of alcohol-based disinfectants handed out• incentives scheme: gave doctors who washed their hands $10 gift cards- results: compliance rose to 80%- applied feedback: after eating lunch, doctors pressed their palms into plates of agar,which were cultured for bacteria- images of the cultures, with myriad bacteria colonies, were shown to doctors (and madeinto a screensaver)- results: compliance rose to nearly 100%- other hospitals use special lights to reveal bacteria on hands
35
Q

Types of feedback

A

• negative feedback: opposes the condition that triggered ite.g., furnace activates on a cold day; A/C turns on when it’s hot e.g., running fast makes you tired, so you run more slowly• positive feedback: increases condition that triggered ite.g., getting to work late means you get a bad parking spot, so you have to walkfarther, which makes you even latere.g., network effect/Metcalfe’s law: one phone in the world is useless; its valueincreases, the more phones there are

36
Q

HVAC SYSTEM

A

see diagram, closed loop. page 11

37
Q

Constraints

A
  • reduce all possible actions to a limited subset which contains the correct one, thus reducing error- like “forcing functions” in engineering, which prevent incorrect actionse.g., scissors that have a smaller opening (for your thumb), and a larger one (for your fingers)
38
Q

Constraints: Physical

A

Physical: actions limited by shape or size - depends on visibility- closely related to real affordancese.g., can’t stick a Muffin in a toaster!

39
Q

Constraints: Logical

A

Logical: use reasoning to determine proper action- based on organized, principled relationship between components of a system - most important component of a good conceptual modele.g., assembling Lego toy–with one piece left(hard to come up with purely logical example, usually physical and logical)

40
Q

Constraints: Semantic

A

• Semantic: restricts choices based on meaning of situation - requires prior knowledge of the contexte.g., positions of Chess pieces at start of the game

41
Q

Constraints: Cultural

A

• Cultural: learned conventions/behaviours shared by a cultural group - based on social norms, traditions, rituals- choice of action may be arbitrarye.g., Chinese wedding invitations are Red(karsten sees lots of overlap between all these types of constraints)

42
Q

Constraints: Types

A
  1. Physical2. Logical3. Semantic4. Cultural
43
Q

Feature Fatigue

A

Experiment 1: participants rated software-based digital video players- player had 7, 14, or 21 features- 21-feature model chosen by 62.3% of people (but seen as less usable) Experiment 2: participants asked to build their own player- selected an average of 19.6 features (out of a possible 25) Experiment 3: working models of the players were actually used- two groups of participants: a) used 7-feature player b) used 21-feature player- both groups asked to choose a player:a) before use: 66% chose the 21-feature playerb) after use: only 44% chose the 21-feature player- conclusion: although more features/choices seem desirable, too many lead to decreased satisfaction

44
Q

Kline & Betel (1994): Design useful signs to indicate action for door - LAB STUDY

A

Do you use…• Words (language-limited?) • Pictures (ambiguous?)• Both (longer processing?)Stages of action:1. see sign/signal, 2. identify/understand it, 3. comply with it-got intended users to come up with signs (production method)-tested identification RT, and conspicuity.conspicuity, sign present vs. absent in 8 signs flashed on screen-also tests preference and meaningfulness (info rich?)results: - faster RTs for mixed-modality signs in ID task- but faster RTs for simpler signs in conspicuity task- symbol-only signs not preferred, less meaningful, generally slower

45
Q

Kline & Betel (1994): Design useful signs to indicate action for door - FIELD STUDY

A

Procedure:- signs placed at average (male) eye-level on doors at a U of Calgary library- doors could be opened by pushing or pulling (baseline: 42% pushed, 58% pulled) - observations made with no external cues (no models; door not still swinging)Results:- signs did affect behaviour- greater compliance with word-only and mixed-modality than symbol-only

46
Q

Kline & Betel (1994): Design useful signs to indicate action for door - CONCLUSIONS

A

Lab vs. Field Research- meaningfulness ranking correlated with compliance: r = .83 - efficiency index:best: hand/horiz.word, next: hand/arrow/horiz.word worst: symbol-onlyConclusions- best signs have little trade-off:words (speed) and symbols (universality)- extra stimuli doesn’t always hinder acquisition of relevant info- lab findings were different from field data–both should be considered

47
Q

Sign recall/recognition/identification affected by

A

• Context: limits and directs possible interpretations e.g., traffic signs• Function: type of task determines meaningfulness of icons- if verbal information required, use text–not icons- images may be more useful if user is less sure of info wanted• Underlying Conceptinfluences ease of mappings from domain concepts or items to meaningful graphical imagese.g., files, folders, documents have physical counterpartse.g., danger, hazard, safety, warning do notRepresentational form: ICON TYPE

48
Q

Icon Types - Representational Form

A

Resemblance icons.use image analogous to underlying concept(falling boulders on side of mountain to watch for)Examplar icons.typical example of concept(Fork and knife for restaurant sign)Symbolic icons.convey underlying meaning more abstract that image(e.g. fragility, the glass with crack in it)Arbitrary icons.have no relation to underlying concept(biohazard sign, designed and intended to not look like anything else)

49
Q

High Mounted Brake Lamp

A

required on passenger cars in 1986thought to reduce crashes by about 50%!!!!eye level, centre of vehicle, redundant with other brake lightsproblems.produces glare, bad for drivers behind.normal brake lights, not high ones, work equally well, still in periphery.difficult to see in vehicles ahead of you?.HMBL did not decrease braking RTweak link between results in lab and real worldin tests, HMBL is much more salientas become more commonplaceafter novelty wore off, reduced crashes only by 4.3%cost of installation was 206millionmedical/property damage savings was 655 million THO

50
Q

Angiolillo & Roberts (1991): What makes a manual look easy to use?

A

-if a manual looks difficult, people won’t even try to use it-ORGANIZATION: The perceived, not actual-Table of Contents, desired whitespace, NOT CROWDED-ease of FINDING INFORMATION: tabs, section headings-section titles: DISTINCTIVE is desirable (type size/font, indents)-color, figures, examples: the more the better, WITHOUT violating the designconclusions:-more organization and VISUAL APPEAL is good-margins (whitespace) and typeface not as important

51
Q

ASong & Schwarz (2008): What makes a manual ACTUALLY easy to use?

A

Study 1)-identical instructions for exercise regimen in either Arial or Brush typeface-in arial, people more open to exercising, believing it would take less time, and feel more fluid and easy. -Brush typeface reported opposite!Study 2)-japanese sushi rolls recipe in either Arial or Mistral (like brush)-arial, less likely to see recipe as time-consuming, and requiring high skill, more apt to try themselves-Mistral, the opposite!conclusion- we perceive things to be more appealing, easier to handle and more efficient, based on how easy they are to understand (“PROCESSING FLUENCY”)- this is also heavily influenced by surface features, such as the font of instructions

52
Q

Mayer (2002): Why won’t we read the manual?

A
  • automakers are required by LAW to provide an owner’s manual- about 20% of issues handled by Subaru customer service are about something covered in the (450-page) owner’s manual.so made an 8 panel folder with FAQ behind visor- as the number of functions performed by devices has increased, manuals have increased in complexity and length–and reading of manuals has DECREASEDhow fix? make devices INTERACTIVEe.g., one microwave oven asks if you are preparing a cooked or uncooked chicken, with or without bones, with or without sauce–it then calculates cooking time and method