operant conditioning

exam 3 psychology of learning

classical conditioning

behavior changes due to the association of two stimuli (CS-UCS) presented prior to the response (CR); focuses on elicited behavior (involuntary); the environment operates on the animal, stimulus evokes response (S>R), animal learns that CS predicts US

operant conditioning

behavior changes as a result of consequences that follow it; focuses on emitted behaviors (voluntary); the animal operates on the environment, stimulus evokes a response to produce an outcome (S>R>O), animal connects context, behavior, and outcome; more powerful and flexible than classical conditioning

Edward Thorndike

“Discovered” instrumental conditioning, which is now called operant conditioning. He studied how cats learned to escape from puzzle boxes; operant conditioning is the process whereby organisms learn to make responses to obtain or avoid a certain outcome; he pioneered experimental approaches to learning

Skinner’s Theory on Operant conditioning

behavior is influenced by consequences that follow it

Thorndike’s Puzzle Box

Thorndike placed cats in a small chamber they disliked with a pulley that allowed them to escape; with repeated exposures to the box, cats initially tried many behaviors and would often escape accidentally at first. Behaviors that opened the box became more frequent over time and behaviors that weren’t useful disappeared

Thorndike’s theory on instrumental learning

an organism’s behavior is instrumental in bringing about certain outcomes; initially a lot of behaviors are tried out, and knowledge begins to guide future behaviors. Behaviors with positive outcomes increase while ones with negative outcomes decrease

Thorndike’s law of effect

responses followed by a satisfying consequence become more likely to occur;, responses followed by an unsatisfying consequence becomes less likely to occur; So>R>O is discriminative stimulus, response, outcome; ABCs of behavior (Antecent, Behavior, Consequence)

R response: What is this R that is learned?

R is a “behavioral unit”- not a single behavior but a class of behaviors producing an effect, some cognitive psychologists would call it a goal or intention

S: Discriminative stimulus (Context)

operant conditioning is about contingencies (if R, then O) however contingencies can change (if i cry, mom picks me up…only if mom is sober!); S tells us which contingencies are in effect, if R, then S -O, however is R and no S, then nothing.; operant conditioning is a three part association, context/discriminative stimulus —> behavioral response —> outcome; S is crucial in this type of learning ensuring that we select behaviors that are appropriate to the situation (you would only put money into a pop machine that is working)

The skinner box

context(S) is a light signaling that the box is “on”, the behavior (R) is the rate at which the mouse lever presses, the outcomes (O) are food delivery as a reward or shock through wires in the floor as punishment; the animal is free in the chamber, no experimenter intervention (free-operant learning)

Shaping

successive approximation builds a complex R incrementally, initially, the contingency is introduced for simple behavior, R, as the rate of R improves, the contingency is moved to a more complex version of R. Gradually, it builds a complex R animals that would never spontaneously produce

Chaining

Chaining builds complex R sequences by linking together S>R>O conditions; initially, train the animal to pick up an object, next reward it for picking it up and then throwing it. allows for a series of behaviors as opposed to shaping, which simply elaborates on a single response

Reinforcement

response strengthened by consequence that follows

punishment

response weakened by consequence that follows

Primary reinforcer

stimuli that organism finds reinforcing because they satisfy biological needs (e.g. food and drink)

Secondary reinforcer

acquire reinforcing properties through association with primary reinforcers (e.g. money and praise)

Positive Reinforcement

occurs when a response is strengthened by the presentation of a stimulus (the positive reinforcer) that follows it

Negative reinforcement

response strengthened by removal (or avoidance) of an aversive stimulus (the negative reinforcer)

escape learning

consequence of behavior is terminating an already present aversive stimulus

avoidance learning

consequence of behavior is preventing the appearance of an aversive stimulus

Factors affecting positive reinforcement

amount (magnitude of reinforcer), quality of reinforcer, delay (timing between behavior and consequence), schedule (relationship between behavior and consequence)

factors affecting negative reinforcement

amount (magnitude of reinforcer), delay (timing between behavior and consequence), and schedule (relationship between behavior and consequence)

continuous reinforcement schedule

every response of a particular type is reinforced; there is more rapid learning, consequences are easier to perceive and extinction is more rapid in continuous reinforcement

partial reinforcement schedule

only some response are reinforced;

Ratio partial reinforcement schedules

certain number of responses are reinforced

interval partial reinforcement schedule

some amount of time must elapse between reinforcements

fixed partial reinforcement schedule

reinforcement occurs after a fixed number of responses or fixed time interval

variable partial reinforcement schedule

reinforcement occurs after an average number of responses or passage of time

fixed ratio schedule

reinforcement given after a fixed number of responses (e.g. pulling a lever every x amount of times will produce an outcome); steady responding until reinforcement, then post-reinforcement pause, a higher ratio means longer pausing after each reward; O - activity slows after reinforcement and then picks up

variable ratio schedule

reinforcement given after a variable number of responses, centered around an average (e.g. slot machines, or notifications of media); constant and high rate of responding due to possibility (gambling, video games, etc); greatest activity of all schedules

fixed interval schedule

first correct response after a fixed time interval is reinforced; behavior before the interval expires has no consequence; at beginning of interval, little to no responding increasing to rapid rate of responding before interval expiration (e.g. checking the time before an appointment); activity increases as deadline nears

variable interval schedule

reinforcement given for the first correct response after a variable time interval, centered around an average; behavior is steady but low rate of responding, such as checking for emails; steady activity results

matching law of choice behavior

response rates to concurrent schedules often correspond to the rate of reinforcement for each schedule

operant extinction

weakening and eventual disappearance of an operant response because the response is no longer reinforced; resistance to extinction is determined by the pattern of reinforcement that has maintained behavior; good alternative to punishment

aversive punishment

response weakened by subsequent stimulus presentation (e.g. spanking or scolding)

response cost punishment

response weakened by subsequent removal of a stimulus such as loss of privileges or money

punishment

Using punishment effectively is very tricky; circumvention → an animal may learn discriminative stimuli that help it avoid punishment, can produce fear/anxiety that can impair behavior, can produce generalized behavior disruptions, can produce aggression, can result in learned helplessnes

punisher

an outcome that decreases the frequency of the behavior

Generalized suppression

a temporary decrease in most behaviors; skinner proposed that punishment does not produce true operant conditioning (learning of a contingency)

How to make punishment effective

manner of introduction → weak punishers habituate, escalating punishers habituate more, it is necessary to start with strongest punishment. immediacy, schedule, and alternatives are the three keys