Transcript J&P Chapter 4 - People Server at UNCW

J&P Chapter 4
Defining Response Class
Strategies of Defining Response
Classes
Unit of Analysis
• The constituent part of the whole
phenomenon that serves as a basis for
experimental study
• Unit of analysis vs. unit of measurement
• Child sat at desk (unit of analysis) for a
duration of x minutes (unit of
measurement)
Skinner’s contribution
• Stimuli defined not just in terms of physical
properties but also by the correlation with
a particular class of responses
• Responses defined NOT by topography
but rather as a class of events controlled
by particular stimuli
Bed making
• Unit of analysis that is itself a relation
between responses and stimuli
– Series of movements
– OR
– Whatever responses result in a bed that is
made to a certain standard
Origins of Response Class
• Behavior occurs within an environmental
context
• Antecedent and consequent events have
either differential or no effect
– Genetic endowment – sweet taste
– Stimulus pairing – stimuli paired with sweet
taste
– NO effect – color of walls in room
What do stimuli do?
• If a stimulus functions to change behavior:
– Organize behavior into response classes
• Class of respondents (reflexes)
– Creating conditioned reflexes = classical, respondent, or
Pavlovian conditioning
– Respondent classes (eye blink) & stimulus classes (all
stimuli that elicit eye blink)
• Classes of operants (individuals responses that
share a particular effect on the environment
Respondent
S–R
Free Operant (2 term contingency)
R–S
Discriminated Operant
(3 term contingency)
S–R–S
Respondent
Operant
Fundamental Units of Analysis
•
•
•
•
Respondent classes
Free operant classes
Discriminated operant classes
These units:
– Smallest bits of integrated behavior
– Not necessary to move to a nonbehavioral level of
analysis
– Not defined on logical grounds – rather defined based
on functional relationship with nature
Functional vs. Topographical Units
of Analysis
Functional Response Classes
• Form or topography of response may vary
• Class includes those reponses whose
occurrence depends upon particular
classes of stimuli that either preceded or
follow responses
Topographical Response Classes
• Specify requirements for form or
topography
• Members of such a response class may
include only a subset of a broader
functional response class
• Example: door opening
Define Self Injurious Behavior
(pgs. 71 & 72)
• Functional definition
• Topographical definition
Consequences of Definitional
Strategies
• Determines what aspects of behavior are
observed and recorded and are then
available for DECISIONS about how the
experiment should be conducted and what
INTERPRETATIONS are warranted
Effects of including different functional response
classes in a defined class
Defined Class
Class A
Class B
Baseline
Intervention
Goals of Definitional Strategies
• Defined behavior should suit the needs of
the experimental questions
• Meet needs for experimental control
• Definitions consequences for
measurement
• Craft definition so that it selects a class of
responses in the subject’s repertoire that
share common functional relations with the
environment.
Tactics of Defining Response
Classes
Functional Response Class
Definitions
• Properly developed = miniature experiment
• Consider the proposed response class in
colloquial terms
• Speculate on events that may preceded or
follow response class members
• Try out possible definitions and see how
they perform
• Example STEALING
Problems and Solutions
• Problem with knowing whether or not
permission given in the stealing example
– Modify research environment to ease
definitional task
• Problem: more than one response class
– Risk increased variability in data
Topographical Response Class
Definitions
• Press the easy button
Variations and Refinements
• Temporal dimensions
• Response products
• Group definitions
Selecting a Unit of Analysis
• Specificity: size of the unit of analysis
– More molar units of analysis may obscure
important changes in responding
• Sensitivity: potential to be influenced by
independent variables
– If under control of extraneous variables, may
not be at all susceptible to influence of
independent variable
Labeling Response Classes
• Taxonomic language of behavior
• How responses classes are defined?
• Generality? Accrue to relations studied,
not labels
• How are researchers to communicate?
What can be done?
• Use sound functional definitional practices
• Be aware of the problem when considering
ways to refer to a response class
• Choose labels that emphasize the relaiton
of the movement to the environment
• No ultimate solution
Operational Definitions
• Not another way of defining behavior
• Operationism
• Relevant to problem of labeling response
classes.