Please enable JavaScript.
Coggle requires JavaScript to display documents.
INTERACTIVE SYSTEM DESIGN & EVALUATION
Errors (NORMAN'S ACTION…
INTERACTIVE SYSTEM DESIGN & EVALUATION
- Errors
-
- NORMAN'S ACTION THEORY
Donald A. Norman (1981) Categorization of Action Slips. Psych Rev, 88, 1-15.
ACTION SCHEMAS
- Schema theory extended to actions
- We learn combinations of actions for common situations
- Sequence of steps
Hierarchical organization of steps within steps
- Open slots for objects (etc) in particular categories filled in by the objects that turn up in context
- Controversial question in Psychology: Does structure of action sequence play casual role in guiding action or does it emerge from learned contingent behaviour?
- Normans analysis of slips based on theory of actions
- Action schema: action sequences to achieve goals
- Like other concepts, memories, etc, activated by activation of related words , concepts, memories, etc
- Triggered when active enough
- Goals } Trigger
- Perceptions } Conditions
- Thoughts in working memory } for Action Schema
- Recent perceptions or related concepts active
- Activation can decrease so necessary action not triggered
- MISTAKES (slide 28)
wrong intention, wrong action
- EXECUTION SLIPS v EVALUATION SLIPS (slide 29)
- CAPTURE SLIP (slide 30)
Automatic activation of a well-learned routine that overrides the current intended activity
e.g. intended to pick something up on the way home but went straight home
- DOUBLE CAPTURE SLIP (slide 31)
Unintended activation of a related strong action schema
Intended to take of shoes but took off socks too
- OMISSION SLIP (Slide 32)
Due to interruptions
e.g. forgot to put flour in
- LOSS OF ACTIVATION SLIP (Slide 33)
Intention forgotten - not active enough to trigger action specification
e..g. got side tracked and forgot
- DESCRIPTION SLIP (Slide 34)
Incomplete or ambiguous specification of intention that is similar to familiar intention
- ASSOCIATION ACTIVATION SLIP (Slide 35)
Activation of similar but incorrect schemas
e.g. after tapping phone screeen. tapped desktop screen
- PERCEPTUAL SLIP (slide 36)
e.g. I added grapes instead of olives
- REVERSE SCHEMA SLIP (slide 37)
Reverse action schema overwrites forward action schema
e.g. got correct fare out, moments later put coins back in
- REPETITION OF ACTION SLIP (Slide 38)
Repetion of an correctly performed action
e.g. nurse repeated radiation therapy 3 times
- CROSS-TALK SLIP (CONCURRENT) (Slide 39)
Action components are exchanged between two or more correctly performed concurrent actions
e.g. started writing in english, after interruption continued in ducth
CROSS-TALK SLIP (SEQUENTIAL) (Slide 40)
Action components are exchanged between two or more sequential actions
e.g. finished on the phone, went to introduce myself and said 'Smith speakin'
HUMAN FACTORS & ERRORS
WHY STUDY HUMAN ERROR?
- Practically
- to avoid catastrophic accidents
- to avoid annoying missteps
- Theoretically
- to predict when one will occur
- Cognitive science
- its the same system that produces correct behaviour most of the time
HUMAN FACTORS
- Key objectives of Human Factors is to design systems that
- people can use
- increase efficiency and performance
- minimise the risk of errors
-
CLASSIFYING HUMAN ERROR
- WHY CLASIFY?
- Understand why different types of errors happen
- Understand how to design them out
CLASSIFICATION SCHEMES
- Various classification schemes exist
- Different scales
- Different pupose
- Make sense of human behaviour
- Make sense of accidents and system failures
- Design to minimize errors
- Discrete action classifications
- Information processing classifications
- Root cause classifications
-
-
ERRORS v VIOLATIONS
- VIOLATION
- Deliberate violation of a rule, procedure or norm
- Routine, situational and exceptional violations
- SLIPS & LAPSES
- Plan is good
- Action is not according to the plan
- Slip(incorrect action), lapse (action forgotten)
- MISTAKE
- Plan is not correct for reaching the goal
- Defficient, wrong, clumsy or dangerous plan
DEALING WITH HUMAN ERROR
SPEED-ERROR TRADEOFFS
- The faster you go, the more likely you are to make mistakes..
- Partly under consciouscontrol
- Subject to incentives.. and laziness
- You'll gradually get quicker, until you make a mistake, then go slower and speed up again
- Designer need to anticipate this
- Human error is inevitable
- Consequences and liklihood can be reduced by
- better recruitment and selection
- training
- better design of equipment procedures and work environment
MURPHY'S LAW
- If theres a wrong way to do something, someone will do it that way
- Design principle: design things so there ISN'tT a wrong way to do something
-
INTERRUPTIONS
- Interruptions are a major source of errors
- Forget to return to task
- Forget where in taks
- Strategies
- Clear signals of where in sequence
*Interlocks to force people to complete task before doing anything else
- Successful lab experiments with timed lockouts to force people to pause to remember what theyre doing
BLAME
- 1st blame only the individual
- Tendency to view error at the operator level
- BUT... Other people involved in the design and operation of the sytem can make errors
- Should therefore consider the entire system
- badly designed or faulty equipment
- poor management practices
- inaccurate or incompletr procedures
- inadequate or inappropriate training
WHY BLAME OPERATOR?
- Shealey (1979) suggest several reasons for blaming operator making mistake
- It is human nature to apportion blame to someone else
- Legal system is geared to apportioning blame
- Easier for management to blame the worker than other aspects
- It is in the interests of the company to blame worker rather than admit deficiencies in their procedures, product or system
ATTRIBUTION THEORY
- We look for causes, and attribute responsibility for what happens, especially when things go wrong
- Attribution Theory is branch of psychology concerned with understanding how people attribute cause or responsibility for actions or events
- Consistent finding that subjects in experiments over-estimate how much of variation in what happens is due to people rather than situations
WHO OR WHAT TO BLAME?
- Beware of bias in attributing blame!
- Natural to blame action that immediately cause the appearance of a problem, hence person who does it
- Focus of attention
- Obvious causal chain
- Last chance to do something else
- NOT necessarily real source of problem
- Human error may be partly or entirely
- Poor system design
- Poor procedures
- Inappropriate use in different environment
- which
- Makes sucessful performance too difficult, or
- Sets up conditions in which people will make errors,
- Results in errors having bad consequences
ACCIDENTS
- A key objective of HF is to reduce accidents and improve safety
- Difficult to define 'accident'
- without apparent reason
- mishap
- unexpected
- chance
- natural event
-
-
-