Memory

This is Unit Six of Psychology, Part Two: getting information back, and losing it. Part One covered encoding and storage. This part covers retrieval — and it opens with three ways of measuring whether something was remembered, which turn out to give three different answers about the same person and the same material. Then forgetting, which has more than one cause. Some of it is interference from other learning, in either direction. Some of it is loss after injury, in either direction. And some of it was never a memory at all, because the information never got encoded. The middle of this part is the most consequential material in the unit. Memory is not a recording. Information encountered after an event can change the memory of that event, a memory's content can be right while its source is wrong, and — the important one — how vivid a memory feels and how confident the rememberer is are not evidence that it is accurate. And the part closes on what actually works. Two findings with direct practical consequences, and a procedure for building a study plan out of them. By the end you will be able to do five things. One: name the three measures and say which demands most. Two: say what a retrieval cue is and how matching helps. Three: tell the two interferences and both amnesias apart. Four: say why confidence is not evidence of accuracy. Five: build a study plan from the research.

Three entries, and they are ordered by how much help each gives. The first gives none. The second hands you the answer and asks whether you have seen it. The third gives no help at all and still detects memory the other two miss.

The most demanding.

Psych 0676 Recall

Retrieving information without cues that contain the answer. The hardest of the three, and the one an essay question asks for.

The easiest.

Psych 0677 Recognition

Identifying previously learned information when it is presented again. Much easier, which is why studying by rereading feels like it worked.

And the subtlest.

Psych 0679 Relearning

Measuring memory by how much faster material is learned a second time. It detects retained memory in cases where neither of the other two measures finds anything at all.

Three entries on what brings something back. A cue is anything that helps — and two entries name a systematic way of arranging for one, by making the conditions at retrieval resemble the conditions at encoding.

The general case.

Psych 0681 Retrieval Cues

Any stimulus that helps bring stored information back into awareness. Which means retrieval is not a property of the memory alone — it depends on what is present.

Outside.

Psych 0683 Context-Dependent Memory

Improved retrieval when the external setting matches the setting of encoding. The room becomes part of the cue set without anybody arranging it.

Inside.

Psych 0685 State-Dependent Memory

Improved retrieval when the internal state matches the state at encoding. The same principle applied to something you carry with you rather than something you stand in.

Two entries. The shape forgetting takes over time — sharply at first and then levelling — and then a category of failure that is not forgetting at all, because there was never anything there to lose.

The shape.

Psych 0687 The Forgetting Curve

Memory for new material drops sharply at first and then levels off. Most of the loss happens early, which is what makes the timing of review matter.

And a different failure.

Psych 0689 Encoding Failure

The absence of a memory because the information was never stored, rather than because it was lost. Part One's first operation failing — and it is indistinguishable from forgetting from the inside.

Four entries in two pairs, and both pairs are distinguished by direction. Older learning disrupting newer, or newer disrupting older. Losing the ability to form new memories, or losing the ones already formed.

Older on newer.

Psych 0691 Proactive Interference

Older learning disrupting the retrieval of newer learning. What you already knew getting in the way of what you learned since.

Newer on older.

Psych 0692 Retroactive Interference

Newer learning disrupting the retrieval of older learning. The same mechanism, running the other way.

Forward loss.

Psych 0694 Anterograde Amnesia

The inability to form new explicit memories after brain injury or illness. Exactly the deficit Part One's hippocampus entry predicts — required to form, and not to retain.

Backward loss.

Psych 0695 Retrograde Amnesia

The loss of memories formed before the onset of injury or illness. The other direction, and a different structural story from the first.

Four entries, and taken together they are the most consequential material in this unit. Memories can be altered after the fact, their source can be wrong while their content is right — and none of that shows up as uncertainty in the person remembering.

Altered afterwards.

Psych 0697 The Misinformation Effect

Information encountered after an event altering a person's memory of that event. The memory changes and the person has no sense of it having changed.

The vivid case.

Psych 0699 Flashbulb Memories

An unusually vivid and detailed memory of the circumstances surrounding a shocking event. Unit Two's amygdala, doing the strengthening — the vividness is real.

And the error.

Psych 0700 The "Vivid Means Accurate" Error

Treating a memory's vividness, or the rememberer's confidence, as evidence of its accuracy. The two come apart — a memory can be intensely vivid, held with complete confidence, and wrong.

One more way.

Psych 0702 Source Monitoring Errors

Remembering information correctly while misattributing where it came from. Part One's semantic memory, detached from its occasion, with the detachment becoming the problem.

Four entries, and this is the part with immediate use in it. Two findings that contradict how most people study, and then two procedures — one an ancient technique for ordered lists, and one for assembling a plan from all of it.

The first finding.

Psych 0704 The Testing Effect

Improved long-term retention from retrieving material rather than reviewing it. Testing yourself is not a way of checking whether you learned it — it is a way of learning it.

The second.

Psych 0706 Distributed Practice

Study spread across separated sessions producing better retention than the same total time massed together. Same hours, better result, entirely from the spacing.

Watch an ordered list get encoded.

Psych 0708 The Method of Loci

Encoding an ordered list by placing each item at a distinct location along a familiar route. It converts a list with no structure into one that borrows the structure of a place you already know.

Watch a plan get built.

Psych 0710 Building a Study Plan from Memory Research

Selecting encoding depth, spacing, retrieval practice and cue matching for a specific body of material. Four decisions, each one traceable to a numbered finding in this unit.

One capstone, and it uses the unit's own structure as a diagnostic. Somebody studied and failed anyway — and the question is which operation broke. Encoding, storage, retrieval, or the choice of method. Four candidates, and the answer determines what to do differently.

The capstone. One failure, located.

Psych 0712 Why You Failed the Test You Studied For

Locating which memory operation broke — encoding, storage, retrieval, or the choice of practice method. The unit's opening division, used as a diagnostic tool rather than as a description.

Five things you can now do, and Unit Six is complete. You can name the three measures — recall, recognition, relearning — and say that the first demands most and the third detects what the others miss. You can say what a retrieval cue is, and that matching the external setting or the internal state at encoding improves retrieval. You can tell the two interferences apart by direction, and the two amnesias by whether new memories cannot form or old ones were lost. You can say why confidence is not evidence: memories can be altered by later information, sources can be misattributed, and vividness comes apart from accuracy entirely. And you can build a study plan from the research — retrieval over review, spacing over massing, meaning over repetition, and cues that will be there at the test.