Unit 4 · Learning Theory · Reading 8 of 13
What Counts as a Reward, and When
Before you read
A baby isn't impressed by a $20 bill. You probably are. Why?
Words you'll own by the end
- Primary reinforcer: naturally rewarding, unlearned: food, water.
- Secondary (conditioned) reinforcer: rewarding only because we've learned to link it with other reinforcers: money, grades, trophies.
- Continuous reinforcement: reinforcing every time the behavior happens.
- Partial (intermittent) reinforcement: reinforcing only some of the time.
1. Primary vs. secondary reinforcers
Primary reinforcers are unlearned and rewarding all on their own. The list is short. Food and water are definitely on it, and some would add shelter, love, and sex.
Secondary reinforcers become rewarding through experience. An animal can be trained to value a poker chip if the chip can be traded for food. The chip has no value by itself. Money is the same: it means nothing to a baby. Promotions, trophies, and a compliment from your coach are all secondary reinforcers.
Try it: Primary or secondary?
Is it rewarding by nature, or only because you learned it leads to something?
A cold drink of water after practice
An A on your report card
Points in a video game you can trade for new skins
A warm meal when you're hungry
A "great job" from your coach
2. Timing matters
Behaviorists like Skinner found that the closer the consequence follows the behavior, the faster learning happens. If your dog sits and you give the treat immediately, it learns quickly. A treat ten minutes later teaches much less.
3. Every time, or only sometimes?
With continuous reinforcement, the behavior is reinforced every single time. It's great for the acquisition stage. A toddler who gets a gold star every time she uses the toilet learns fast.
The catch: once the rewards stop, the behavior tends to undergo rapid extinction. The learner notices right away that the deal is over.
That's why behaviorists recommend switching to partial (intermittent) reinforcement once a behavior is learned. When rewards only come some of the time, the learner can't tell when the next one is coming, so the behavior is much more resistant to extinction. There are four partial schedules. They're next.
Check-in: A vending machine always works, then breaks. A slot machine pays only now and then. Which do people keep trying longer after it stops paying, and why?
Am I ready for a pop quiz?
Check a box only if you can say the answer out loud without scrolling up.
Come back to your "before you read" question
The $20 bill is a secondary reinforcer. You learned its value from what it gets you, and a baby hasn't learned that yet.