05 10 2012 I Can any behavior by operant A Some people think so B Relating to the changing of the distribution of behaviors based on reinforcement II Belongingness in Instrumental Conditioning A Response reinforcer belongingness effects are also found in instrumental conditioning B Thorndike s puzzle boxes 1 Scratching self yawning as escape responses are harder to train via reinforcement C Breland and Breland 1961 attempted to condition response chains in various animals for entertainment in zoos 1 Some responses would change from the reinforced type to an unwanted form instinctive drift behaviors would drift into a form that was more natural for the response reinforcer relationship D Shettleworth found that food reinforcement could increase environment directed activities in hamsters but not self care responses 1 The giving of free reinforcement can actually activate the foodsearch behaviors 2 Reinforced self care responses do not increase III Quantity and Quality of the Reinforcer A Hutt 1954 varied both the amount of liquid and the quality acid normal sweet of the liquid used to reinforce lever pressing behavior 1 Larger amounts of liquid within a certain range produce more vigorous responding 2 For all amounts sweet water normal water acidic water a Acidic water still produces a positive response lever pressing B What about a change in quantity of a reinforcer 1 Mellgren 1972 a Rats are trained to run down a runway i Small reward ii Large reward b In phase 2 half the rats in each condition are switched to the other condition c Measure speed i Phase 1 high reward groups produce a slightly higher response faster than the low reward group ii Phase 2 low high group increased speed to much faster than the hh group s phase 1 or 2 levels o Positive contrast effect High low group decreased speed far below that of the low group s phase 1 level o Frustration disappointment Negative contrast effect 2 Tinklepaugh and his monkeys a We previously learned that the reinforcer itself wasn t part of the association but clearly these animals are learning not just what to do but what to expect IV Contiguity vs Contingency A Contiguity how close in time the reinforcer follows the response temporal relationship 1 How can we overcome longer delays in reinforcement a Secondary conditioning conditioned reinforcement immediately follow the correct response with a stimulus that has previously been paired with the reinforcer b Marking procedure marking each instance of a response correct or not i Example press a lever light turns on for 5 seconds 25 seconds later food The light is the mark ii Marks make your response more salient and improve performance B Contingency to what degree delivery of the reinforcer depends on the occurrence of the instrumental response 1 Is the contingency between the response and reinforcer necessary to strengthen an instrumental response a Skinner discovered superstitious behavior in pigeons when food was delivered every 15 seconds irrespective of behavior i Adventitious accidental reinforcement of whatever behavior the pigeon happened to be engaged when food was delivered ii Relation to human superstitious behavior b Staddon Simmelhag 1971 w behavior systems theory the food is engaging the behavior system as they come to expect the food they ll move from general search behaviors towards focal search behaviors i Terminal responses ii Intrim responses V Controllability of reinforcers A If a subject has experienced a lack of control over a reinforcer he she will show a decrement in learning to make a response B To degrade contingency present the reinforcer in the absence of the target behavior 1 All bar presses are followed by food but not all food was predicted by a bar press C Learned helplessness 1 Impairs the ability to learn avoidance techniques a Making can rescue and animal from learned helplessness D What about attentional deficit 1 They learn to pay less attention to their own actions than do animals in a group where their actions influence the environment E Why doesn t escapable shock impair learning 1 Sensory feedback stimuli accompanying the escape response sensations of your own body that you create can signal the end of the shock session a These associations reduce fear 2 If you signal the end of the shock in an escapable group they will be more likely to learn an escape behavior in the next phase