PSYC2015 Lecture 7

• Lecture discusses the effects of choice on behavior.
• Two main topics include the effects of choice and stimulus-guided choice.

Effects of Choice

• Concurrent choice involves the simultaneous availability of different choices.
• Matching Law is a principle describing how behavior is distributed among different options based on their reinforcement rates.
• Differential reinforcement of other responses (DRO) involves reinforcing the absence of a specific behavior for a period of time.
• Inter-temporal choice involves making decisions between options available at different points in time.
• Temporal discounting refers to the decrease in the subjective value of a reward as the delay to receiving it increases.
• Improving self-control includes methods to increase the ability to delay gratification for larger, later rewards.

Instrumental Behavior

• Desirable outcomes are considered good, while undesirable outcomes are considered bad.
• Current behavior is selected by its consequences; behaviors leading to desirable outcomes are repeated, and behaviors leading to undesirable outcomes are inhibited.
• People learn to discriminate situations to repeat or inhibit behaviors based on previous outcomes.
• People have the chance to take more than one action, each producing a different consequence, and learn to choose actions that maximize desirable outcomes.

Belief-Desire Models

• Motivation can be broken down logically: a person knows an action leads to an outcome, desires the outcome, and performs the action.

Goal-Directed Behavior

• Monkeys can learn to choose a pyramid (Sd1) for a cherry reward (Rf1) and a cube (Sd2) for a peanut reward (Rf2).
• Reward devaluation can change behavior: if a monkey overeats cherries, it will then prefer the cube over the pyramid, indicating instrumental behavior is guided by the expected outcome.

Response Rate and Reinforcement

• Behavior rate (B) is proportional to reinforcement rate (R), with a slope constant (k), shown as B = kR.

Concurrent Schedules

• Concurrent schedules involve two behaviors (B1, B2) performed at different rates.
• The relative rate of responding of B1 is calculated as B1 / (B1 + B2).
• These behaviors are reinforced at different rates (R1, R2).
• The relative rate of reinforcement of B1 is R1 / (R1 + R2).
• Matching Law predicts that the distribution of behavior matches the distribution of reinforcement: B1/(B1+B2) = R1/(R1+R2). The relative rate of behavior matches the relative rate of reinforcement.

Reinforcement and Punishment

• Four possibilities relating contingency to outcome: Positive Reinforcement, Positive Punishment, Negative Punishment, and Negative Reinforcement.
  • Contingency is relationship between response (R) and outcome (O).
  • In positive reinforcement, R causes O, and can give something good with response inceasing.
  • In positive punishment, R causes O, and can give something bad and response decreases.
  • In negative punishment, R prevents O, and can remove something good, and response decreases.
  • In negative reinforcement, R prevents O, remove something bad and response increases.

Punishment

• The availability of an alternate, non-punished response affects the effectiveness of punishment. If there is no alternative, punishment can suppress behavior, but if an alternative is available, the individual may switch to the non-punished response.

Negative Punishment/Omission

• In negative punishment/omission, if R1 prevents O1, then R1 decreases.

Differential Reinforcement of Other Behaviors (DRO)

• If R1 prevents O1 and R2 causes O1, then R1 decreases. DRO is a method of reducing unwanted behavior, and involves reinforcing an alternate but competing behavior.
• People can only do one thing at a time, and has the best results if the alternate behavior the same function as unwanted behavior.
• DRO avoids the negative effects of using aversive consequences to reduce behaviors.

Inter-Temporal Choice

• Involves choices between a current outcome and a future outcome rather than between two current outcomes. Examples: dieting snack vs future health, quitting satisfying craving vs future health.
• Self-control involves delaying immediate gratification for a bigger, longer-term reward.
• Impulsivity is demonstrated forgoing a larger long-term reward for immediate gratification

Risk of Losing Reward

• Risk is compounded for each period of time.

Temporal Discounting

• A more immediate but smaller reward can have a larger subjective reward value at the time of choice than a delayed but larger reward.
• For each time (Delay, D) there is a risk (r) that reward (A) may be lost. B = Ae^ (-rD).

How to Increase Self-Control:

• Control antecedents by removing options to do impulse behavior, removing triggers for compulsive behavior, and making longer-term rewards salient.
• Control competing choices by punishing less preferred option, and delay the impulsive choices

Stimulus Guided Choice

• Monkeys can learn to chose cube (Sd2). Reward devaluation gets monkeys to eat eg: cherries then given a choice between pyramid and cube, which causes a change to prefer cube
• Pavlovian to Instrumental Transfer guides choice behaviors

Two-Factor Behavior

• Consists of Pavlovian cues and instrumental behavior
• Pavlovian cues affects the choices, and signal availability of outcome
• Instrumental or operant conditioning is one factor
• Classical or Pavlovian conditioning is the other factor
• Affects motivation, and induce craving

Cue-Induced Craving

• Cue-induced craving increases drug use
• It is incubation of cue-induced craving: cue induced craving becomes stronger at longer abstinence periods, then drops off.
• Can be explained with "two factor processes in drug seeking behaviors" with avoidance of fear and craving.

Readings

• A good general reading on behaviour modification procedures – excellent for the self-control section: Martin, G. & Pear, J. (2007).
• Choice and self-control pp 260-272. (272-276 is also interesting but not essential) is interesting, and can be found in Bouton.
• Premack Principle pp277-282 is interesting and can be found in Bouton.
• Domjan, M., & Delamater, A. R. (2023). The essentials of conditioning and learning (5th ed.). American Psychological Association. It can be accessed via the library: https://doi.org/10.1037/0000363-000. Chapters 7 and 8 are important to read
• More information about Marshmallow test is available at https://journals.sagepub.com/doi/full/10.1177/0956797618761661