Presentations: how to build slides and the spoken talk around them
Question: how should slides and the talk around them (research talk, conference presentation, lecture) be built so the listener understands and remembers what matters, given that a listener cannot set the pace or reread.
Scope: only what changes when a speaker paces the medium and is part of it, including motion on slides and in video (strand 7) and colour (strand 8). Text on a page is covered elsewhere: learning science (section 5, multimedia principles), typography and layout, and the general principles in the standard.
Steps: frame · protocol · sources · review · state of the art · position · against · baseline · bibliography
Claims
| Claim | Grade | Where |
|---|---|---|
| Showing slides does not by itself raise learning; design decides | measured (meta-analysis, 48 studies) | review.md 2 |
| A slide set that bundles sentence headlines, visual evidence and little text beat a default topic-and-bullets set (d about 0.8 to 0.9, n = 110) | supported with caveat (one lab experiment, one team, bundled factors) | review.md 2 |
| Slide faults of text density and clutter are common in every field, academic talks included; one UK medical school showed text-heavy slides 84.4% of the time | measured, descriptive | review.md 2 |
| Narration plus verbatim on-screen text does not beat text alone; key terms on screen beat verbatim text | measured (meta-analysis, 57 studies) | review.md 1 |
| Learners prefer the redundant slide that teaches them less | supported with caveat (two experiments) | review.md 1 |
| Speaker gestures give listeners a moderate comprehension benefit | measured (two meta-analyses) | review.md 3 |
| A visible instructor in video has no consistent learning benefit | supported with caveat (12 studies) | review.md 3 |
| Conference talks have a rhetorical structure that differs from the paper | measured, descriptive (44 talks) | review.md 5 |
| Minimum projected sizes (18, 24, 30 point) are conventions with no study of comprehension at distance found | convention, untested | review.md 4 |
| The "attention falls after 10 minutes" rule has little primary support | measured (reviews of primary studies) | against.md |
| Practice (30-point rule, 20 slides in 20 seconds, memos instead of slides) runs ahead of the evidence; the one randomised trial of 20 by 20 found equal test scores | convention, untested | review.md 6 |
| Good multimedia design matters more in system-paced settings such as lectures (overview of 29 reviews) | measured | review.md 0 |
| Talk advice from experts agrees across 91 papers, while experiments are scarce | measured, descriptive | review.md 0 |
| Accessibility guides ask speakers to say slide text aloud, which is untested against the redundancy result | convention, untested | review.md 6 |
| Key points shown in step with the narration beat the whole slide at once in two small recorded studies (n = 99, n = 40) | supported with caveat | review.md 7 |
| Live coding taught no more than static code in one controlled study; active live coding taught no more than traditional live coding (about 530 students) | supported with caveat | review.md 7 |
| Video added to teaching has a larger effect (g = 0.80) than video replacing it (g = 0.28) | measured (105 trials, abstract) | review.md 0, 7 |
| Animation against slides in explainer videos: engagement and liking differ, comprehension mostly does not | supported with caveat (small studies) | review.md 7 |
| Pleasant colours with faces give a small gain on retention and transfer; colour alone is 5 of 33 samples, faces and colour together 13 | measured (meta-analysis, 33 samples, N = 2,924) | review.md 8 |
| Colour coding of text and picture raised retention and transfer in one eye-tracking experiment (n = 52) | supported with caveat | review.md 8 |
| Rainbow-like colour maps distort data and defeat colour-vision deficiency; a model finds some uses | convention, argued; one counter-study | review.md 8, against.md |
| Dark text on a light ground was most legible, projected too; advice for dark rooms says the reverse | measured (three experiments) against convention | review.md 8 |
| Colour harmony is preference; no study of harmony and learning was found | measured for preference | review.md 8 |
| Colour is never the only cue (WCAG 1.4.1) | convention (a standard) | review.md 8, baseline.md |
| 20 of 39 grey sources reach 10 of 20; the rest are examples of practice | convention (a quality score) | sources.md |