Familiar—or remembered from here?: sources & model
Explore real picture cards, try remembering with and without clues, and compare recognition with context. Keep a private record, inspect published learning results and plan a review.
The source and model records are available for inspection. No external scientific reviewer has signed off yet.
memory-tray-record-1 · content 1 · setup format 1
What supports the explanation?
Cued versus noncued recall under particular list conditions
Original publisher abstract checked. No numerical figure or universal effect size is copied. Tray-context cues differ from the study’s category-name cues.
Distinguishing item judgments and accompanying context
Original old/new, Remember/Familiar and source tasks. Their latent-process estimates are not turned into a personal product metric. The 2022 correction concerned funding.
Table 1 provides condition means and participant SDs in 89 Spanish-speaking undergraduates. No ERP truth detector or individual picture-task prediction is inferred.
Published restudy and retrieval means at three delays
Experiment 1 values are explicitly reported on page 251. Original newly drawn charts show the factual six percentages; the publisher’s figure and prose passages are not redistributed.
Three classroom experiments with children aged 9–11. Their actual results motivate care in applying adult evidence; this interface has not undergone those trials.
Original adult trivia-learning study with actual day/month delays. Tested, interpolated and fitted optima differ. The local planner performs date arithmetic only.
Actual standardized photographs and naming variation
Selected Figure 2 crops. The source states Creative Commons Attribution without a version number; that notice is preserved rather than relabeled. Adult norms do not validate a child memory score.
Selected Figure 1 crops, CC BY 4.0, with original TIFF, exact crop operations and derivative hashes. No full-database item norms or author endorsement are implied.
Educational demonstration, not a validated memory assessment, health screen or intelligence measure.
The same-item sequence changes cue availability, order, earlier retrieval and elapsed time together.
Four pictured targets and four foils are interface choices, not a universal cognitive capacity.
The actual BOSS assets are licensed figure crops, not the complete database or child-specific normative data.
Text-card, photo, paper and screen-reader experiences provide different information; no performance comparison between these modes is implied.
A limited local name matcher cannot infer every intended description. Raw attempts remain frozen.
Visible time records page exposure, not attention, effort or a neural event.
Repeated tasks or viewing the authored example change familiarity. A replay is not a fresh naive trial or proof of improvement.
Published group means come from particular participants and tasks; no personal retention forecast or forgetting curve is calculated.
Private responses remain in this page’s memory. Shared setups and videos contain no personal answers.
Tasks are not transparent brain readouts: Free recall, cued recall and recognition are operational task definitions. Their outcomes can involve studied information, associations, decision criteria, guesses and task history. An observed difference does not uniquely identify one internal process.
Cues and accessibility: Tulving and Pearlstone’s 1966 study used categorized word lists and category-name cues. Recall depended on the cue and organization. Our neutral tray context differs from their category manipulation; no effect size or hidden-memory guarantee is transferred to this activity.
Recognition is not familiarity alone: Research separates old/new judgments, remembered detail, familiar feelings and source questions in different ways. Such measures and latent-process estimates rely on assumptions. This lesson records choices and context answers without estimating how much recollection or familiarity someone has.
One sequence has several confounds: The same items move from study to recall, context cues, recognition and source questions. Earlier attempts, candidate exposure, time and order all change. This is an explanatory sequence, not a randomized efficacy trial of learning methods.
Why uncertainty needs its own place: “Not sure” is not silently converted into “new.” Unanswered trials are omitted, not guessed. In a balanced four-old/four-new set, calling everything old gives four hits and four false alarms. Looking only at hits would conceal the errors.
Source answers need their own denominator: The context question follows every “old” answer, including false alarms. A definite correct tray answer for a recognized old item differs from an item-recognition hit. Source uncertainty and omissions remain visible; no eight-item percentage is presented as a psychological measurement.
Words can create apparent scoring errors: The photographic-stimulus studies measured variation in naming, familiarity and other properties in adults. An exact word match can confuse vocabulary or spelling with remembering. The local matcher handles a small declared set of alternatives, flags duplicates and leaves other descriptions for self-checking.
Actual delayed-test results: Roediger and Karpicke’s 2006 Experiment 1 enrolled 120 undergraduates aged 18–24. Participants learned prose passages scored in 30 idea units. Learning condition was within participants; final delay was between participants. The initial retrieval tests had no feedback. The visible percentages are published group means, not each person’s result.
Children may need support: Karpicke and colleagues’ 2014 elementary-school experiments examined retrieval activities with children aged 9–11. Difficulty with unsupported recall motivated guided activities. The findings support attention to task design and help; they do not validate this specific four-card interface or a universal amount of guidance.
Spacing is not “the longer, the better”: Cepeda and colleagues varied study-to-review gaps and later test delays in an adult trivia-learning study across days and months. Best tested gaps differed from fitted or interpolated optima. The lesson therefore uses a date worksheet, not a model-generated forgetting curve or an optimal schedule for the visitor.
Related information can be endorsed by mistake: Cadavid and Beato’s 2016 study reported mean “old” responses of 65.30% for studied words, 39.31% for related unstudied lures and 11.14% for unrelated new items. The source used Spanish word lists and undergraduate participants. Those three condition means are not parts of one total or a forecast for this photo task.
An exposure record has limits: The activity records which readable cards were presented, their phase and visible page intervals after loading. Visibility is not attention or mental effort. Context labels and candidate displays change available information. A repeat within the page is labeled; earlier encounters elsewhere cannot be reconstructed by this local task.
What has been checked
Analytical reference cases, conservation or transition invariants, finite drawing commands, bounded setup parsing, discovery and route integrity are checked automatically. These checks do not establish anatomical fidelity, learner outcomes or browser/device compatibility. Independent subject review, learner trials, comprehensive accessibility review and browser video encoding checks remain pending.
Each source supports the associated claim. Sources do not certify this implementation or its visuals.
About the cover illustration
Original SUN and MOON card trays on a desk. Closed lids preserve the activity’s unseen photographs. This is physical context, not a brain model.