WK12 Reading Instrumental Learning and Skill

Cards (80)

  • Instrumental learning
    Learning by trial and error, interacting with the environment and observing the consequences
  • People and animals are able to learn how to do things by interacting with their environments and observing the consequences of those interactions
  • Sutton & Barto (1998): 'The idea that we learn by interacting with our environment is probably the first to occur to us when we think about the nature of learning. When an infant plays … it has no … teacher, but it does have a direct sensorimotor connection to its environment. Exercising this connection produces a wealth of information about cause and effect, about the consequences of actions, and about what to do in order to achieve goals.'
  • 4 phases to Learning by trial and error
    1. Trying different things
    2. Observing whether the desirable consequence occurs
    3. Discarding unsuccessful attempts
    4. Retaining successful attempts
  • Intelligent method of trial and error
    Try to determine through thinking and reasoning which variants of behaviours in your existing repertoire are able to solve the problem
  • Unintelligent method of trial and error
    Try out random variations of the behaviours in your repertoire that are in some way related to what you want to achieve
  • The learning process uses information about success or failure - feedback information - to develop the tendency to perform the successful behaviour(s) and not to perform the unsuccessful ones
  • Edward Thorndike
    Pioneered the study of this kind of learning in the 19th century
  • Thorndike's puzzle boxes
    • Box A: Pull down on a loop to release the catch and open the door
    • Box K: Depress a treadle, pull on a string and push up or down on a bar to release the catches and open the door
  • When first put in a puzzle box, cats would engage in what appeared to be species-typical, escape-related behaviours that would not lead to escape unless, by accident, they resulted in the effective action
  • The animal does not have to learn how to perform a new kind of action: it already has the ability to pull on loops, push down on bars and depress treadles
  • Law of Effect

    Of several responses made to the same situation, those which are accompanied or closely followed by satisfaction to the animal will, other things being equal, be more firmly connected with the situation, so that, when it recurs, they will be more likely to recur. Those which are accompanied or closely followed by discomfort to the animal will, other things being equal, have their connections with that situation weakened, so that, when it recurs, they will be less likely to occur.
  • According to the Law of Effect, the consequences of actions only play a role in learning as a result of the internal states of 'satisfaction' or 'discomfort' they produce
  • The Law of Effect proposes that learning is a process of strengthening and weakening stimulus-response (S-R) connections
  • Reinforcer
    Something that is capable either of increasing the likelihood that the behaviour that produced it will be performed again or of otherwise increasing the strength of that behaviour
  • Reinforcement
    The delivery of a reinforcer to (or the acquisition of a reinforcer by) an animal or person
  • Positive reinforcement
    Getting something desirable
  • Negative reinforcement
    Termination of an aversive situation
  • Punishment
    The delivery of something that decreases the likelihood that a behaviour will be performed again (or otherwise weakens the behaviour)
  • Positive punishment
    The delivery of a punishing stimulus
  • Negative punishment
    The removal or loss of something desirable or the prevention of access to something desirable
  • Negative punishment
    The removal or loss of a something desirable or the prevention of access to something desirable
  • Reinforcer
    Something that is capable either of increasing the likelihood that the behaviour that produced it will be performed again or of otherwise increasing the strength of that behaviour (also called REWARD)
  • Prison would be an example of both negative and positive punishment
  • Prison is negative punishment because it punishes by taking something away (your liberty and freedom to do the things you want to do)
  • Prison is also positively punishing because being confined and being told what to do is intrinsically aversive
  • Instrumental learning
    The kind of learning that concerns behaviours that are instrumental in bringing about consequences
  • Instrumental conditioning procedure
    Experimental situations in which executing a particular behaviour results in the delivery of reinforcement or punishment
  • Operant conditioning
    The term used by B.F. Skinner to refer to instrumental conditioning procedures
  • Instrumental conditioning procedures involve making the delivery of a reinforcer or punisher conditional upon the trainee performing a particular action
  • Instrumental conditioning procedures in rats (examples of reinforcer)
    • Positive reinforcement (rat presses lever, food pellet delivered)
    • Negative reinforcement (escape or avoidance procedures)
  • Instrumental conditioning procedures allow investigators to determine whether animals or people can discover what to do so as to obtain reinforcers or avoid punishment
  • According to the Law of Effect, only connections between stimuli and responses (S-R connections) are changed during instrumental learning, not connections between responses and outcomes (R-O connections)
  • The Law of Effect implies that animals do not learn that actions (responses) have consequences, and cannot learn to do things in order to achieve outcomes ( their behaviour is not goal-directed/ oriented)
  • Reinforcer devaluation studies have shown that animals can learn about the relationship between responses and outcomes (form R-O connections)
  • Habitual behaviour (habit)
    An instrumental behaviour whose production is insensitive to the value of its outcome (associated reinforcer)
  • Goal-guided behaviour
    An instrumental behaviour whose production is sensitive to the value of its outcome (associated reinforcer)
  • The Law of Effect seems to imply that animals do not learn how to do new things or learn new ways of doing old things, as all that changes is the likelihood that the animal will do a particular thing
  • The Law of Effect seems to imply that animals do not learn how to do new things or learn new ways of doing old things
  • All that happens during the learning process is that behaviours already in the animal's repertoire are more or less strongly connected to stimuli (S-R connections are strengthened or weakened)