Retroactive interference (RI) occurs when new learning disrupts the recall of previously learned material. If you learn French vocabulary and then study Spanish, the Spanish words may interfere with your ability to recall the French words. RI is the complement of proactive interference: where PI involves old memories disrupting new learning, RI involves new learning disrupting old memories. Together, proactive and retroactive interference account for a substantial portion of everyday forgetting.
Key Structures
- Hippocampus — A medial temporal lobe structure essential for the formation of new declarative memories and spatial navigation — one of the most studied structures in cognitive neuroscience.
- Proactive Interference — A memory phenomenon in which previously learned information impairs the ability to learn and remember new information, as old memories interfere with the formation of new ones.
- Memory Consolidation — The process by which newly formed, fragile memories are stabilized into durable long-term representations, involving molecular changes, sleep, and systems-level reorganization.
- Recall — A form of memory retrieval in which previously learned information must be produced from memory without the item being physically present as a cue.
Classic Demonstrations
The standard RI paradigm involves three conditions. The experimental group learns List A, then learns List B, then is tested on List A. The control group learns List A, rests (or performs an unrelated task), then is tested on List A. RI is measured as the decrement in List A recall caused by interpolated List B learning. McGeoch (1932) demonstrated that RI increases with the similarity between the two lists — the more similar the interfering material, the greater the interference — and with the degree of original List B learning.
Mechanisms
Two mechanisms have been proposed. The unlearning hypothesis (Melton and Irwin, 1940) suggests that learning new responses to old stimuli weakens (unlearns) the original associations. The response competition hypothesis proposes that both associations remain intact but compete at retrieval, with the more recently learned response dominating. Modern evidence supports both mechanisms: some forgetting due to RI reflects true weakening of old memories, while some reflects retrievability problems that can be overcome by reinstating the original encoding context.
Jenkins and Dallenbach (1924) found that memory for nonsense syllables was better after a period of sleep than after an equivalent period of waking activity. This classic finding was initially interpreted as showing that sleep reduces RI by limiting exposure to interfering material. Modern research suggests a more active role for sleep: sleep-dependent memory consolidation actively strengthens new memory traces, making them more resistant to subsequent interference. This finding has practical implications — studying before sleep may be more effective than studying before a day full of potentially interfering activities.
Everyday Implications
RI is ubiquitous in everyday memory. Studying similar subjects back-to-back produces more RI than interleaving dissimilar subjects. Updating information (learning a new phone number, moving to a new address) inevitably creates RI for the old information. In educational settings, spacing and interleaving study sessions reduce RI by allowing consolidation between learning episodes and by promoting discriminative encoding that helps distinguish similar materials.
Disorders
- Contributes to forgetting in amnesia
- worsened by hippocampal damage
- relevant in eyewitness memory distortion