Unformatted text preview: 9/30/08 Thomas G. Brown, Ph.D. Rules Sensitive, describing when reinforcers happen reliable baseline – Motivation – Punishment – Drug effects – Etc
Search for lawfulness External validity Constant Schedules of Reinforcement
– Every response treated the same – All reinforced or all not reinforced
CRF  continuous reinforcement EXT  extinction Intermittent Schedules of Reinforcement
– Sometimes a response is reinforced – Not every time, but sometimes
Four simple ones 1 9/30/08 Determining Factor Count Time Unchanging Fixed Ratio Fixed Interval Variability Variable Ratio Variable Interval Changing Patterns Rates of responding Pauses What do we need to control for?
and magnitude – Reinforcement density
Frequency Two methods – Yoked control – Within subject 2 9/30/08 Yoked control  uses two chambers VR in one and VI in other VI is determined by VR
– Each time the animal gets food on VR, the next response in the other chamber is reinforced Guarantees equal rft distributions while keeping different contingencies in place Faster on ratio schedules
– Ferster and Skinner (1957) – VR: 23/sec and VI: 1/sec Within subject design Alternate a VR (green) and VI (red) Each VI period determined by previous VR (how long it took) VR faster then VI (again) Herrnstein (1964)
Why? In ratio schedules:
– Differential reinforcement for higher rates – Work faster and get food sooner In interval schedules:
– No such differential reinforcement – Working faster doesn’t result in sooner food Molar level analysis
– Explains “pressure” to respond faster on ratios But there is also “pressure” to respond more slowly on interval schedules 3 9/30/08 Interresponse time  IRT
– Time between responses Longer IRT means slower responding Assume a FI 60”
– What is probability of rft if IRT is 15 seconds? – If 30 seconds? – If 60 seconds?
So “pressure” is to lengthen IRT
– to slow down and increase rft probability Assume a FR 60 – What is probability of rft if IRT is 15 seconds? – If 30 seconds? – If 60 seconds?
So there is no “pressure” is to lengthen IRT – slowing down either increases or decreases rft probability  for a given response Molar Level analysis
– In ratio schedules:
Work faster and get food sooner Working faster doesn’t result in sooner food – In interval schedules:
– So there is “pressure” to respond faster on ratios
Molecular Level analysis
– In interval schedules:
Slower responding (longer IRT) increases rft probability Slower responding does not increase probability of reinforcement – In ratio schedules:
– So there is“pressure” to slow down on intervals 4 ...
View
Full Document
 Spring '08
 BROWN
 interval schedules, ratio schedules, Thomas G. Brown

Click to edit the document details