Chapter 10: Conditioned Reinforcement

1
Chapter 10Conditioned Reinforcement
2
Conditioned Reinforcement

• Conditioned reinforcement occurs when behavior is
  strengthened by events that have an effect
  because of a conditioning history. The critical
  aspect of this history involves a correspondence
  between an arbitrary event and a reinforcer.
• Once the arbitrary event increases the frequency
  of an operant, it is called a conditioned
  reinforcer.

3
Chain Schedules and Conditioned Reinforcement

• One way to investigate conditioned reinforcement
  is to construct sequences of behavior.
• A chain schedule of reinforcement involves two
  or more simple schedules (CRF, FI, VI, FR, etc.)
  each of which is presented sequentially and is
  signaled by an arbitrary stimulus (each has its
  own SD). Only the final or terminal link in this
  chain results in primary reinforcement.

4
Multiple Stimulus Functions

• In a sequence of two schedules such as an FR150
  FI 120 seconds in which distinct SDs do not
  signal the different components would be called
  an tandem schedule but it is the same as an
  unsignalled chain.
• In equivalent tandem vs. chain schedules,
  performances will differ on the chain showing
  that the distinct signals serve as both SDs and
  conditioned reinforcers.

5
Homogeneous and Heterogeneous Chains

• Operant chains are classified as homogeneous when
  the topography or form of response is similar in
  each component, i.e., a similar response
  requirement is in effect in all components.
• A heterogeneous chain requires different
  responses in each link.
• Chain schedules illustrate how sequences of
  behavior are maintained by conditioned
  reinforcement.

6
Determinants of Conditioned Reinforcement-Strength
of Conditioned Reinforcement

• Frequency of Primary Reinforcement
• Variability of Primary Reinforcement
• Establishing Operations
• Delay to Primary Reinforcement

7
Establishing Conditioned Reinforcement

• New-response method
• Established-response method
• Pairing, Discrimination and Conditioned
  Reinforcement
• S-S account of conditioned reinforcement
• Discriminative stimulus account of conditioned
  reinforcement

8
Information and Conditioned Reinforcement

• A mixed schedule of reinforcement
• Will subjects respond to make observing responses
  to reveal good news, stimuli correlated with
  reinforcement as well as bad news, stimuli
  correlated with extinction?
• Good news will function as a conditioned
  reinforcer bad news will not

9
Delay Reduction and Conditioned Reinforcement

• Delay-reduction hypothesis
• Stimuli closer in time to positive reinforcement,
  or further in time from an aversive event, are
  more effective conditioned reinforcers.
• Stimuli that signal no reduction in time to
  reinforcement (S?) or no period of safety from an
  aversive event (Save) do not function as
  conditioned reinforcement.

10
Concurrent-Chains Schedules of Reinforcement

• To assess the effects of delay, organisms must be
  able to choose between stimuli associated with
  different reductions in time to primary
  reinforcement.
• Use a two-key concurrent-chains procedure with
  equivalent initial links but different lengths of
  terminal links.

11
Generalized Conditioned Reinforcement

• A generalized conditioned reinforcer is any event
  or stimulus that is associated with or
  exchangeable for, many sources of primary
  reinforcement. Generalized reinforcement does not
  depend on deprivation or satiation for any
  specific reinforcer.
• Generalized social reinforcement for human
  behavior approval, attention, affection, praise

12
Tokens, Money and Generalized Reinforcement

• Other conditioned reinforcers are economic since
  they are exchangeable for goods and services.
  Probably the most important such reinforcement is
  money.
• A token economy is a set of contingencies based
  on token reinforcement the contingencies specify
  when and under what conditions, particular forms
  of behavior are reinforced with tokens. Tokens
  are exchangeable for a variety of backup
  reinforcers.