sci-slides

name	sci-slides
version	1.0.0
description	Comprehensive guide for creating effective academic presentation slides for STEM fields. Use this skill when users ask to create scientific presentations, review slide design, need guidance on slide content density, bullet point limits, text vs visuals, equation presentation, figure adaptation, or want to avoid common presentation mistakes. Applies to conference talks, lab meetings, journal clubs, and all academic presentation contexts.

Scientific Presentation Slides

Evidence-based guidelines for creating effective academic presentations in STEM that respect cognitive principles and enhance audience comprehension.

---

Overview

This skill provides research-backed guidance for creating scientific presentation slides that communicate complex STEM content effectively. The guidelines are grounded in cognitive load theory, multimedia learning research, and empirical studies of presentation effectiveness.

Key insight: 100% of analyzed academic presentations violated basic cognitive principles. Following evidence-based guidelines provides significant competitive advantage while better serving scientific communication.

When to use this skill:

• Creating new academic presentations
• Reviewing and improving existing slides
• Adapting publication figures for presentations
• Presenting equations or technical content
• Seeking guidance on slide structure and design

---

Core Design Principle: Assertion-Evidence

Replace topic headings with full-sentence assertions that state findings.

Traditional approach (avoid):

• Slide title: "Results"
• Content: Bullet points listing observations

Evidence-based approach:

• Slide headline: "Catalyst efficiency increases 40% at higher temperatures"
• Content: Supporting graph or visual evidence

Why this works:

• Forces clarity about the actual message
• Provides clear takeaways even if audience misses narration
• Eliminates cognitive conflict between reading and listening

Rule: Each slide should communicate ONE clear message supported by visual evidence, not bullet points.

---

Typography & Readability Standards

Font Requirements

Element	Specification
Minimum size	24 pt (28 pt for challenging conditions)
Font family	Sans-serif only (Arial, Helvetica, Calibri)
Contrast	Dark gray text on white backgrounds

Test: Every slide must be readable from the farthest seat in the presentation room.

Text Density Limits

Element	Maximum
Lines per slide	6-9 lines
Bulleted items per list	5 items
Lines per bullet	2 lines
Words per bullet	2-10 words (not complete sentences)

Key principle: Audiences read faster than speakers talk. Text-heavy slides create unsolvable cognitive conflict.

---

Visual Evidence Guidelines

Core Rules

1. Replace text with visual evidence whenever possible
2. One visual message per slide (respect 3-4 chunk working memory limit)
3. Maximize data-ink ratio - remove decorative elements
4. Large, simple figures over complex multi-panel figures
5. Direct annotation - use arrows and text boxes to guide interpretation

Adapting Published Figures

Critical distinction: Journal figures can include extensive detail because readers control viewing duration. Presentation figures must communicate key findings in seconds.

Adaptation workflow:

1. Identify the specific point the figure should support
2. Simplify by reducing data series
3. Enlarge all labels (minimum 24 pt)
4. Remove secondary information
5. Consider creating entirely new visualizations focused on the key message
6. Add direct annotations (arrows, text boxes) highlighting critical features

Common error: Copying complex multi-panel figures directly from papers into slides. Audiences lack time to decipher them.

---

Presenting Data

Graphs vs Tables

Strongly prefer graphs (processed 60,000x faster than text)

Use graphs for:

• Trends over time (line graphs)
• Relationships between variables (scatter plots)
• Comparisons between groups (bar charts)
• Patterns and distributions

Reserve tables only for: Exact numerical lookup requirements

Data Visualization Rules

Do:

• Remove gridlines unless essential for reading values
• Eliminate 3D effects (distort perception)
• Remove shadows, gradients, decorative patterns
• Choose graph types that match the data story

Avoid:

• Software defaults that serve no scientific purpose
• Misleading scales or inappropriate graph types
• Excessive chartjunk

---

Mathematical Content

Equation Presentation

Golden rule: Only include equations explicitly explained during the presentation.

Best practices:

1. Decompose complex equations into components
2. Discuss each term separately while graying out others
3. Provide plain-language explanations alongside equations
4. Show graphical behavior through plots demonstrating how the equation works
5. Define all terms both verbally and visually
6. Use LaTeX for equation formatting (import as high-resolution images)

For short talks: Minimize equation density. Displaying complex equations without decomposition and explanation fails to communicate.

---

Cognitive Science Foundations

Working Memory Limitations

Key constraint: Humans can hold only 3-4 chunks of information simultaneously in working memory.

Implication: Text-heavy slides + narration = cognitive overload on the same mental resources.

Solution: Employ dual modality by combining visual graphics (processed by visual channel) with auditory narration (processed by verbal channel).

Retention Research

After three days:

• 10% retention from purely verbal presentations
• 35% retention from purely visual presentations
• 65% retention from combined verbal and visual presentations

Mayer's Key Principles:

1. Redundancy principle: Graphics + narration outperform graphics + narration + identical text
2. Coherence principle: Extraneous elements interfere with learning
3. Multimedia principle: Words + pictures significantly outperform words alone

---

Critical Mistakes to Avoid

1. Excessive Text

Problem: Creates cognitive conflict - audiences read in 10 seconds, then face boredom or divided attention.

Solution: Limit to 6-9 lines maximum, or replace with visual evidence.

2. Reading From Slides

Problem: Signals poor preparation; written prose sounds pompous when read aloud.

Solution: Use conversational language in delivery; slides should support, not script, the talk.

3. Unreadable Content

Problems:

• Fonts below 24 pt
• Insufficient contrast
• Acknowledging "I know you can't read this"

Solution: Test readability from the farthest seat. Never present unreadable content.

4. Decorative Elements

Problem: "Seductive details" (clip art, background patterns, sound effects) interfere with message retention. Audiences remember decoration instead of findings.

Solution: Remove all elements that don't convey scientific information.

5. Assuming Specialized Knowledge

Problem: Using jargon or acronyms without definition, even at specialized conferences.

Solution: Define technical terms on first use. Mixed expertise levels exist in all audiences.

---

Workflow

When helping users create or review scientific presentations:

1. Structure Development

• Identify 2-5 clear messages audiences should remember
• Create full-sentence assertion headlines for each message
• Determine visual evidence that supports each assertion

2. Apply Design Standards

• Font: Sans-serif, minimum 24 pt
• Text: Maximum 6-9 lines, 5 bullets, 2 lines per bullet
• Visuals: Simplify published figures, annotate directly
• Equations: Only include if explicitly explaining; decompose complex ones

3. Content Review

• Remove decorative elements
• Replace text with visual evidence where possible
• Verify one message per slide
• Check readability from back row

4. Cognitive Load Check

• Does the slide respect 3-4 chunk working memory limit?
• Is there cognitive conflict between reading and listening?
• Are verbal and visual channels used effectively?

5. Mistake Prevention

• No text-heavy slides (≤6-9 lines)
• No reading from slides (conversational delivery)
• All content readable from back row
• No decorative elements
• All jargon defined

---

Success Criteria

A well-designed scientific presentation:

• Has 2-5 clear messages audiences will remember
• Uses full-sentence assertion headlines
• Presents visual evidence, not text bullets
• Respects working memory limitations (3-4 chunks)
• Is readable from the back row
• Eliminates decorative elements
• Adapts published figures for quick comprehension
• Explains equations explicitly or omits them
• Maintains engagement through variety and enthusiasm