applied Flashcards
what do health psychs do?
could b clinical psychs working w/ ppl with other health issues, could b promoting health, could work on therapeutic relationship between patient + other healthcare workers
environmental psych
how ppl relate to their environment, might look at factors how neighbourhood u live in could influence behaviour
scientist-practitioner model
need to do research + apply it
descriptive research methods examples:
- observations - observe ppl’s behaviour
- focus groups - talking 2 a group of ppl
- epidemiological studies - looking at disease trends
- incident analysis - why r things going wrong, how could they b better
absolute validity
to what extent do the results u get realistically reproduce what u get irl
face validity
how measure appears to participants, do they see it as valid?
relative validity
do we get same pattern of results across different methods
what indicates practical significance + not just statistical significance?
how big is effect size, how meaningful is effect
ergonomics (more physical work)/human factors (more psych aspects) brief def
concerned w/ understanding interactions between humans + other elements of a system
when did human factors get popular?
mid-20th century, triggered by WW2 - getting psychologists 2 work on optimising the work of military + air force personnel etc.
what does HFE stand 4?
human factors n ergonomics
HFE characteristics
- takes a systems approach - looking at how humans work w/ tech + other systems, as well as overall system in which humans operate (e.g. laws)
- design driven + outcome-focused (performance, safety, user satisfaction)
key difference between HFE + org psych:
org psych is people working w/ people, HFE is people working w/ things/tech
usability
how easy + pleasant is it 2 use
what does the design aspect of HFE look @?
- usability
- equipment (physical ergonomics)
- tasks (look at way it’s being carried out - is it effective)
- environment
- organisation/regulation
ambient environment
people’s immediate work space
e.g. of issues w/ workplace processes
carrying out abortion on wrong person due 2 workers not following procedures as they should + having communication difficulties
error of commission
action that shouldn’t have been done or was done incorrectly
error of omission
action that wasn’t done (failure to execute action), more common in healthcare
human error template/HET
- fail to execute
- execution incomplete
- executed in the wrong direction
- wrong task executed
- task repeated
- executed on wrong interface element (e.g. B-17 using wrong control)
- executed too early
- executed too late
- executed too much
- executed too little
- misread information
- other
HTA/hierarchical task analysis def
working out what the tasks r
Stanton et al., 2010
- 8 undergrads trained 2 use HET
- compared errors that they predicted w/ alternative methods 4 error prediction + actual errors that ppl had made flying planes
- gave them diff scenarios, undergrads worked out how likely particular errors were
- key limitation of using this is that likelihood is really easy 2 quantify, but criticality is harder 2 quantify - differing criticality for diff individuals
vigilance
ability 2 maintain attentional focus n remain alert over prolonged periods, can involve detecting discrete events (e.g. warning light that smth’s gone wrong)
key features of vigilance tasks
- monitor 1+ info task
- prolonged period - between 5 mins-2 hrs for research vigilance tasks, irl example could b a security guard working 8 hr shift
- detect low probability signal
- signal requires response
features of a continuous performance task
- items r presented 1 at a time on the screen
- task is to respond to a specific item (e.g. respond only if x follows a)
- interested in attention span + memory
d prime def
factors in the hit rate + the false alarm rate = measure of sensitivity, how good r ppl at discriminating the signal from noise
vigilance decrement def + description
task performance declines w/ time
1. occurs w/ in 5-15 mins
2. can b explained by how demanding, stressful, + unengaging vigilance tasks r (self-reports)
neuroergonomics
application of neuroscience 2 ergonomics
how often does target typically appear in visual search research and why?
abt half the time, although this is v diff from real-life
1. this is bc when target is low prevalence (2% in security screening ?), ppl change search strategies 4 the worse bc they think they’re not gonna come across it
low prevalence effect def
when rare targets r missed at a disproportionately high rate
1. happens no matter what, is usually quite large n pronounced
how 2 reduce low prevalence effect:
- increase prevalence - not practical irl
- give false performance feedback (make them think they’re missing more than they r) –> makes them more vigilant, but often not possible 4 the task
- getting software 2 project false images onto luggage –> keeps ppl vigilant, works where a computer interface is involved (most effective when false high prevalence occurs in short bursts)
- give breaks
TAIC (NZ)
investigate accidents involving planes, ships, n other forms of non-road transport
human factors analysis + classification system (HFACS)
latent failures:
1. organisational influences
2. unsafe supervision
latent and/or active failures
1. preconditions 4 unsafe acts
active failures:
1. unsafe acts
errors def + types (HFACS)
generally unintentional, unsafe acts of operators
1. skill-based errors - know what they should b doing, but not following procedure properly
2. decision errors (e.g. exceeded ability)
violations def + types (HFACS)
know that this isn’t what u should b doing, unsafe acts of operators
1. routine - everyday actions, e.g. I don’t really Need 2 b doing this
2. exceptional - not just disregarding a small move, equivalent of doing doughnuts in ur plane
condition of operator (preconditions; HFACS)
- adverse mental states - temporary
- adverse physiological states - temporary
- physical/mental limitations - them just being incompetent ig
personnel factor (preconditions; HFACS)
- crew resource management - ur not working well as a team
- personal readiness - ur own competence
enviro factors (preconditions; HFACS)
- physical environment (not as important 4 indoor job)
- technological environment
unsafe supervision (HFACS)
- inadequate supervision - ur a bad mentor
- planned inappropriate operations - ur bad at planning
- failed 2 correct a known problem
- supervisory violations - doing stuff in ur supervision role u Know is wrong
resource management (HFACS)
- human resources
- monetary/budget resources - excessive cost-cutting
- equipment/facility resources (e.g. shitty aircraft)
organisational climate (HFACS)
- structure (e.g. delegation of authority)
- policies
- culture
organisational process (HFACS)
- operations - incentives, time pressure
- procedures (e.g. performance standards)
- oversight - how much ur looking over things, make sure it’s safe
Shappell et al., 2007 (looking @ uses of HFACS)
tend to attribute mistakes more 2 individuals than actual govt. or bigger structure (bias??)
1. found most accidents involved unsafe acts rather than organisational influences or unsafe supervision
automation def n description
execution by a machine agent of a function that was previously carried out by a human
1. not binary, exists on a continuum
2. leaves humans to do difficult tasks that can’t b automated
how reliable is automation usually?
most of the time automation is pretty reliable (80-90%), but we may think of it as 100% reliable + trust it blindly → more error prone than not having automation available
what level of automation should be used for easy vs difficult tasks?
- easy task 4 operator, lower automation
- harder task, increase automation
info acquisition n automation (Parasuraman et al., 2000)
info acquisition - equivalent 2 human sensory processes
1. recommended relatively high automation here
info analysis (Parasuraman et al., 2000)
working memory, inferential processes - making predictions, identifying trends
1. recommended relatively high automation
decision selection (Parasuraman et al., 2000)
equivalent 2 human decision making, AI just needs info acquired n info analysis 2 come to a decision
1. 4 high risk task, low automation
2. 4 low risk task, high automation
action implementation (Parasuraman et al., 2000)
replaces human voice, hand movements etc
1. moderate degree of automation, lower automation if there’s high risk
Rasmussen, 1997
competing pressures:
1. economic failure (unsustainable costs)
2. unacceptable workload (unsustainable work requirements)
want 2 stay in space of possibilities - safe n sustainable work
work teams
2 or more ppl working together on a collective goal
task work
any actions or behaviours that ppl do that r directly related 2 a specific task, both teams n individuals can do taskwork
teamwork
behaviours inherent 2 the team (how the team works 2gether), can only b undertaken by the team
framework 4 understanding teamwork
input (e.g. u + ur teammates) → processes (can be behavioural, cognitive, emotional - how does team work 2gether) → output (e.g. group presentation)
3 types of teamwork processes r:
- transition phase
- action phase
- interpersonal
transition phase (teamwork)
evaluating n planning work that team will do. 3 key aspects:
1. mission analysis
2. goal specification
3. strategy formulation
diff types of strategy formulation (teamwork)
- deliberate planning - plan A
- contingency planning - plan B
- reactive strategy adjustment - if ur plan doesn’t work, have 2 decide if ur gonna change strategies or abort task altogether