Forgetting

All 9 Terms & Definitions

Trace decay theory

The proposal that memory traces (engrams) fade over time through disuse, like an unused path becoming overgrown. Ebbinghaus's (1885) forgetting curve showed that forgetting is most rapid in the first hours after learning and then slows — consistent with decay. However, decay theory has been criticised: it lacks a neurological specification of how traces decay, and the finding that forgetting slows during sleep (when time still passes) suggests interference — not mere passage of time — is the primary mechanism.

Proactive interference (PI)

When older memories interfere with the acquisition or recall of new material. "Proactive" means the earlier learning acts forward in time. Example: having learned French, a student may find French words intruding when trying to recall Spanish. Underwood (1957) showed that participants with more prior list-learning experience forgot significantly more of a new list — attributable to accumulated proactive interference from earlier learning.

Retroactive interference (RI)

When new learning disrupts the recall of earlier memories. "Retroactive" means the newer learning acts backward in time. Example: learning a new work phone number causes difficulty recalling your old one. RI is strongest when old and new material are similar. Müller and Pilzecker (1900) provided early evidence, and McGeoch's (1932) interference theory proposed that RI (along with PI) is the primary cause of forgetting in verbal memory.

Retrieval failure

Forgetting that occurs not because information is absent from memory but because appropriate cues are unavailable at retrieval. Tulving and Pearlstone (1966) demonstrated this by showing that providing category name cues dramatically improved recall of "forgotten" items — proving the information was available but not accessible. Retrieval failure is distinguished from storage failure: it can be reversed by reinstating appropriate cues.

Cue-dependent forgetting

The umbrella term for forgetting that results from the absence of appropriate retrieval cues — including both context-dependent and state-dependent varieties. Based on the principle that cues present during encoding become part of the memory trace, and retrieval is most effective when these cues are reinstated. Encompasses context-dependent (external environmental) and state-dependent (internal physiological or emotional) memory effects.

Context-dependent forgetting

Forgetting that occurs when the physical environment at retrieval differs from the environment during encoding. Godden and Baddeley (1975) found that divers recalled word lists ~40% better when learning and recall contexts matched (both underwater or both on land) versus mismatched. Environmental cues present during encoding become part of the memory trace, and reinstating those cues facilitates retrieval — an application of Tulving's encoding specificity principle.

State-dependent forgetting

Forgetting that occurs when the internal physiological or psychological state at retrieval differs from the state at encoding. Goodwin et al. (1969) found that words learned while intoxicated were better recalled when intoxicated than when sober. Bower (1981) demonstrated mood-congruent recall: happy people recall more positive memories; sad people recall more negative memories. State-dependent effects arise because internal states become part of the encoded memory context.

Encoding specificity principle

Tulving's (1983) principle that a retrieval cue is effective to the extent that it overlaps with the information encoded at the time of original learning. What matters is not the strength of the cue's general association with the target, but whether the specific cue was encoded alongside the target during learning. This unifies context-dependent and state-dependent forgetting: both reflect the failure to reinstate the original encoding context at retrieval.

Reconstructive memory

Memory is not a video recording but an active reconstruction at retrieval — fragmentary traces are combined with schema-based expectations, post-event information, and context. Loftus & Palmer's (1974) car crash experiments are the canonical demonstration: a single word in a question ("smashed" vs. "contacted") altered speed estimates and caused participants to falsely remember broken glass a week later. Reconstructive memory explains why eyewitness accounts are fallible, why leading questions are dangerous, and why memory errors are systematic rather than random.