Peterson & Peterson (1959): STM duration

Nearby memory studies get swapped, and a wrong finding scores nothing. Keep them apart: Peterson and Peterson = duration of short-term memory (under 10% by 18 seconds); Miller = capacity, about seven items; Baddeley = how memory codes information; Murdock = how list order affects recall. "The capacity is seven items" answers a different study.

The mark depends on specific figures, so "they forgot the letters quickly" measures nothing and does not score. Give two anchors: about 80% recalled after 3 seconds, under 10% after 18 seconds. A safe sentence: recall fell from about 80% at 3 seconds to under 10% at 18 seconds, so short-term memory holds unrehearsed information for roughly 18 seconds.

A result is the figure — under 10% recalled after 18 seconds. A conclusion is what it tells us about memory — short-term memory holds unrehearsed information for only about 18 seconds, so rehearsal is needed to keep it there. A conclusion that just restates numbers is credited as a result and scores zero. The test: a claim about memory, or only about this data?

A weakness is something wrong with the study — the trigrams are meaningless, so ecological validity is low. An improvement is what you would change next time, plus its benefit: repeat with meaningful material such as short phrases, so the duration found applies to material people must remember. An improvement is a change plus its benefit; restating the weakness is not.

"A strength is that it was controlled" names a point and stops — flagged every sitting. A creditworthy evaluation has three parts: the point, why it follows, and the consequence. E.g. the study was tightly controlled because only the interval changed while the trigrams, counting and pacing stayed the same, which means the fall in recall is due to the interval itself.

Candidates often set the study out perfectly and never touch the scenario — AO1 only, and because AO3 cannot be awarded without AO2, it can lose the evaluation marks too. Not application: the character's name, the stem, or a lifted scenario line. Application means taking a specific detail — the interruption that blocked rehearsal — and reading it through the study.

"Short-term memory holds 18 items" swaps duration for capacity: this study measured how long (about 18 seconds unrehearsed), not how much. "Memory only lasts 18 seconds" is too broad — it holds only when rehearsal is prevented, only for meaningless material, only for short-term memory. And counting was not what was tested: it is a distractor task blocking rehearsal.

"The study is useless because it is not real life" loses marks. The artificial material is deliberate: meaningless trigrams stop links to existing knowledge, buying the control. Name the trade-off — confident on cause and effect, cautious on everyday memory. Sharper lines: the interference explanation and the narrow sample. Ethics is the weakest criticism here.