Do different learning types require separate mental processes?
Informing people about how often stimuli will appear together changes their expectations of future events, but does not change how much they like those stimuli.
Source
Dissociations between Learning Phenomena do Not Necessitate Multiple Learning Processes: Mere Instructions about Upcoming Stimulus Presentations Differentially Influence Liking and Expectancy
What they did
The researchers analyzed data from 181 participants in the first experiment and 208 participants in the second experiment. They instructed participants about the pairing of a nonword with an unpleasant scream, setting up four conditions that crossed high or low contingency with high or low co-occurrence. Participants then completed two trials where they rated their expectancy of the scream and their liking of the nonword.
What they found
In both studies, instructions about statistical contingency altered how much participants expected the negative sound, but did not alter how much they liked the nonword. The combined analysis revealed a significant interaction between stimulus type, contingency, and the measure used, with a p-value of .030.
The limits
What it doesn't show
First, the study only used instructions rather than actual experiences of paired stimuli, meaning it does not directly compare instruction-based learning with experience-based learning. Second, because it is a single dissociation where contingency affected expectancy but not liking, it is possible that liking is simply less sensitive to the manipulation rather than driven by a completely different process. Finally, the researchers failed to find a strong, consistent effect for the number of co-occurrences on liking ratings, meaning their manipulation of 4 versus 8 pairings may have been too weak.
Key terms
- Evaluative conditioning
- A learning phenomenon where a person's liking of a stimulus changes because it is repeatedly paired with another positive or negative stimulus.
- Expectancy learning
- A learning process in which pairing stimuli leads a person to expect the occurrence of one stimulus when the other is presented.
- Statistical contingency
- The relative frequency with which two stimuli are presented together, alone, or absent altogether, which determines how strongly one predicts the other.
- Propositional theory
- A theoretical framework suggesting that learning occurs through the formation of conscious, rule-based beliefs or propositions about the relations between stimuli.
- Referential learning
- A proposed automatic form of learning where a stimulus simply triggers the mental representation of another paired stimulus without any expectation or prediction.
- Dissociation
- An experimental finding where a manipulation affects one psychological outcome or behavior but has no effect on another, often used to argue for separate mental processes.
Flashcards
Want these cards to stick?
Save the deck to NoteFren and study it with spaced repetition.
Quiz yourself
What is the fundamental difference between how evaluative conditioning (EC) and expectancy learning (EL) are defined?
Common questions
Why did the researchers use nonsense words and a scream instead of real-world stimuli?
Using nonsense words like MORAG and an unpleasant scream ensures that participants have no prior associations or biases, allowing researchers to measure the pure effect of the instructions given during the experiment.
Does this study prove that evaluative conditioning and expectancy learning are controlled by the same brain system?
No, it does not prove they use the same brain system, but it demonstrates that a single mental process (forming propositional beliefs) is capable of explaining differences between the two, meaning we do not need to assume separate learning systems exist.
What is the difference between contingency and co-occurrence in this context?
Co-occurrence is simply how many times two things appear together, while contingency is the overall predictive relationship, which also takes into account how often one appears without the other.
Why did the researchers measure both immediate and retrospective ratings?
They measured ratings at different times to see if participants' expectancies and liking changed over the course of the experiment, and to test if memory of the instructions influenced their final judgments.
More on Learning