The Process of Encoding Experiences Into Long-Term Memory

The study of memory, and how memory is encoded, has long been debated within psychology. While some researchers focus on the different parts of long-term memory, this essay focuses on how experiences encode into long-term memory, with a specific focus on predictor error signals, closely linked to the Rescorla-Wagner rule, and consolidation. Concisely, a predictor error signal can be described through the notion that when an individual predicts an event incorrectly, the prediction error will be large, and a great deal will be learned. 

This theory is linked closely to the Rescorla-Wagner rule as it emphasises the importance of surprise and unexpected events in determining whether an experience is encoded into LTM. However, other researchers focus on the process where, after the acquisition, memory splits into either short-term or long-term state, and memory consolidation is when experiences in the short-term memory encode itself into the long-term memory. 

This theory can easily explain how there appears to be an aspect of unconscious selectivity in what experiences are encoded into long-term memory. Despite this, I would argue that predictor error signals provide a more apt theory to explain why experiences are encoded into the long-term memory than consolidation, as there’s a wide bank of research to support the theory and also real-life situations where psychologists can assume predictor error signals are taking place.

There is much evidence to support the theory of prediction error signals. This could be seen through the developmental psychologist Baillargeon’s ‘Violation of Expectation’ research, where they compared infant reactions to expected and unexpected, surprising, events. The findings of this study indicated that infants looked significantly longer at the impossible event, and when the infant learned from the impossible event, they soon became disinterested. 

Therefore, this research could support the theory of prediction error signals being vital for learning, as Baillargeon indicated that infants are more intrigued and ready to learn when faced with an unexpected task that violated their expectations of reality. This is also closely linked to and provides real-life evidence for, the Rescorla-Wagner rule of ΔV= α β (λ - V), where the more an individual knows (ΔV), the next amount of learning becomes smaller. Despite this, the Rescorla-Wagner rule has often been criticised for its vastly theoretical approach and non-biological evidence for this encoding of memory. 

However, psychologist Berns et al (2001) conducted research that indicated through fMRI scans, when learning occurs in a human brain, there is consistent activation of the nucleus accumbens, an area where dopaminergic neurons project to the ventral tegmental area, in humans. This therefore indicates that there is biological evidence to support the theory that prediction error signals exist and help information be encoded into the long-term memory. 

While this research does not indicate how prediction error signals specifically regulate learning, which is what the Rescorla-Wagner rule suggests they do, it does prove that prediction errors do exist in humans, providing sound biological evidence and rejecting the view that the Rescorla-Wagner rule is wholly theoretical. In turn, it appears that predictor error signals are vital in the process of encoding experiences into long-term memory, as indicated by Baillargeon’s infant studies and the biological evidence within the nucleus accumbens of human brains.