6.14 Learning Theories and Observational Learning

There exist learning phenomena that aren't neatly categorized into classical or operant conditioning.
These types of learning allow for learning without direct and explicit reinforcement.

Latent Learning: Learning that occurs without immediate reinforcement or realization, often automatic and passive.

Edward Tolman, a psychologist, famously used rats in mazes to investigate latent learning.

Three groups of rats were tested under differing reinforcement conditions:
- Group 1 (Purple Line): Rats that never received reinforcement regardless of maze completion.
- Group 2 (Teal Line): Rats that were always reinforced with a reward (e.g., banana pellets) upon reaching the maze's end.
- Group 3 (Dark Blue Line): Rats not reinforced for the first 10 trials but began receiving rewards from the 11th trial onwards.

Group 1 (Never Reinforced): Maintained more errors throughout the trials.
- Lack of reinforcement led to little incentive to learn the maze efficiently.
Group 2 (Always Reinforced): Showed a significant reduction in errors over time due to consistent reinforcement.
Group 3 (Delayed Reinforcement): Initially, performed similarly to Group 1 with many errors, but exhibited a sharp decrease in errors once reinforcement was introduced after the 10th trial.
- This indicated they had acquired knowledge of the maze without expressing that learning until reinforcement was present.

Learning occurs even in the absence of reinforcement.
Demonstrates a form of knowledge retention that is not expressed until required or relevant.
Real-world example: Navigating a campus without explicit reinforcement can result in an internal map that can be utilized when needed (e.g., meeting at the Bell Tower).

Observational Learning: Learning that occurs by witnessing the consequences of others' behaviors rather than through direct reinforcement or punishment.

Found in New Caledonia, these crows demonstrate tool-making behaviors, breaking twigs and fashioning them to extract grubs from trees.
The methods of tool creation vary among communities, showing that techniques are learned through social observation and not genetically determined.

Bandura's experiment aimed to investigate whether children learn through observation.
Bobo Doll: An inflatable clown toy that returns upright after being hit.

Children were divided into groups:
- One group observed an adult exhibiting aggressive behavior towards the Bobo doll, hitting and kicking it.
- Another group saw an adult interact peacefully with the doll without aggression.
After observation, children were allowed to play with toys unsupervised.

Children exposed to aggressive adult behavior imitated those violent actions toward the Bobo doll.
Conversely, children who witnessed peaceful interactions replicated non-violent actions.
Children are influenced by adult behavior models, which can lead to unintended learning outcomes, including swearing or aggression.
Highlights the adaptive nature of observational learning as it can prevent individuals from experiencing direct risks associated with certain behaviors.

Developed four requirements for successful observational learning:
1. Attention: The learner must focus on the behavior of others; without attention, no learning occurs.
2. Memory: The learner needs to remember the observed behavior for future recall.
3. Motor Reproduction: Physical capability to reproduce the observed behavior is necessary; one can’t replicate skills without the necessary strength or flexibility, as with athletic examples.
4. Motivation: There must be an expectation of reinforcement when reproducing the behavior; without motivation, the likelihood of using the behavior as a guide diminishes.