Kamin Blocking
The Rescorla-Wagner model links learning to surprise.
Reinforcement and Animal Learning
Kamin blocking is important because it shows that learning cannot always be understood by examining one conditioning event in isolation. What an animal has already learned can change what it learns later. This idea connects the study of reinforcement with broader attempts to explain animal learning and classical conditioning.
The term reinforcement has changed meaning across historical traditions. In the context of animal learning, the source traces an early use to the 1927 English translation of Pavlov's monograph, where reinforcement referred to an event that was not dependent on an action. Later, Mackintosh used reinforcement for either strengthening or weakening a pattern of behavior, while Skinner used it only for strengthening behavior and treated weakening as an effect of punishment.
The Two-Stage Learning Pattern
Kamin blocking describes a two-stage conditioning situation. First, one stimulus is repeatedly involved in conditioning. Later, that previously conditioned stimulus appears together with a new stimulus. The key question is whether the new stimulus will also become a predictor. In blocking, prior conditioning prevents or greatly reduces the expected new learning about the added stimulus.
Tracing a blocking procedure
Suppose an animal first experiences Cue A repeatedly in a conditioning arrangement. Later, Cue A is presented together with a new Cue B. Explain why Cue B may fail to become a strong new predictor.
Stage 1: establish a predictor: Cue A is repeatedly involved in conditioning. The animal develops prior conditioning involving Cue A.
Stage 2: add a new cue: Cue B is introduced alongside Cue A. The compound presentation contains both the previously conditioned cue and the new cue.
Stage 3: evaluate the later learning: Because the prior conditioning has already prepared the animal for what will happen, the added Cue B contributes little or no additional learning in the blocking case.
Locate the change: The important change occurs between the first and second stages: the animal's prior conditioning alters the conditions under which the later compound is learned.
Kamin blocking is the pattern in which prior conditioning involving Cue A blocks later learning about the added Cue B.
Surprise as the Learning Signal
The Rescorla-Wagner model links learning to surprise. Learning is expected when what occurs differs from what the animal was prepared to expect. If earlier conditioning already accounts for the important outcome, adding another stimulus may produce little or no additional learning because the outcome is not sufficiently surprising.
This is why the model directs attention to the change between trials rather than treating every trial as an isolated event. In the blocking case, the later compound presentation must be interpreted in light of the animal's earlier conditioning history. The prior history changes the conditions for later learning.
Blocking Compared with New Learning
| Conditioning history | Later presentation | Role of prior learning | Expected result |
|---|---|---|---|
| A stimulus has already been involved in conditioning | The previously conditioned stimulus appears with a new stimulus | The prior stimulus already prepares the animal for what will happen | The new stimulus receives little or no additional learning in the blocking case |
| No prior conditioning history is being used in the comparison | A new stimulus is considered as a possible predictor | There is no previously learned predictor described as accounting for the outcome | This provides a contrast for seeing how prior learning changes the later outcome |
The comparison is not simply between two different stimuli. It is between two different learning histories. In the blocking procedure, the later compound is interpreted after one component has already been conditioned. That history is what changes the outcome for the newly added stimulus.
Alternative Conditioning Models
The Rescorla-Wagner model is one model among several used to study classical conditioning. The source lists models associated with Klopf, Grossberg, Mackintosh, Moore and Stickney, Pearce and Hall, and Courville, Daw, and Touretzky. This broader list is a reminder that Rescorla-Wagner is one explanation within a larger history of animal learning theory.
| Model or model family | How to use the comparison |
|---|---|
| Rescorla-Wagner | Study the connection between learning and surprise, especially how prior conditioning affects later learning. |
| Klopf | Recognize this as another model associated with the study of classical conditioning. |
| Grossberg | Recognize this as another model associated with the study of classical conditioning. |
| Mackintosh | Recognize this as another model associated with the study of classical conditioning and animal learning. |
| Moore and Stickney | Recognize this as another model associated with the study of classical conditioning. |
| Pearce and Hall | Recognize this as another model associated with the study of classical conditioning. |
| Courville, Daw, and Touretzky | Recognize this as another model associated with the study of classical conditioning. |
The source names these alternatives but does not provide their mechanisms in this section.
Common Misunderstandings
Treating each conditioning trial as independent
Kamin blocking depends on the earlier conditioning history involving Cue A.
Fix:
Trace the learning across stages and ask what the animal already expects before the later trial.Assuming that adding a stimulus automatically produces new learning
Prior conditioning may already account for what happens, leaving little or no additional learning about the added stimulus.
Fix:
Ask whether the new stimulus creates surprise beyond what the earlier cue already predicts.Treating reinforcement as a term with one permanent meaning
The source describes different uses associated with the Pavlov translation, Mackintosh, and Skinner.
Fix:
Identify the historical tradition and the meaning of reinforcement being used.Presenting Rescorla-Wagner as the only conditioning model
The model is part of a broader history of theories of classical conditioning.
Fix:
Place Rescorla-Wagner alongside the models associated with Klopf, Grossberg, Mackintosh, Moore and Stickney, Pearce and Hall, and Courville, Daw, and Touretzky.
Practice Trace
An animal has a prior conditioning history with Cue A. In a later stage, Cue A appears together with a new Cue B. Trace the two stages and explain why the later presentation may produce little or no additional learning about Cue B.
Hints
- Begin by stating what was learned about Cue A in the first stage.
- Next, identify what changes in the second stage: Cue B is added to the previously conditioned cue.
- Finally, connect the outcome to surprise and explain why prior conditioning can reduce new learning.
- Kamin blocking is best understood as a change across conditioning trials. Prior conditioning involving one stimulus can block later learning about a stimulus added alongside it. The Rescorla-Wagner model explains this pattern by emphasizing surprise: when earlier learning already prepares the animal for the outcome, the later event provides little additional surprise. The meaning of reinforcement has also changed across historical traditions, and Rescorla-Wagner is one model within a broader group of theories of classical conditioning.
Key Takeaways
- Kamin blocking demonstrates that prior conditioning can affect subsequent learning.
- The blocking procedure has two stages: first condition one stimulus, then present it with a new stimulus.
- The Rescorla-Wagner model links learning with surprise and emphasizes the change between trials.
- Different historical traditions have used the term reinforcement in different ways.
- Rescorla-Wagner is one of several models used to study classical conditioning.