(PRE) accelerated resistance to extinction after occasional reinforcement instead of after continuous reinforcement. Commonly referred to as partial reinforcement extinction effect (PREE).
In psychology, what is the "Partial Reinforcement" effect?
a) Definition of partial reinforcement
B.F. Skinner was known for his Operant Conditioning concepts. Part of which is Partial Reinforcement, which is the intermittent reinforcement that occurs after a response.
b) Importance of partial reinforcement
Those who have kids or nephews or nieces, especially those aged 4-7, might have noticed that rewarding good behavior with something like chocolate if they stay still tends to make them behave and be more obedient. However, after a certain time, it becomes harder to get them to do the desired behavior on the same reward model. The impact of chocolates decreases over time. What the rewarder was doing was an example of continuous reinforcement that we knew would eventually lose its effect in the long run.
The importance of partial reinforcement is brought into play by neutralizing the negative possibility of continuous reinforcement. Later on, we will learn about the real-life samples of this phenomenon, but first, we will get to know its different types.
Types of partial reinforcement
There are four types of partial reinforcement and to help you better understand, we will break down the terminologies before going to each type. The first concept is “Ratio” which is defined as the amount of a certain thing or for our discussion, the number of responses triggered by the reinforcement.
Next is the concept of “Interval”. When you encounter the word Interval in our discussion, it refers to an amount of time.
Each of these types/schedules can either be Fixed or Variable, meaning the reinforcement can be consistent (fixed) or there may be variations (variable). So if you combine all the concepts discussed, we now have the four types: a.) Fixed Ratio b.) Fixed Interval c.) Variable Ratio and d.) Variable Interval.
A. Fixed Ratio - In this type/schedule, reinforcement is given after a fixed number of responses have been made. A classic example is the hungry rat in the Skinner Box which gets a food pellet every time it pressed the lever five times. Therefore, reinforcement is activated in every fifth response.
B. Fixed Interval- this type/schedule gives weight to the aspect of time regardless of the amount of response. If the fixed interval is set, for example to every 1 hour, the subject is given a reward every hour no matter the number of responses. Take for example the case of Skinner’s rat, the number of times it pressed the lever will not earn the rat a food pellet unless the set time has lapsed. It will only get a food pellet every hour regardless if the rat pressed the lever only once or even a hundred times.
C. Variable Ratio- this type/schedule focuses on giving reinforcement for every variable number of responses that have been made. In the sample shown in Fixed Ratio, the rat has to press the lever five times to gain a food pellet. However, in the variable ratio, the rewarder can set the reinforcement in varied instances or several presses in this case as long as the average or mean of the variables will equal five. An example is shown in the table below:
Number of Lever Press | Result |
4 | First Reinforcement |
7 | Second Reinforcement |
3 | Third Reinforcement |
6 | Fourth Reinforcement |
Average: 5 |
As shown in the table above, there was a variety of responses the rewarder set for the rat to earn a food pellet. The first reinforcement required four presses, seven for the second reinforcement, and so on. During the four instances, the average is five presses, the same as the Fixed Ratio.
D. Variable Interval- in this type/schedule, different amounts of time are set to elapse before giving reinforcements. Reinforcement is dependent on time but the interval is random. A sample is a kid waiting under an apple tree with the fruits falling at different times of the day. The subject (kid) is unable to determine the time when an apple falls (reinforcement); hence, the rate of response is consistent.
How does partial reinforcement affect behavior?
The type/schedule of reinforcement affects the behavior as shown in the table below:
Type of Partial Reinforcement | Behavior |
Fixed Ratio | Response rate is high when aiming for the reinforcement requirement. There is a decline in the rate of response in between reinforcements known as Post-Reinforcement Pause. Example, if the rat already earned a food pellet after five presses in the lever, it has the option to rest and would decide to have another round of five presses to gain another pellet. The reinforcement is certain. |
Fixed Interval | Moderate response rate with significant pauses after reinforcement. Example, if the rat is conditioned to get a pellet every five minutes, it will press the lever more if the time is about to reach the fifth minute of each interval. The response rate is highest during the moment of reinforcement and with significant pauses (decline in response rate) after reinforcement. |
Variable Interval | Moderate yet steady response rate. The unpredictability on when the reinforcement is given makes this type of reinforcement a better type compared with the fixed types. |
Variable Ratio | Reinforcement is delivered after an unpredictable number of responses. This type of partial reinforcement is considered the best as it yields high response rate due to unpredictable response requirement. |
Examples
Here are some of the real-life examples of how each partial reinforcement is applied:
Fixed Ratio:
1. A car salesman earns a bonus for every five cars he sells regardless of the time aspect. If he sells five a week, he gets a bonus. If he sells another five the next week, he’ll get a bonus. If he sells five cars the next month, he earns a bonus.
2. A factory worker gets paid if he fills ten crates of finished products regardless of how early he is done.
3. A fruit picker gets a day’s wage if he harvests a hundred apples.
Fixed Interval:
1. A car salesman gets a salary at the end of the month regardless of how many cars he sells. If he sells one car he gets the same amount of salary at the end of the month and still earns the same wage even if he sells ten cars.
2. A factory worker gets paid if he fills or not ten crates of finished products at the end of the working day.
3. A fruit picker gets a day’s wage at the end of the day even if he harvests ten or a hundred apples.
Variable Interval:
1. A car salesman is given a bonus by his supervisor if he is seen engaging with customers. Since the car salesman does not know when the supervisor will visit and see him engage with customers, he will always have to engage with clients hoping his supervisor comes and sees him working hard.
2. At an unannounced schedule, undercover members of the Quality Control Division visit outlets of a fast food chain and reward the crew on that shift $40 bonus for their clean resto and quality service. The store manager is always on his toes to remind the crew to perform at its best.
3. Unannounced quizzes always have students burn the midnight oil as tomorrow’s class may have a surprise test to open the session.
Variable Ratio:
1. Slot machines are the best examples of variable ratio reinforcement. Since the gambler does not know when the next bet produces a jackpot prize; hence, the difficulty of leaving the gambling area.
2. Social media scrolling. You never know what posts or notifications you get every time you open your account which is why you cannot resist the urge to check. Unannounced quizzes always have students burn the midnight oil as tomorrow’s class may have a surprise test to open the session.
3. Scratch-offs in your nearby convenience store gives you the thrill of being the next millionaire which is why you cannot stop buying and scratching off the next coupon.
References:
https://www.researchgate.net/publication/358797170_Operant_Conditioning_Theory-_BF_Skinner