It’s not your fault - bad user experience is the fault of the designer.

Design Principals

  • Affordances - “is for” Perceived and actual properties of an object. make appropriate actions perceptible. a chair is for sitting. materials, textures.
    • https://www.interaction-design.org/literature/topics/affordances - Affordances are an object’s properties that show the possible actions users can take with it, thereby suggesting how they may interact with that object. For instance, a button can look as if it needs to be turned or pushed.
  • Conceptual Models - how people perceive something operates. Finding meaning in the events around us.
    • mapping of what you want to do to what appears possible.
    • a good conceptual model allows us to predict the effects of our actions.
    • without a good model we do operations as we are told to do them without fully appreciating why or what to expect, using rote knowledge.
    • the conceptual model can be very simple we only need to associate the relationship between the controls and outcomes.
    • the conceptual model must be easily discernible from the controls
    • The design model is the designers conceptual model, the user’s conceptual model is a mental model developed through the visible controls and interaction with the system.
      • Ideally the design and user models are the same, but a poor design can lead the user astray.
    • the visible part of the device is the system image.
  • Visibility - the correct parts must be visible and they must convey the correct message.
  • Mapping - The relationship between two things
    • the mapping between controls (and how they move) and the results
    • natural mappings take advantage of physical analogies and cultural standards, “common sense”
      • spatial analogies: move a control left to go left, up to go up, down to go down. orient a control to what it controls, light switch on the left turns on/off lights to the left
      • turn a dial clockwise for more
        • slide a slider right for more, or to direct something to the right (stereo left/right balance)
        • flip a switch up to turn a light on
        • steering wheel - turn clockwise (right from the top) to turn right, counter-clockwise (left from the top) to turn left
      • additive dimension, louder, brighter, longer, and darker show incremental increase
      • substitutive dimensions: pitch, taste, color, location, there is no natural concept of more or less
      • check a box to indicate affirmation
    • one or more controls per function is easier, less controls than functions adds difficulty
      • radio preset channel buttons + a tuner knob and channel frequency indicator (slide or digital)
        • note setting a preset is harder, for example press the button for less than 3 seconds to go to that channel, more than 3 seconds to make the current channel that preset value
  • Feedback - send back to the user what action has been done or what result has been accomplished. - cause and effect, react to a user action.
    • tactile - push a button
    • audible - clicks, tones, beeps, verbal responses
  • Constraints - constrain the choices a user can make

The power of observations - observe how people act, interact with objects.

  • Design - things are designed for people.
    • non-functional requirements:
    • manufacturing - how will it be produced, how much will it cost, can it be mass produced
    • maintenance - is it easy to repair
    • aesthetics - does it look good
    • usability - is it easy, pleasant to use?
    • safety
  • Mental Models - are conceptual models of how we perceive things to work based on our knowledge and perception.
    • often knowledge is fragmentary (you don’t know everything)
    • the mental model can only be as good as the information used to form it. If the design doesn’t inform the model then it is likely to be wrong. In the absence of information people use imagination to fill in the missing details.
    • people frequently misconstrue coincidence as cause and effect.
    • learned helplessness - failures lead people to blame themselves
    • taught helplessness -

Psychology of Everyday Things

false assessment - once we have an explanation for discrepant events (right or wrong) we are complacent.

Naive Theories - non-scientific sound sensible/plausible but may be wrong

  • Seven stages of Action pg 47
    • perceive the state of the world
    • interpret the perception
    • evaluate the interpretations
    • goals
    • intent to act
    • sequence f actions
    • execution of actions

to get something done you must start with a notion of what is wanted - the goal. next you act to accomplish the goal. next you asses if the goal was met.

  • goal is what you want, not how to get it. vague
  • intentions are the statements of what is to be done. an intention to act so as to achieve the goal
  • plan a sequence of actions to do
  • execution of the actions in proper sequence

I don’t like the word intention for how it is being used. definition: intention - a thing intended; an aim or plan. synonyms: aim, purpose, intent, objective, object, goal, target, end, design, plan, scheme, resolve, resolution, determination, wish, desire, ambition, idea, dream, aspiration

would “solution” make more sense - goal -> solution -> action plan -> execute plan -> evaluate

goal - I need more light intention - I need to turn on the lamp.

To me goal and intention seems like two levels of goals, an abstract or high level goal and concrete goal - the light is too simplistic. or maybe sometimes people start at the intention level as the goal.

data-driven behavior - responding to events versus starting with a goal.

Gulfs of execution and evaluation

difficulty with mapping and feedback. the relationship between the actions, controls and outcome. a gulf is a difficulty in deriving the relationship between the mental intentions and interpretations and the physical actions and states.

gulfs separate mental states from physical ones. they each reflect an aspect of the distance between mental representations of the person and the physical components and states of the environment.

gulf of execution - does the system provide actions that correspond to the intentions of the person? Is it clear what actions must be performed to meet the intention? gulf of evaluation - the amount of effort to or difficulty in interpreting the physical state of the system and if the intentions are met.

The seven stages of action as design aids

provide a checklist of questions to ask to ensure the gulfs are bridged (minimize gulfs)

each stage of action has its own design strategies.

  • can the user tell what actions are possible
  • determine mapping from intention to action
  • perform an action
  • see if the system is in the desired state (before and after an action)
  • interpret the system state desired or otherwise

Knowledge in the head and in the world

discrepancy between the precision of behavior and imprecision of knowledge - a typist can type correctly, but most couldn’t describe the location of letters on the keyboard.

not all knowledge must be in the head, it can be distributed between the head, the world and constraints in the world. (internal plus external)

  • information in the world
  • great precision is not required
  • natural constraints are present
  • cultural constraints are present

two kinds of knowledge

  • knowledge “of” - declarative - facts and rules, easy to write and teach. a red light means stop.
  • knowledge “how” - procedural - how to perform actions - hard to teach or write down - taught through demonstration and learned by practice - play an instrument - procedural knowledge is primarily subconscious.

knowledge by delta - we don’t need to know every detail of things, just the differences that distinguesh choices. Size and color of coins.

the structure of mental memory

  • short term - 5-7 items immediately recallable - can be quickly lost when distracted
  • long term - memory of the past - it takes time to store and retrieve - how experiences are stored - not exact recall but interpreted through our existing memory/knowledge/understanding/experience.
  • three categories: ** arbitrary knowledge - facts - no back story ** meaningful relationships - facts+ (emotion?) sensible structure - makes sense (can be made to make sense), fit into mental model - interpretive ** through explanation - can be derived from mental model - understanding

knowledge in the world - external memory

  • out of sight, out of mind - specific to time and place
  • reminders ** rehearsal - keep in your mind by thinking about it ** notes
  • two aspects ** signal - string on the finger ** message - what to remember

trade off between knowledge in the head and world

retrieve-ability learning efficiency of use first time aesthetics

knowing what to do

the difficulty of dealing with novel situations is directly related to the number of possibilities. what are the controls and what can be done how do we deal with novel situations

  • information in the head ** transfer of knowledge from using a similar object ** instruction/taught
  • information in the world ** the design ** affordances - suggest a range of possibilities ** natural/physical constraints - limit the possibilities

there are four types of constraints

  • physical -
  • semantic - meaning in the situation - knowledge of the situation and the world.
  • cultural - vary by location, change/created over time. social rules
  • logical - natural mappings

applying affordances and constraints

switches

  • group related together
  • use shapes to distinguish
  • physical layout to match what they control ** vertical or horizontal rows of light switches don’t map to the lights ** a room “plan” type layout with the switches laid out where the lights are in the room

visibility and feedback

  • visibility - what parts move, how do they move, where do you pull/push/slide/turn.

  • feedback - what signifies a state change, what is decorative

  • visibility versus aesthetics - hiding controls for aesthetic effect.
  • where am I in a sequence of steps - VCR programming - gulf of execution

  • did the steps produce what I intended.

  • sound for visibility ** beeping for button presses (informative and annoying)
  • auditory icons

to Err is human

  • slip - subconscious mis-action - pressing the wrong key typing - these are typically small actions - execution phase of a plan. usually easy to detect and correct. ** happen from lack of attention, multi-tasking. ** six categories: capture, description, data-driven, associative activation, lack of activation, mode. ** capture - a frequently done activity is done instead of the intended activity. two different actions have a similar starting sequence and you switch from the intended, typically less used, to the more used. ** description - the internal description of an action is not precise and you do the action out of context. the “right” action on the wrong object. putting your coffee in the refrigerator instead of the microwave to heat it. flipping the wrong light switch. ** data driven - mixing up data - if your adding a series of numbers and someone in a different conversation says a number and you add it in. giving your old phone number instead of your new one. ** associative - internal thoughts interfering in actions. ** loss of activation - forget why you went into the kitchen. ** mode errors - when a control has more than one function “modes of operation”

slips are usually easy to detect because the result doesn’t match the goal, but if there is no feedback it may not be possible to know.

level problem - goal or intention versus lower level actions.

error correction seems to start at the lowest level and works out to higher level. This new code is broken, so start looking at the new code, but maybe the tester was testing on the wrong server or deployed the wrong version.

design lessons from slips

  • prevention ** asking for confirmation before deleting, although it usually is more annoying than helpful and you don’t realize that you selected the wrong file until later. confirmation of the action versus that you have the right file. having the trash can is much better because you can easily recover by pulling the file out of the trash.

  • detection and correction

  • mistake - choosing the wrong goal - conscious deliberations - thinking your appointment is tomorrow when it was today - can be catastrophic. can be hard to detect because the execution was correct, but done for the wrong reason, over generalize a situation - jump to the wrong conclusion. ** poor decision, misclassified situation, fail to take all relevant factors into account. non-rational thought. ** rely on past experience/memory instead of systematic analysis. ** human thought is not pure logic - jumps and unfounded connections ** memory - schema theory, frame theory, semantic networks * there is logic/order in individual structures *** memory is associative *** deductive thought applies the information from one schema to another schema to deduce the properties of it. *** default values and exceptions *** connectionist approach - neural nets ** discrepant memories (out of the ordinary) carry equal weight to the “normal” ones that happen more frequently - bias behavior to them.

  • wide and deep structures - chess moves
  • shallow structures - may have many top level options, but no deep associations - selecting an ice cream: flavor, amount, container, topping
  • narrow structures - sequential, very limited variety - recipe

####conscious vs. unconscious thought

  • unconscious thought is fast, it uses experience of “regular” or generalized information appropriate to the situation, things you do everyday, biased to the normal.

  • conscious thought is slow and laborious, takes different options into account, ponders the situation and rationalizes the explanation. limited by short term memory ** short term memory - 5-6 “items”, can be overcome by organizing individual items into structures, using subconscious structures, understanding and explanations, more can be reasoned about. *** mistakes are made by mismatch, applying the wrong organization structure, understanding or explanation to the wrong situation. this leads to “rare” situations being interpreted as either common or unique resulting in error. ** explaining away errors because they are the rare situation is common.

  • social pressure has a strong influence on behavior. ** threat of punishment ** risk and reward ** time and economic pressures - taking a short-cut in proper procedures to save time or get back on schedule or to avoid extra costs.

  • designs can make it easy to err, but difficult or impossible to detect or recover. ** it should not be possible for one person with one simple error to cause catastrophic failure.

  • design to minimize cause of errors and easy to recover. ** easy to undo an action or hard to do actions that can’t be undone. ** easy to discover errors and correct them ** think of errors as approximations in actions to complete a task, as the normal way of doing things. ** don’t punish a user for making mistakes and don’t take offense, most of all don’t ignore it. ** normal behavior isn’t always accurate.

how to deal with errors

  • warning signals - car beeps when you open the door with the engine running - can’t distinguish deliberate actions from erroneous ones, frequently ignored such that when it is correctly signaling you, you will ignore it as usual.

  • forcing function - can’t lock keys in car, you must lock the door from the outside. ** it isn’t easy to force people to do something they don’t want to do, ex: seat-belts. ** if you use a forcing function it must be reliable and distinguish legitimate cases from illegitimate. ** not every case affords forcing functions. ** interlock - forces proper sequence - ex: microwave can’t be turned on with the door open. ** lockin - prevent premature stopping of a process. ex: you have unsaved files or unsaved data on a form dialog. ** lockout -

  • put the required knowledge in the world, don’t require all the knowledge to be in the head. ** allow for more efficient operation when the user has learned the operations.
  • use the power of natural and artificial constraints, physical, logical, semantic, and cultural. ** use forcing functions and natural mappings.
  • narrow the gulfs of execution and evaluation. ** make things visible, operations readily available, outcomes apparent.
  • make the system state easily available and consistent with the goals, intentions and expectations of the user.

Design Challenge

natural design process - good design evolves - design -> test -> problems -> repeat.

  • time constraints - lots of iterations are slow
  • feedback mechanism required

design is a balance of aesthetics, usability, economics and manufacturing.

why designers go astray
  • reward structure is loaded towards aesthetics.
  • designers aren’t typical users
    • big difference in the expertise required to be a designer versus a user.
    • designers can use their own device with knowlege in the head where as a novice user requires knowlegde in the world.
    • designers must work with real users from the beginning before any decisions are made.
  • the user/target audience isn’t always the client/buyer
    • builders and landlords buy appliances and equipment based on cost and aesthetics more than usability
    • the government frequently makes decisions based on cost

infinite gulf of evaluation - levels of bureaucracy, feedback/desires filtered through multiple levels.

complexity of the design process

design is the successive application of constraints until only unique product is left

  • The variety of possible solutions is enormous.
    • the range of expression granted the designer is vast.

  • the level and number of details to account for is astounding.

  • varied and obscure uses - “use a pen to set the clock in a car” does the designer of the pen know this requirement?

  • designing for special people
    • there is no such thing as the “average person” - physical anthropometry
    • make things adjustable to accommodate different users
    • left handed versus right handed - how can you adjust for mirror image use?
    • various capabilities: vision, hearing etc.
  • selective attention: problem of focus
    • limits to what can be focused on
    • attend to immediate problems and ignore the rest
    • designers must consider the problems of focus in the design.
      • use of forcing functions for safety
      • proper feedback to avert evaluation problems
two deadly temptations for the designer
  • creeping featurism
    • tendency to keep adding new features
      • require more controls, more feedback to display, or instructions to know
      • adding more complexity
    • if adding a feature can’t be avoided, can features be organized better? modularity.
  • worshipping false images
    • features for showing off, not using - fancy copiers to make you look cool to clients.
    • features for the appearance of technical sophistication.
  • computer systems
    • programmers should not be responsible for the computer’s interactions with the user, it is not their expertise.
    • many programs are too abstract for users
    • how to do things wrong
      • make things invisible, widen the gulf of execution, give no feedback
      • be arbitrary - non-obvious command names (or key combos)
      • be inconsistent - change the way things are done from screen to screen
      • use idiosyncratic language - indecipherable error messages
      • make operations dangerous - lose work by clicking the wrong button.

User Centered Design

  • Make it easy for the user to know what actions are possible at any time
  • Make things visible
    • the conceptual model
    • alternative actions
    • results of actions
  • Make it easy to evaluate the current state of the system
  • Use natural mappings
    • between intentions and the possible actions
    • between actions and results
    • what is visible and how it relates to system state

in other words

  • The user can figure out what to do
  • The user can tell what is going on
Seven Principles for Transforming Difficult Tasks into Simple Ones
  • Use knowledge in the world and in the head
    • people learn better/feel comfortable when knowledge is externally available.
      • explicit or derived through constraints
    • knowledge in the world is only useful if it is natural/easily mapped to actions and outcomes.
    • allow users to use knowledge in the head to make things faster.
      • (key-chords versus navigating through menus)
      • the design shouldn’t impede a user that knows what they want.
    • good conceptual models make things easier.
      • natural - works the way the user thinks it should.
      • consistent - similar goals use similar actions
      • three models
        • design - designers conceptualization of how it works
        • user - users self explanation of how it works
        • system image - the user facing interface to the system
        • ideally the user model = design model, for this to happen the system image must communicate the design image to the user
        • the user aquires all knowledge of the system through the system image.
      • manuals/help - if you need them, something is wrong.
  • Simplify the Structure of tasks
    • people only have limited resources for performing tasks
      • memory
        • short term - limited space
        • long term - needs to be triggered (hints), slower access
          • sensible versus rote
      • concurrency
        • people can only concentrate on one thing at a time
          • lose your place if distracted, where was I?
        • context switching can be expensive
          • help the user with feed back if they get distracted
      • provide mental aids - blinking cursor, highlight the field I’m on, tell me how many characters I’ve typed and what is the maximum. identify typos, validate fields
      • make the invisible visible
    • automate tasks so users don’t have to do tedious things
      • don’t destroy the fun/pleasurable aspects of a process, maybe?
      • don’t take away control
    • make tasks easier for the user
      • velcro straps instead of shoe laces.
      • digital versus analog clock.
  • Make things visible for Execution and Evaluation
    • people make explanations for the things they see which may be wrong.

Once a satisfactory product has been achieved, further changes may be counterproductive, you have to know when to stop. pg 150

It is a lot easier to determine what is obvious after it has happened pg 129

technology changes quickly, people change slowly.

People are explanatory creatures - back story, mental models, interpretations pg 38

When ever the number of functions and required operations exceeds the number of controls, the design become arbitrary, unnatural, and complicated. pq. 31