Week 3: Persuasive Imagery

Week 3 Lecture: Persuasive Imagery

Discussion

1 Introduction

⏱ 10 min

Week overview, canvas handover from Week 2, visual persuasion brief

Content

2 Theory and Application

⏱ 30 min

Cognitive fluency, dual coding, semiotics, visual hierarchy, accessibility, AI imagery, evidence standards

Practice

3 Worked Example

⏱ 15 min

Applying the evidence framework to the Villa College visual asset set

Assessment

4 Attendance and Exit Check

⏱ 5 min

QR attendance, Moodle concept check

Module Arc

Weeks 1–2

Foundations and Planning

→

Weeks 3–5

Content and Media

→

Weeks 6–8

Acquisition Channels

→

Weeks 9–12

Relationships and Measurement

→

Weeks 13–15

Governance and AI

We move into Weeks 3 to 5: Content and Media. Your Week 2 canvas is now your creative brief. Every visual decision this week must trace back to a canvas entry.

Today’s Agenda

Lecture: nine theory and practice sections

Why images persuade: the 13-millisecond window
Cognitive fluency theory: familiarity and processing ease
Dual coding theory: verbal and visual congruence
Semiotics and connotation: denotation, connotation, and audience-specific meaning
Visual hierarchy and Gestalt principles: directing the eye through contrast, scale, and placement
Platform format requirements: to-scale formats and safe zones
Accessibility standards: WCAG 2.2 and colour contrast
AI-generated imagery: uses, limits, and disclosure obligations
Image evidence standards: the five-level evidence pyramid

Tutorial preview

You will build a real campaign asset in Canva:

Write a three-line design brief from your Week 2 canvas
Build your hero asset at correct platform dimensions (40-minute sprint)
Swap screens for a three-question critique carousel
Fix, export a second size, and run the contrast check on your own colours
Write alt text and complete the one-page design decision record
Sign up for a free Canva account before the session

Learning Objectives (1 of 2)

After studying this chapter, you should be able to:

Explain how cognitive fluency and dual coding theory predict audience responses to visual stimuli.
Apply semiotic principles to interpret and produce digital images with intentional connotative meaning.
Describe the principles of visual hierarchy and explain how they direct audience attention in digital formats.
Evaluate a visual asset against platform format requirements and accessibility standards.

Learning Objectives (2 of 2)

After studying this chapter, you should be able to:

Distinguish between uses of AI image generation that support campaign planning and uses that undermine evidence standards.
Produce a visual asset set aligned to a positioning claim, a hierarchy stage, and an ELM processing route.
Apply the evidence framework to visual claims: classify image choices as evidence-based, inferred, or assumed.

All seven objectives are assessed in your Week 3 submission. Objective 7 connects directly to the evidence discipline from Week 1. Trace every image choice to a canvas entry, or label it as an assumption.

Why Images Persuade

A well-designed image can establish relevance, credibility, and emotional alignment before the audience has decided whether to pay attention. A poorly chosen image can end engagement before the argument is ever encountered.

Visual response precedes verbal processing. This is a perceptual fact established by measurement rather than design opinion.

Vote: Which Ad Stops You?

Hands up: which one would make you stop scrolling? Why?

Cognitive Fluency (Reber et al., 2004)

Cognitive fluency is the subjective ease with which mental content is processed. When a stimulus is easy to process, that ease is attributed to familiarity, truth, and quality, and the attribution happens without the audience noticing it.

The table below shows where that attribution gets triggered in campaign design, element by element.

Design element	High fluency	Low fluency
Contrast	High contrast text on plain background	Light grey text on white background
Typography	Familiar face, appropriate size, sufficient spacing	Decorative or very small face, compressed leading
Layout	Conventional structure for the category	All conventions violated simultaneously

The practical balance: use familiar structures and conventions to build processing ease, then introduce one focal element of controlled novelty to earn attention, because novelty the audience struggles to process simply goes unnoticed.

Live Test: The 13-Millisecond Flash

The protocol:

Your lecturer shows a campaign asset for one second, then blanks the screen
Write down (or say): what do you remember seeing?
Compare: is the remembered element the campaign’s claim, or its decoration?

Whatever survives one second of exposure is the de facto dominant element of the design, whatever the designer intended. Recognition happens in about 13 milliseconds (Potter et al., 2014). Your one-second memory is a generous test.

Try It: Flash This Image

J. Howard Miller, *We Can Do It!*, 1943. Public domain, National Museum of American History, Smithsonian Institution.

This 1943 Westinghouse poster is public domain and a genuinely excellent teaching example: it was designed for exactly the same problem your students face this week, earning attention in a crowded space with one image and a handful of words.

Run the same protocol as the previous slide: show it for one second, blank the screen, then ask what the room remembers. Expect near-total agreement on two things: the raised fist and the words “We Can Do It.” Nothing else in the image survives the second, because nothing else competes for it.

Why this asset is a strong test case, for your own reference rather than the slide: it has one figure, one gesture, and one claim, so no secondary element competes for attention. The pose and the words reinforce the same meaning, which is the dual coding congruence test from the next section. The high contrast between the navy caption band and the yellow background does the fluency work before anyone reads a word.

Use the room’s agreement to make the point concrete: this is what a one-second dominant element looks like when hierarchy, fluency, and dual coding are all working together, and it is the standard their own asset should be tested against this afternoon.

[Timing: 2 minutes]

Dual Coding Theory (Paivio, 1986)

The brain processes verbal and non-verbal information through two distinct but interconnected systems. When a message activates both systems simultaneously, memory encoding is stronger and recall is more reliable than when only one system is engaged.

Dual Coding in Practice: Congruence Test

Headline	Congruent image	Incongruent image
“Learn at your own pace”	Student working independently at a quiet table with headphones	Students in a regimented classroom with a visible clock
“Our guesthouse puts you at the centre of the reef”	Snorkeller underwater with coral in the foreground	Generic beach with no water activity visible
“Graduate with a live campaign portfolio employers can open”	Laptop screen showing a live analytics dashboard	Diploma certificate in a frame on a wall

The visual argument test: if a viewer follows only the visual hierarchy without reading the text, do they still understand the campaign’s primary claim?

Infographics and data visualisations exploit dual coding systematically. A statistic as a number activates the verbal system only. The same statistic as a proportional visual activates both systems and produces more durable recall. Campaign teams that present evidence only in text are leaving a substantial encoding advantage unused.

Discussion: One Image, Two Audiences

The image: a wooden overwater bungalow at sunrise. Empty. Serene. No guests visible.

Left half of the room, answer as an international first-time visitor:

What does this image say to you?

Right half of the room, answer as someone who lives in Malé:

What does this image say to you?

The same pixels. Different meanings. Connotation lives in the audience’s cultural frame, and image selection must start from the audience’s perspective rather than the designer’s.

Gestalt Principles

The visual system organises elements into patterns and wholes before conscious attention is applied. Gestalt principles describe the rules of that organisation.

Violations of these principles are typically experienced as visual noise rather than as creativity.

Ten Gestalt Principles for Campaign Design

Ten principles, the same figure from your chapter reading. Each predicts a specific perceptual response, and three carry the most weight in campaign design.

Proximity is the most commonly violated. Evidence elements placed randomly around an asset feel unorganised. Moving the testimonial and the statistic adjacent to the claim they support makes the layout read as a coherent argument.

Figure-ground is the accessibility principle in disguise. When the figure lacks sufficient contrast with the ground, it fails both the Gestalt test and the WCAG contrast test simultaneously.

Continuity is about implied motion through the layout. A well-designed asset has a reading sequence: the eye enters at the largest element, moves to the secondary claim, reaches the evidence, and arrives at the call to action. Continuity governs that sequence.

The other seven (symmetry, common region, common fate, connectedness, focal point, similarity, closure) are worth a fast pass: ask students to name one campaign layout choice that would violate each.

[Timing: 2 minutes]

Visual Hierarchy: Directing the Eye

Visual hierarchy is the arrangement of elements in a layout so that the viewer’s eye moves through them in the intended sequence, achieved through contrast, scale, colour, placement, and white space.

In the absence of intentional hierarchy, audiences impose their own reading order, and that reading order may work against the campaign.

Visual Hierarchy: Six Design Levers

Principle	Mechanism	Application
Contrast	Difference in tone, hue, or texture between elements	Place the primary call to action against a high-contrast background; ensure headline text meets WCAG AA contrast ratio
Scale	Larger elements are perceived as more important	Make the most important element the largest; keep the hero image and the headline at different scales so neither competes with the other
Colour	Hue and saturation guide attention and carry associative meaning	Use your brand palette consistently; reserve a high-saturation accent colour for the primary action only
Proximity	Elements grouped together share perceived meaning	Group evidence elements near the claim they support
White space	Empty space increases the salience of surrounded elements	Give the primary message space; a cluttered layout reduces the salience of all elements equally
Alignment	Elements on a shared axis are perceived as related	Align text blocks to a grid; misaligned elements signal low production quality

Eye-Path Patterns: F and Z

Place the headline at the top left. Place the call to action at the bottom right (Z-pattern) or where the left-edge scan arrives (F-pattern). Place evidence elements along the scanning path rather than in the corners.

Eye-tracking research shows these two dominant patterns. Text-heavy pages, like long landing pages, tend to produce F-patterns. Visual-dominant ads tend to produce Z-patterns.

The practical implication for campaign assets is specific. For a social media image: headline top left or centre, primary claim as the largest element, evidence mid-layout, call to action bottom right. For a landing page with dense text: the first two lines of every section carry the most attention, so use them for the claim and the key evidence rather than for preamble.

Keep the call to action away from the top right in a Z-pattern layout: the eye arrives there second after starting at top left, before the evidence has been encountered. The decision to act should follow the evidence rather than precede it.

[Timing: 1.5 minutes]

Accessibility Standards: WCAG 2.2

An inaccessible visual asset excludes a portion of the audience. In digital marketing contexts, accessibility is both an ethical requirement and a practical one, measured against WCAG 2.2, the international standard for digital content accessibility.

WCAG 2.2 Colour Contrast Requirements

Minimum contrast ratios

Text type	WCAG AA (minimum)	WCAG AAA (enhanced)
Normal text (below 18pt)	4.5:1	7:1
Large text (18pt or 14pt bold)	3:1	4.5:1
UI components and graphics	3:1	unspecified

Free checking tools (no sign-up required): - WebAIM Contrast Checker: webaim.org/resources/contrastchecker - Coolors Contrast Checker: coolors.co/contrast-checker - Colour Contrast Analyser (desktop app, free download)

Beyond contrast: colour-blind-friendly design

Approximately 8 per cent of men and 0.5 per cent of women have some form of colour vision deficiency.

Avoid red-green combinations as the sole distinguishing element
Use shape, pattern, or text label as a secondary distinguisher alongside colour
Test your palette in a colour-blindness simulator before production
Never rely on colour alone to convey meaning (error states, calls to action, data categories)

Alternative Text: The Second Accessibility Requirement

Every image used in a web page, email, or accessible document requires a text alternative.

The standard: alt text should describe what the image communicates rather than what it literally depicts.

Image	Poor alt text	Strong alt text
Three guests on a Maldivian beach at sunset	“Three people at a beach”	“Three guests celebrating their first dive day at a Maldivian guesthouse, Rasdhoo”
Analytics dashboard showing a campaign result	“Screenshot of a graph”	“Bar chart showing 34% increase in direct bookings after the summer social campaign”
A student working at a laptop	“Person using a computer”	“A Villa College student completing the Week 3 visual asset task in the campus study room”

Alt text is the image’s verbal equivalent. Where the image makes an argument, the alt text carries that argument for users who rely on screen readers.

AI-Generated Imagery

AI image tools can produce visual concepts rapidly, but direct photographic evidence from a real context remains irreplaceable: AI carries no guarantee of accuracy for depicted settings or people, and introduces significant transparency obligations.

The core risk: AI-generated people and places presented as real customers or locations.

AI Image Decision Framework

AI Image: Permitted and Restricted Uses

**PERMITTED:** abstract texture, nothing real implied

**PERMITTED:** icon illustration, no photographic accuracy claimed

**PERMITTED WITH DISCLOSURE:** concept sketch for briefing, labelled AI-generated

**RESTRICTED:** a generated person shown as a real customer

**RESTRICTED:** a photorealistic setting implying a real, bookable resort

**RESTRICTED:** an AI stand-in implying manufacturing accuracy

Every AI-generated image used in any submission must appear in the AI prompt log with the exact prompt, tool, version, and output used. Images without a log entry must be removed.

Visual Evidence Standards

An image used in a campaign is making an implicit claim: that the product looks like this, that customers feel this way, that the environment is like this. Those claims can be evidence-based, inferred, or assumed.

The evidence framework from Week 1 applies to visual choices in exactly the same way it applies to text claims.

The Visual Evidence Pyramid

The standard concerns documentation rather than origin: any image type is acceptable when the choice is documented, justified, and consistent with the positioning claim. If you use stock photography, your workbook must state why it was selected and when it will be replaced with owned photography.

Platform Format Requirements: To Scale

Platform Format Requirements: Full Specification

Platform	Recommended size	Aspect ratio	Safe zone	Max file size
Facebook Feed	1200 x 628 px	1.91:1	Central 80% for text	30 MB
Facebook Story	1080 x 1920 px	9:16	Avoid top and bottom 14%	30 MB
Instagram Feed (square)	1080 x 1080 px	1:1	Full frame visible	30 MB
Instagram Story / Reel	1080 x 1920 px	9:16	Avoid top and bottom 14%	30 MB
LinkedIn Single Image	1200 x 627 px	1.91:1	Central 80% for text	5 MB
Google Display (leaderboard)	728 x 90 px	8.09:1	Legible at display size	150 KB
Google Display (rectangle)	300 x 250 px	1.2:1	Legible at display size	150 KB
WhatsApp Status	1080 x 1920 px	9:16	Avoid top and bottom 14%	16 MB

Platform specifications change without notice. Always check the current official documentation before the production phase. Cite the specification source and the date you accessed it in your workbook.

These specifications are verified as of May 2026. Any one of them may change. The LinkedIn file size limit of 5 MB is considerably tighter than the others: assets that are file-size compliant for Instagram may need recompressing before LinkedIn upload.

The safe zone requirement for Story formats eliminates the top 14 per cent and bottom 14 per cent from usable text space. Those zones are covered by the platform UI: the profile picture and the progress bar at the top, the “send message” bar at the bottom. Any text placed in those zones will be hidden on delivery.

Document the specification source with its access date in your workbook. If the specification changes between your planning date and your submission date, you have an audit trail showing you used the correct specification at the time.

[Timing: 1.5 minutes]

Worked Example: Villa College Certificate Visual Asset

Campaign context: the Villa College Certificate in Digital Marketing, positioned at the Knowledge-to-Preference transition stage. ELM route: central (the audience is evaluating options and will read the argument).

Format: Facebook / LinkedIn Feed 1200 x 628 px · body text 14pt minimum, claim 24pt+ · navy #002147 on #edf1f7 = 12.6:1 (AA pass) · dashed line marks the 10% safe zone

The AI Protocol for Week 3

AI may assist with: generating abstract backgrounds and textures, producing concept sketches for team briefing, suggesting colour palette options, drafting alt text for review, and generating icon-style illustrations.

AI use is restricted for: generating photorealistic images of people or settings presented as real, replacing consented photography without disclosure, or producing visual content used without a documented prompt log entry.

Week 3 AI Protocol: Full Specification

Permitted and useful	Restricted
Generate abstract background textures or colour palette options	Generate photorealistic images of Maldivian beaches, hotels, or resorts presented as real
Generate concept sketches for team briefing (labelled as AI concepts)	Generate a person to represent a target segment persona without clear AI disclosure
Ask AI to check whether a colour palette is colour-blind-friendly	Replace product or service photography with AI-generated output without legal and ethics review
Ask AI to draft alt text for a photograph and then review it for accuracy	Submit AI-generated imagery without a prompt log entry
Generate icon-style or illustrative elements for infographics	Use AI to invent platform specification data or contrast ratio calculations

Every AI image used in any form in your submission must appear in the prompt log with: the exact prompt used, the tool and version, the output file, and your decision about what was accepted, modified, or rejected.

Exit Ticket

Complete the Week 3 Quiz in Moodle before moving to the tutorial.

10 questions, unlimited attempts, 70 per cent to unlock the assignment.

This is the weekly quiz (quiz-week-03.gift in the Moodle pack), separate from the three-question concept check that opens Activity 1 of the tutorial. The concept check belongs to the tutorial rather than this exit point, so keep the two distinct on this slide.

Remind students the weekly quiz can be attempted more than once and feeds their independent study, so five minutes now to start it (or finish it, if already underway) keeps them ahead of the Sunday deadline rather than starting cold on submission day.

The tutorial session starts now. The brief is on Moodle. You have two hours to build your hero asset in Canva, take a critique, export a second size, check your contrast ratio, and complete the design decision record.

[Timing: 1 minute. Total lecture: approximately 50 to 55 minutes with the three live activities.]