4A: Punishment, Generalisation, Discrimination & Stimulus Control Flashcards
Positive punishment
Giving a punisher to reduce a behaviour
Negative punishment
Removing a pleasant stimuli to reduce a behaviour
Time out
The loss of access to positive reinforcers for a brief period of time following a problem behaviour
Response cost
The removal of a specific reinforcer following a problem behaviour
Intrinsic punishment
Punishment is an inherent aspect of the behaviour
Extrinsic punishment
Not an inherent part of the behaviour, but just follows the behaviour
Problems with punishment
- Escape or avoidance of the punishment
- Aggression towards the punisher
- Apathy or suppression of all behaviour
- Abuse
- Imitation of the punisher
The persona delivering the punishment could become a SD for punishment such that the unwanted behaviour is only suppressed when that person is present.
Benefits of punishment
- Punishment can sometimes lead to an increase in social behaviour
- Punishment sometimes results in an improvement in mood
- Punishment can increase attention to the environment
To maximise effectiveness, punishment should be…
- Immediate
- Consistent
- Intense enough to supress the target behaviour – but
balanced so that it is not abusive - Negative punishment is preferable to positive punishment
- Combined with an explanation (where possible)
- Combined with positive reinforcement for appropriate behaviour
Contingency
The degree to which a punishing event is dependent on the behaviour.
The stronger the correlation, the more effective the punishment and the faster a behaviour will change.
Contiguity
The gap between a behaviour and its consequence
In general, the longer the delay, the less effective the punisher will be.
Punisher intensity
The stronger the punisher is the more effective it is in reducing an unwanted behaviour.
Introductory level of punisher: Is it better to start with the strongest intensity of a punisher or begin with a small or weaker punisher and build up the intensity if the behaviour continues?
Both can be problematic:
• If we build up the intensity, it can increase the threshold that the subject can withstand.
• If we start with a strong punisher we have no way of knowing the exact appropriate level.
Motivating operations
Anything that changes the effectiveness of a consequence, either in terms of increasing or decreasing its effectiveness. e.g. food deprivation
Two-process theory
Punishment involves both operant and classical conditioning - paid lever with shock, conditioned to fear lever (shock) then reinforced when avoid lever.
One-process theory
Punishment only involves operant conditioning - punishment weakens behaviour
Conditioned Suppression Theory
Punishment does not weaken a behaviour but instead produces an emotional response that interferes with the occurrence of the behaviour
The premack approach to punishment
A low-probability behaviour can be used to punish a high probability behaviour
Generalisation
The tendency for the effects of a learning experience to be transferred to other situations or environments.
4 types of generalisation
Generalisation across people (also called vicarious generalisation)
Generalisation across time (response maintenance)
Generalisation across behaviours (response generalisation)
Generalisation across situations (stimulus generalisation)
Generalisation Gradient
Graphic representation of generalised data.
Shows the tendency for a behaviour to occur in situations that differ systematically from the training situation.
Response generalisation
The tendency for changes in
one behavior to spread to other behaviors.
Stimulus generalisation
The tendency for changes in behavior in one situation to spread to other situations.
Increase generalisation
Providing the training across a variety of settings
Vary the consequences
Reinforce generalisation
Stimulus discrimination
The tendency for behaviour to occur in certain situations but not others.
Discrimination training
Any procedure for establishing a discrimination. Discriminations can be established through both Pavlovian and operant procedures.
e.g. pair one stimulus (CS+) with a US, another stimulus (CS-) appears alone; one stimulus (S+) is reinforced, while another (S-) is never reinforced.
Simultaneous discrimination training
An operant discrimination training procedure in which the S+ and S- are presented at the same time.
Successive discrimination training
The S+ and S- alternate, usually randomly. When the S+ appears, the behavior is reinforced; when the S- appears, the behavior is not reinforced.
Matching to sample
The task is to select from two or more alternatives (called comparison stimuli) the stimulus that matches a standard (the sample). The comparison stimuli include the S+ —the stimulus that matches the sample—and one or more S-.
Oddity matching / mismatching
The subject is shown a sample and must select the opposite disk in order to receive reinforcement
Stimulus control
Describes the relationship through which a discriminant stimulus reliably affects the probability of the behaviour.
The tendency for a behavior to occur in the presence of an S+ but not in the presence of an S-.
Complex schedules
Often examine stimulus control through complex schedules
Multiple schedules: consist of two or more independent schedules presented in sequence – order can change but reinforcement occurs in each (differs from chained schedules)
Stimulus control evidenced through different response patterns for each schedule
Behavioural contrast
Change in rate of reinforcement on one part of multiple schedule results in opposite change in rate of response in another part
Positive contrast effect
Decrease in rate of reinforcement (or use of punishment) on one leads to increase in rate of response on other component
Negative contrast effect
Increase in rate of reinforcement on one leads to decrease in rate of response on other component
Anticipatory contrast
Change in rate of responding in anticipation that there will be a change in the rate of reinforcement
Application of stimulus control
Animal trainers
Human applications include use of cues for behaviour, e.g. medication compliance
Pavlov’s theory
Discrimination training produces physiological changes in the brain.
The CS+ is associated with an area of excitation in the brain and the CS- is associated with an area on inhibition.
• If a novel stimulus is similar to the CS+ it will excite an area of the brain near the CS+ area and elicit the CR.
• If a novel stimulus is similar to the CS- it will excite an area of the brain near the CS- area and inhibit the CR.
Spence’s theory
The tendency to respond to any given stimulus is a result of the interaction between the increased and decreased tendencies to respond as reflected by a gradient curve
• Excitatory Strength is increased when a response to a specific stimulus is reinforced
• Inhibitory Strength is increased when a response to a specific stimulus is not reinforced
• The tendency to respond to any stimulus is based upon its Net Excitatory Strength
Lashley-Wade theory
Prior experience with stimuli similar to those used in training impacts the generalisation gradients.
Discrimination training increases the steepness of the gradient because it teaches the subject to tell the difference between the target stimulus and other stimuli.
Generalisation occurs because a subject has not had much experience with the stimuli involved and as such is not able to discriminate between them.
Learned helplessness
Is a reduction in learning ability that results from repeated exposure to uncontrollable aversive events.
