section 3- generalisation & VB & schedule of R Flashcards
(60 cards)
1
Q
discrimination
A
- narrow stimulus control
- when a limit spectrum of stimuli occasion A response
2
Q
generalisation
A
- large spectrum of stimuli occasion certain responses
- critical for human to survive and thrive
2 types of generalisation
- stimulus generalisation
- response generalisation
3
Q
stimulus generalisation
A
- responding to antecedent stimuli that sharing certain aspects of the ORIGINAL SD
- the person responds to sth in the SAME way that resemble the original thing he learned
- a BROADENING of the spectrum of stimuli that occasion certain responses
- program stimulus generalisation: teach a person to respond in the same way to similar but not identical stimuli
- over-generalisation/inappropriate generalisation: emit a response that appropriate to some contexts in an inappropriate context
e. g. call all women ‘mommy’
e. g. prejudice
4
Q
response generalisation
response induction
A
- a person exhibits NOVEL responses that are FUNCTIONALLY equivalent to the trained target response
- intervention effects expand from a targeted response to a SIMILAR non-targeted response
- response induction = introduction (novel responses)
- enable to shape NEW behavior
- if the response/behavior change, then it’s response generalization
5
Q
plan for generalisation
A
- select target behaviors that will meet with NATURAL contingencies of R
- functional behavior
- relevance-of-behavior rule: only choose behaviors that generate R after intervention ceases
- 2 types of contingencies: naturally existing contingency, contrived contingency
- specify ALL desired variations of B & situations that it should/not occur after instructions ended
* *do this at the beginning of the planning stage**
6
Q
7 strategies to promote generalisation
A
CLEMING
- common stimuli
- ensure the same SD exists in both the instructional & generalization setting
- correct response is more likely to occur in the generalization setting if there’s lots of similarity between the instructional & generalization settings.
- loosely train
- alter NON-critical elements of the teaching settings in arbitrary ways –> the individual is LESS likely to too narrowly discriminate some non-critical elements –>non-critical elements are LESS likely to acquire exclusive control over the target response
- expand the heterogeneity of SD - exemplars/multiple exemplar training
- the MORE examples used in teaching, the better
- provide the person OPPORTUNITIES to respond correctly to multiple examples of ANTECEDENT stimuli
- provide multiple RESPONSE examples - mediation
- instruct OTHERS who will help maintain & generalize the newly acquired Bx
* * ETHICAL**- your ethical responsibility to collaborate with others to maintain the clients’ progress after your services terminate
- the CODE: cooperate with other professionals to serve your client effectively & appropriately
- in-discriminable contingencies
- contingencies when an individual is UNABLE to discriminate when his responses will be reinforced –> Bx continue at a HIGH rate
- make the contingencies UNCLEAR in generating setting
- intermittent schedules of R: start with continuous R (CRF), then move to intermittent R (INT)- all indiscriminable contingencies involve INT but not all INT are indiscriminable*
- delayed rewards: unclear what stimuli will produce R
- all indiscriminable contingencies involve INT but not all INT are indiscriminable*
- negative teaching examples
- teach individuals when it is NOT appropriate to display a certain B: settings, times, conditions- strengthen discrimination skills
- general case analysis/strategy
- ensure to teach ALL different STIMULUS & RESPONSE variations in generalisation, post-intervention environment
- help to learn SIMILARITIES & DIFFERENCES of stimuli WITHIN a stimulus class
- ensure to teach ALL different STIMULUS & RESPONSE variations in generalisation, post-intervention environment
7
Q
terminate successful interventions
A
- must SYSTEMATICALLY terminate successful interventions
- should access how INTRICATE/COMPLEXITY the intervention is, how QUICKLY the intervention produced the desired change, the AVAILABILITY of NATURAL contingencies of R for the newly acquired skill
- attempt to reduce the need to generalize FROM THE BEGINNING
- PROBE for generalization BEFORE, DURING, AFTER intervention
- mediators should have responsibility in the generalization process
8
Q
maintenance
A
- after remove the intervention, the extent that a particular response remains in the person’s REPERTOIRE OVER TIME
- program maintenance *
- use intermittent/variable R –> promote persistence under EXT
- use variable ratio/interval schedule of R (VR/VI)
9
Q
verbal B
A
- SKINNER created VB
- VB book published 1957
10
Q
private events
A
- inside the skin
- thoughts & feelings
- are behaviors
- only accessible to 1 individual, not 2 or more –> you are the only person who can feel it
11
Q
technical definition of VB
A
- B that is reinforced through the MEDIATION of OTHER’s Bx
- define by FUNCTION of the response, NOT topography
- vocal & non-vocal Bx
- any response FORM can be verbal: baby cry, gestures, writings etc
- involve SOCIAL INTERACTION between SPEAKER & LISTENER
- speaker: gain access to R, controls environment through the B of listener. SKINNER’s VB mainly concerned B of speaker
- listener: must learn how to R the speaker’s VB
- communication helps ppl get what they desire & avoid undesirable
12
Q
form & function of VB
A
- form: formal properties of language involve topographies
e. g. classify words as nouns, verbs, prepositions - function: EFFECTS of the response
13
Q
verbal operant
A
- the UNIT of VB
e. g. mands, tacts - MO/SD–>response–> consequence
- verbal repertoire: a set of verbal operants emitted by sb
- use verbal operant as the basis for language assessment*
- assess current effectiveness of each verbal operant
- start by mand repertoire
- FIRMLY establish repertoires for each verbal operant before move on to more complex ones
- VB assessment: VB-MAPP
- functional analysis of VB is ongoing
- must determine what the correct source of control should be and how the source can be established
14
Q
Skinner’s 6 elementary verbal operants
A
EMITTT
- echoic
- mand
- intraverbal
- tact
- textual
- transcription
15
Q
echoic
verbal imitation
A
- occurs when the speaker repeats the VB of another speaker
- controlled by the verbal SD: whatever the speaker said controls what the listener is going to say: the speaker says ‘echo’, then listener says ‘echo’
- the ability to echo is essential for identifying objects & actions
- produce GCSR, e.g. praise, attention
- verbal SD has 2 things in common with response:
- point-to-point correspondence: verbal stimulus & response match in beginning, middle, end
- formal similarity: controlling antecedent stimulus & response share same SENSE mode (e.g. both visual, auditory, tactile) & PHYSICALLY look exactly the same
16
Q
echoic training
A
- bring verbal responses under the FUNCTIONAL control of verbal SD that have point-to-point correspondence & formal similarity with the response
- GOAL: to enable the speaker to repeat your sounds –> eventually TRANSFER the response form to other, more advanced verbal operants
- teach echoics*
- shaping: teacher presents a vocal verbal stimulus, reinforces the student’s SUCCESSIVE APPROXIMATIONS towards the sample
- at beginning: choose sounds that SIMPLE & WITHIN his repertoire
- teach systematically:
- teach consonants after mastered all vowels
- oral motor imitation: build physical strength: blow on items, use stimuli etc
- prompt level: physical (e.g. hands on face), touch face, gradually shape mouth formations
17
Q
mand
A
- the speaker asks for what he needs or WANTS
- controlled by MO not SD
- occur due to a state of deprivation / aversive stimulation
- manding is reinforced by ATTAINING the manded items
- allows the LISTENER to infer what EO may be affecting the speaker
- mands often become STRONG forms of VB coz the specific R received
- mands are the FIRST verbal operants acquired by humans
18
Q
mand training
A
- FIRST verbal operant to teach
- bring verbal responses under the functional control of MOs
- initiate teach mand training*
- assess motivation: an MO needs to be in effect at the time of training
- the time when your client’s motivation is STRONG
- control & contrive the motivation
- make a list of potential motivators & reinforcers
- select the 1st few words to teach
- at beginning choose*
- items related to STRONG motivation, e.g. food, toy
- items you can CONTROL access
- items can be available for a SHORT period of time initially
- items are easy to DELIVER &/ REMOVE
- items that are CONSISTENTLY strong
- items the client prefers
- words involving a response form already in the client’s repertoire
- next, teach complex mands*
- mands with adj & prepositions
- increase the length of the mands
- mand for information, attention, ppl do things for you
19
Q
2 types of mands
A
- regular mand: can actually be reinforced
- extended mand: emit mands to objects /animals that CANNOT possible supply an appropriate reinforcing response
e. g. while drive on the freeway tell another driver ‘move your car’
a. superstitious mand: R sometimes occur incidentally.
b. magical mand: R NEVER occurred in the past; wishing. e.g. ‘I wish I had a million dollars’
20
Q
intraverbal
A
- speaker DIFFERENTIALLY responds to other ppl
- ANSWER A QUESTION
- a verbal SD evokes a verbal response that does NOT have point-to-point correspondence with the verbal stimulus
- allow a person to talk & think abt things that are NOT physically present
- produces GCSR
21
Q
intraverbal training
A
- bring verbal responses under the functional control of verbal SD that LACK point-to-point correspondence with the response
- prerequisites: has acquired 50 mands & tacts
- teach*
- use prompting, fading, chaining
- focus on what interests the learners & manipulate the EOs
- teach in the NATURAL environment as much as possible
- CONTINUE teach NEW mands & tacts while teach intraverbals
- at beginning*
- begin with fill in the blank activities using favorite songs, characters
- teach to give one’s name, gender etc. (e.g. social questions)
- use object & animal sounds
- WH questions
- teach systematically*
- teach in natural environment promote generalisation
- short sessions: 1-15 minutes
- conduct training at table, on floor, playground, etc.
22
Q
tact
A
- speaker names things & actions that the speaker has DIRECT contact with through any of the SENSE mode
- label the environment when the object, event, stimulus is PRESENT in your environment
- verbal response in the PRESENCE of the thing tacted
- if a person is naming something but the thing is NOT presented to the person –> the person CANNOT tact it
- controlled by NON-VERBAL SD
- produce GCSR
- tact extensions/extended tacts *
- not 1 name for 1 thing: many ways to describe the same thing
- a new stimulus being SIMILAR to another known stimulus may evoke a response LIKE the original stimulus
- 4 types SMMG
- solistic extension: poor use of language, substandard VB, slangs
- metaphorical extension: the novel stimulus shares SOME but not all of the features of the original stimulus. (e.g. his heart is as black as coal)
- metonymical extension: the novel stimulus shares NONE relevant features of the original stimulus, some irrelevant but related feature has acquired stimulus control
- generic extension: the novel stimulus shares ALL relevant features of the original stimulus = stimulus generalization
23
Q
tact training
A
- bring verbal responses under the functional control of NONVERBAL SD
- prerequisites: echoics, some labeling of vocabulary, 5-10 mands that occur without echoic prompts
- teach*
- similar to teach mands, with verbal stimulus ‘what is that’ ‘what do you see’ –> not a pure tact (i.e. tact with only the non-verbal SD)
- continue with mand training simultaneously with tact training sessions
- teach systematically
- teach with mand component: with MO for the items, (e.g. cookie), R with access to the item
- teach without mand component: without MO for the item.
24
Q
textual
A
- without any implications, the readers UNDERSTANDS what is being read
- read WRITING words
- involves intraverbal behavior & receptive language
- occurs when a verbal SD has point-to-point correspondence but NO formal similarity between the stimulus & response
- produce GCSR
25
transcription
- WRITING & SPELLING words spoken to you
- DICTATION
- occurs when a SPOKEN verbal SD controls a WRITTEN, typed, finger-spelled response
- has point-to-point correspondence but NO formal similarity between the stimulus & response
26
duplic & codic
Jack Michael 1982 article created 2 new strategies A
27
codic
- 3 defining features
1. response form is controlled by verbal SD
2. point to point correspondence
3. NO formal similarity
- codic in Skinner's elementary verbal operants
a. textual: SD is verbal & visual, response is speaking words
b. transcription: SD is verbal & spoken, response is writing, typing, finger-spelling
- other examples: Jane reads a Braille book (for blind ppl) out loud
28
duplic
- 3 defining features
1. response form is controlled by verbal SD
2. point to point correspondence
3. FORMAL similarity
- responses can be speaking, writing, signing, etc
- duplic in Skinner's elementary verbal operants:
a. echoic
- other examples: imitating sb else's signs
29
listener training
- often speaker & listener reside the same skin: listener behaves simultaneously as a speaker, happens during conversations
- a verbal episode requires a speaker & listener
- role of listener: a mediator of R & SD for the speaker's B
- listener behavior can be non-verbal, e.g. receptive language
30
autoclitic
- VB abt one's own VB
- a SECONDARY verbal operant where some ASPECT of a speaker's own VB functions as SD / MO for additional speaker VB
- modifies other forms of VB
- effects are very RAPID & occur in the emission of a SINGLE SENTENCE composed 2 levels of responding
e.g. 'don't look now, but I think I see Marty with his new girlfriend.'
31
schedule of R
definition:
- a rule that describes a contingency of R
- the environmental arrangements that determine conditions by which Bx will be reinforced
- continuous R (CRF)
- intermittent R (INT)
- EXT
* CRF & EXT are the boundaries of all schedules of R
32
```
continuous R (CRF)
(FR1)
```
- R for EVERY occurrence of the target B
- utilized for strengthening NOVEL Bx when first initiate for a new skill
e.g. after a person flips on the light switch, the light comes on
33
intermittent R (INT)
- between CRF & EXT
- R some not all occurrence of B
- to maintenance Bx that ALREADY been established
- help to FADE from artificial to NATURAL environment
e. g. lottery, slot machine 老虎机, find parking at the supermarket
- FR, VR, FI, VI
34
fixed ratio - FR
- a CERTAIN/constant number of occurrences of the behavior have to occur before 1 response produces R
* pattern of responding produced by FR*
- individual completes required responses with LITTLE hesitation
- POST-REINFORCEMENT PAUSE: when the individual does NOT respond for a certain time following R: large ratio = long duration
- FR often produce HIGH rates of responses coz quick responding produces FASTER rate of R: large ratio = high rate
graph:
high rate of responding --> post-R pause --> contingency met & R is provided
35
variable ratio - VR
- strongest basic schedule of INT
- varaibale=changing, variable criteria, average
- a number (average) of occurrences of the target B have to occur before 1 response is R
* pattern of responding produced by VR*
- CONSISTENT, STEADY rates of response
- NO post-R pause: may coz the absence of info when the next response will produce R: responding remains steady coz the next response may produce R.
- FAST rate of response
- large ratio = faster rate
graph:
high rate of responding --> contingency met & R provided
36
fixed interval - FI
- one basic schedule of INT
- a SPECIFIC amount of TIME elapses before a single correct response is R
e. g. FI5: R is delivered after the 1st CORRECT response that occurs after 5 mins
* pattern of responding produced by FI*
- post-R pause occurs ONLY during the early part of the interval
- FI SCALLOP: at the END of the interval, a GRADUALLY ACCELEBRATING rate of response toward the end of the interval--> round curve in graph
* rate of response*
- slow to moderate rate
- larger fixed interval = longer post-R pause
graph:
post-R pause --> increase in responding --> contingency met & R
37
variable interval - VR
- one basic schedule of INT
- a specific average amount of time elapses before a single correct response is R
e. g. VI12: R is delivered after the 1st correct response that occurs after an average of 12 mins
* pattern of responding produced by VI*
- CONSTANT, STABLE rate of response
- few hesitations between responses
* rate of response*
- low to moderate rate of response
- larger average interval = lower overall rate of response
graph
steady rate of responding --> contingency met & R
38
FR & FI
- both produce post-R pause
| - variable schedule: NO post-R pause
39
schedule thinning
| thinning INT
- gradually INCREASE ratio/duration
| e. g. CRF-->FR2/VR3
40
ratio strain
- a result of abrupt increases in RATIO (e.g. FR1-->FR20) requirements when moving from denser to thinner R schedule, e.g. non-compliance
- common behavioral characteristics: aggression, avoidance, etc.
*reduce ratio strain*: reduce ratio requirements
41
limited hold=limited time
| LH
- a restriction placed on an INTERVAL schedule requiring the primed (1st response following termination of the required interval) must occur within a specified span of time following that interval
- can be imposed on any type of schedule
- use LH to speed up the response rate
e.g. FR5 with LH2 mins: the person must complete the 5 tasks within 2 mins
42
FR & FI
- both produce post-R pause
| - variable schedule: NO post-R pause
43
schedule thinning
| thinning INT
- gradually INCREASE ratio/duration
| e. g. CRF-->FR2/VR3
44
ratio strain
- a result of abrupt increases in RATIO (e.g. FR1-->FR20) requirements when moving from denser to thinner R schedule, e.g. non-compliance
- common behavioral characteristics: aggression, avoidance, etc.
*reduce ratio strain*: reduce ratio requirements
45
limited hold=limited time
| LH
- a restriction placed on an INTERVAL schedule requiring the primed (1st response following termination of the required interval) must occur within a specified span of time following that interval
- can be imposed on any type of schedule
- use LH to speed up the response rate
e.g. FR5 with LH2 mins: the person must complete the 5 tasks within 2 mins
46
3 variations of basic INT
- schedule of differential R of rates of responding
- R is contingent on Bx occurring at HIGHER or LOWER rates than a specified criterion
- DRH
- DRD
- DRL
47
DRH
| differential R of HIGH rates of responding
- R for emitting Bx that are AT OR ABOVE a pre-established rate
- to INCREASE B that displays too INfrequently
48
1. concurrent schedules of R (conc)
- when 2 or more contingencies of R operate independently and SIMULTANEOUSLY for 2 or more Bx
- choice-making: make a choice in 2 concurrent Bx regarding 2 kinds of R schedule --> matching law--> choose B that has the highest rate of R
* matching law (Herrnstein, 1960s): organisms MATCH their responses according to the PROPORTION of PAYOFF during CHOICE situations
- B1/(B1+B2)=R1/(R1+R2)
e.g. if a B is R 60% of time in 1 situation & 40% in another, then B tends to occur 60% of the time in 1st situation & 40% in the 2nd.
49
2. multiple schedules of R (mult)
- present 2 or more basic R scheduels in alternating sequence for only 1 or more Bx, usually RANDOM
- basic schedules occur SUCCESSIVELY & independently
- 1 SD is correlated with each basic schedule & present as long as the schedule in effect
e.g. VR5 with teacher at school & VI2 with tutor at home
50
3. chained schedule of R (chain)
- 2 or more basic R scheduels occur SUCCESSIVELY & has 1 SD correlated with each independent schedule with 1 or more Bx
3 elements
- basic schedules occur in a SPECIFIC order
- B may be the same for all elements of the chain or different Bx may be required for different elements in the chain
- conditioned R for the 1st B in the chain is the presentation of the 2nd element and so on.
51
4. mixed schedules of R (mix)
- identical to multiple schedules (random order), except mixed schedule has NO SD correlated with independent schedules --> the individual is not sure what schedule is in effect at any given time
e. g. mix FR10 FI1: R sometimes occur after completing 10 responses, sometimes occur after 1-min interval from the preceding R -- random order
52
5. tandem schedules of R (tand)
- similar to chained schedule (successive, specified order) except for NO SD
- an unsignaled chain
53
6. alternative schedules of R (alt)
- R when the requirement of EITHER a ratio or interval schedule is met, regardless of which schedule requirements is met first
- EITHER / OR schedule
54
7. conjunctive schedules of R (conj)
- R when the completion of the response requirements for BOTH a ratio & interval schedule have been met
- BOTH schedule
55
adjunctive (附属的) Bx
| schedule-induced Bx
- Bx that are brought about by schedules of R during times when R is unlikely to be delivered
- time-filling or interim Bx
e.g. doodling, smoking, drinking
56
schedules of P
- similar to R schedules
- continuous schedule of P (FR1) max. the suppressive effect
* ethics*
- should recommend R rather than P whenever possible
- include R for alternative B when using P
- always recommend least restrictive procedures
57
DRD
- R when # of responses in a specified time period is LESS than or equal to a prescribed limit
- help to decrease B that occurs too frequently
- NOT to eliminate entirely
e. g. R when taking only 10 bites of food in 5 minutes
58
DRL
- R ONLY IF the B occurs following a specific period of TIME during which it did NOT occur or since the last time it occurred
- interresponse time (IRT): duration of time between 2 responses
- IRT & rate of response are FUNCTIONALLY related: longer IRT--> lower rate
- increase IRT --> decrease rate of B
- decrease but NOT eliminate B
e. g. R only upon talking a bite of food after 10 secs pause
59
progressive schedules of R
- systematically THIN each SUCCESSIVE R opportunity INDEPENDENT of B
- use for reinforcer assessment to identify Rs that will maintain treatment effects across increasing schedule requirements
* typically thin progressive schedule to the breaking point when NO responding
- use for intervention: systematically INCREASE the requirement for R
* use to measure strength, potency, effectiveness of scheduled Rs
- there's direct relation between how hard an organism will work for access to an object & the potency of the reinforcer
60
7 compound schedule of R
combine CRF, INT, (DRH, DRD, DRL), EXT
CMCMTAC
1. concurrent schedules of R (conc)
2. multiple schedules of R (mult)
3. chained schedule of R (chain)
4. mixed schedule of R (mix)
5. tandem schedule of R (tand)
6. alternative schedule of R (alt)
7. conjunctive schedule of R (conj)
