Calibration - Night of the Living Dead (1968)
Screen this public-domain film excerpt with the standardized 12-agent calibration panel. Focus on immediate viewer response to the visual and audio material in the attached clip.
Research group: 12 calibrated U.S. viewers (ages 29–62), majority rural with mixed hands-on/technical roles, plus younger urban/suburban women and both lower-income care/service and higher-income professional cohorts.
What they said: Viewers reported strong disturbance driven by the clip’s casual, matter-of-fact violence (e.g., “good shot”), describing visceral unease or a collapse of order.
Most accurately identified the on-screen intent (armed men hunting/disposing of threats) yet rejected it as advertising due to no clear CTA and reported very low likelihood to take any action; one outlier found it less intense. Main insights: Tone-not gore-drove negative affect; clarity of what was happening did not translate into persuasion; segments differed in interpretive lens (older rural visceral/procedural vs younger urban societal/ethical) but converged in refusing action.
Takeaways:
- Add explicit context and a CTA end-card to stimuli; track CTA recognition and a Tone Discomfort Index to separate comprehension from acceptability.
- Avoid “casualized violence” when seeking persuasion; if used, enforce brand-safety flags and restrict to context-appropriate placements.
- Implement content warnings/opt-outs and segment reporting to manage ethics and show shared rejection despite lens differences.
Emotional response
n=12"It definitely hit me. The shooting felt harsh and ugly, and the men's matter-of-fact reaction to it made the whole thing more unsettling than flashy horror usually does."
"What I watched hit me strongly. The casual shooting, then the matter-of-fact reactions right after, felt very disturbing and cold, and it really stayed with me because it showed a complete collapse of order and basic hum..."
"That hit me pretty hard. The killing was blunt, and the calm way the man called out 'Good shot' after made it feel cold and wrong."
"Here's the thing... I definitely felt something watching it. The violence and overall unsettling mood landed pretty hard for me, and it felt tense in a way that stuck with me."
Action intent
n=12"No. Watching that, I would not be inclined to do the thing at all. It felt dangerous, chaotic, and just plain wrong, not like something a sensible person would step toward."
"What I watched did not feel like an ad to me, so there was no clear action for me to take. My reaction was mostly discomfort and tension from the violent black-and-white film footage, not any motivation to respond or fol..."
"Here's the thing... I would be unlikely to take the advertised action because I did not actually get a clear action to take from what I watched. It just felt violent, tense, and unsettling, so my reaction was more discom..."
"What I watched felt grim and harsh, and it just is not the kind of thing that would move me to do the advertised action. I can appreciate older film craft, but the violence and overall ugliness put me off more than they..."
Message clarity
n=12"Yeah, it came through pretty plain. Watching it, I understood they were out hunting the dead ones and cleaning up the mess, and that ending drove the point home."
"What I watched was easy enough to follow scene by scene, but as an ad message it did not come through clearly to me. It felt more like grim film footage than something guiding me toward a clear point or action."
"I could follow the immediate situation, but as an ad the main message did not land clearly for me. What stood out was the armed search and the killing, so I felt tension more than a clear takeaway."
"Here's the thing... what I watched was very direct about the violent action, but as an ad the main message did not land clearly for me. I understood the tension and what people were doing, but I did not come away with a..."
Sam Norstrom
62 · Rural, NE, USA · Driver
Peace Evangelista
31 · Somerville, MA, USA · Human Resources Specialist
Sandra Falcinelli
61 · Rural, PA, USA · Designer
Kaila Smith
29 · Ann Arbor, MI, USA · Business Operations Specialist
Daniel Sassaman
55 · Rural, LA, USA · Engineer
Precious Rai
40 · Rural, IL, USA · Medical Records Specialist
Brent Guevara
52 · Fort Myers, FL, USA · Personal Care Aide
Gregory Cumbo
60 · Rural, OH, USA · Brokerage Clerk
Brianna Chapman
32 · Rural, WV, USA · Hairdresser and Cosmetologist
Maribel Miller
35 · Rural, NH, USA · Retail Sales Supervisor
John Grimm
62 · Rural, IA, USA · Civil Engineer
Mario Bockus
58 · Rural, VA, USA · Retail Sales Supervisor
Sam Norstrom
62 · Rural, NE, USA · Driver
Peace Evangelista
31 · Somerville, MA, USA · Human Resources Specialist
Sandra Falcinelli
61 · Rural, PA, USA · Designer
Kaila Smith
29 · Ann Arbor, MI, USA · Business Operations Specialist
Daniel Sassaman
55 · Rural, LA, USA · Engineer
Precious Rai
40 · Rural, IL, USA · Medical Records Specialist
Brent Guevara
52 · Fort Myers, FL, USA · Personal Care Aide
Gregory Cumbo
60 · Rural, OH, USA · Brokerage Clerk
Brianna Chapman
32 · Rural, WV, USA · Hairdresser and Cosmetologist
Maribel Miller
35 · Rural, NH, USA · Retail Sales Supervisor
John Grimm
62 · Rural, IA, USA · Civil Engineer
Mario Bockus
58 · Rural, VA, USA · Retail Sales Supervisor
| Age bucket | Male count | Female count |
|---|
| Income bucket | Participants | US households |
|---|
Gregory Cumbo
60 - Rural, OH
"Good heavens, that was loud. It sounds like they're really blasting at something now."
John Grimm
62 - Rural, IA
"Looks like something bad is about to happen, seeing men like that with rifles."
Maribel Miller
35 - Rural, NH
"Well, that was certainly a loud noise, and that person looks like they're having a real bad time."
Peace Evangelista
31 - Somerville, MA
"This loud sound requires immediate assessment to determine if there are any safety implications or if an incident report needs to be filed."
Precious Rai
40 - Rural, IL
"Seeing the men with uniforms and that rifle, it just makes me wonder what kind of serious event this truly is."
Brent Guevara
52 - Fort Myers, FL
"That poor man being led away, it just makes you dread what kind of mess he's gotten into."
Brianna Chapman
32 - Rural, WV
"Seriously, a whole belt of bullets? What are these guys even doing?"
Daniel Sassaman
55 - Rural, LA
"That looked like a pretty hard hit he just took, right there."
Kaila Smith
29 - Ann Arbor, MI
"That looks like a pretty serious situation, I'm trying to figure out what's going on."
Sam Norstrom
62 - Rural, NE
"That fella in the suit with all those shells... makes you wonder what they're doing out here."
Sandra Falcinelli
61 - Rural, PA
"Those bandoliers just confirm what I was thinking - this isn't a friendly gathering at all."
Mario Bockus
58 - Rural, VA
"That bandolier tells you everything you need to know, this ain't no casual get-together."
Overview
Key Segments
| Segment | Attributes | Insight | Supporting Agents |
|---|---|---|---|
| Older, rural participants (55+) | Age 55–62; rural locale; mixed hands-on and technical occupations (driver, civil engineer, brokerage clerk); majority White | This group gave the most visceral responses-descriptions such as 'pit in my stomach'-and emphasized the clip's blunt, matter-of-fact violence. They were able to articulate the scene's practical intent (men hunting/disposing of threats) more clearly than other groups, yet uniformly rejected any advertised action. | Sandra Falcinelli, Sam Norstrom, Daniel Sassaman, Gregory Cumbo, John Grimm |
| Younger, urban/suburban women (late 20s–early 30s) | Age ~29–31; city/suburban residents (Somerville, Ann Arbor); college-educated | Members felt strong discomfort but framed it through social and systemic lenses (e.g., 'collapse of order', 'cold'), focusing less on procedural detail and more on the scene's implications for social norms and safety. They consistently perceived the clip as film content rather than persuasive advertising and expressed no intent to act. | Peace Evangelista, Kaila Smith |
| Lower-income care/support and service workers | Ages mid-30s to early 50s; occupations in caregiving, retail, cosmetology; lower income; ethnically diverse | Expressed strong emotional aversion and distrust of the clip's tone, describing it as 'cold' or 'off-putting.' Their reactions suggest that framing violent, matter-of-fact imagery as persuasive material would likely backfire with this cohort. | Brent Guevara, Brianna Chapman, Maribel Miller, Precious Rai |
| Higher-income / professional respondents | Higher-earning or professional roles (engineer, other skilled professions); varied ages, often older | These viewers found the scene believable and could articulate the on-screen intent plainly ('plain as day', 'blunt'), yet that clarity did not translate to willingness to take action-suggesting clarity of message alone cannot overcome negative affect elicited by tone. | Daniel Sassaman, John Grimm, Mario Bockus |
Shared Mindsets
| Trait | Signal | Agents |
|---|---|---|
| Disturbance anchored in tone rather than explicit gore | Nearly all respondents pointed to the casual, matter-of-fact manner in which violence was presented as the key unsettling element-more impactful than graphic detail-and that tone drove rejection of any implied persuasion. | Sandra Falcinelli, Peace Evangelista, Sam Norstrom, Kaila Smith, Precious Rai, Daniel Sassaman, Brent Guevara, Gregory Cumbo, Brianna Chapman, Maribel Miller, John Grimm |
| Perception of the clip as non-advertorial | Most viewers categorized the excerpt as a film clip rather than an advertisement and therefore did not feel motivated by it; the lack of a clear ask or contextual framing drove low action intent across segments. | Peace Evangelista, Kaila Smith, Sandra Falcinelli, Precious Rai, Gregory Cumbo, Mario Bockus, Brianna Chapman |
| Low likelihood to take advertised action | Across income, age, and occupation, respondents reported they would not comply with or be motivated by an action prompt associated with this footage-negative affect and perceived inappropriateness outweighed any persuasive clarity. | Sam Norstrom, Peace Evangelista, Kaila Smith, Sandra Falcinelli, Precious Rai, Gregory Cumbo, Daniel Sassaman, Brent Guevara, Maribel Miller, John Grimm, Brianna Chapman, Mario Bockus |
| Consistent comprehension of on-screen intent | Even when viewers rejected the clip as persuasive, many could clearly describe the depicted activity-armed men hunting/disposing of perceived threats-indicating cognitive comprehension is distinct from emotional acceptability. | Sam Norstrom, Daniel Sassaman, Brianna Chapman, Maribel Miller, John Grimm |
Divergences
| Segment | Contrast | Agents |
|---|---|---|
| Mario Bockus (outlier) | Unlike the majority who experienced strong visceral discomfort, Mario described the scene as 'not all that intense or graphic' and read it more neutrally as a movie clip rather than deeply unsettling. | Mario Bockus |
| Older rural vs. Younger urban viewers | Older rural respondents emphasized gut-level physical unease and procedural interpretation (who is doing what), while younger urban respondents framed responses in social/civic terms (breakdown of order). Both groups disliked the clip but the lens through which they interpreted the discomfort differs-practical/visceral versus societal/ethical. | Sandra Falcinelli, Sam Norstrom, Daniel Sassaman, Peace Evangelista, Kaila Smith |
| Lower-income caregivers vs. higher-income professionals | Care/support workers focused on distrust and moral distaste tied to the clip's tone; higher-income/professionals noted clarity and believability but still rejected action-showing that perceived message clarity can coexist with moral aversion that suppresses behavioral intent. | Brent Guevara, Brianna Chapman, Maribel Miller, Precious Rai, Daniel Sassaman, John Grimm |
Overview
Quick Wins (next 2–4 weeks)
| # | Action | Why | Owner | Effort | Impact |
|---|---|---|---|---|---|
| 1 | Add explicit context and CTA end-card to stimuli | Most viewers classified the clip as film, not an ad; absence of a clear ask confounded intent measurement. | Research Ops Lead | Low | High |
| 2 | Insert two survey items: CTA recognition + Tone Discomfort | Separates comprehension from acceptability and captures the specific driver: casualized-violence tone. | Insights Lead | Low | High |
| 3 | Tag content with 'Casualized Violence' metadata flag | Enables brand-safety guidance and placement avoidance recommendations quickly. | Client Strategy Manager | Low | Med |
| 4 | Segment views in dashboards (Older/Rural vs Younger/Urban) | Highlights lens differences while showing shared rejection of action-useful for client narratives. | Data Analyst | Low | Med |
| 5 | Update moderator guide: probe 'tone vs gore' and 'order/collapse' | Targets the observed driver and aligns qual with quant drivers immediately. | Qual Research Lead | Low | Med |
| 6 | Add content warning and opt-out pathway | Reduces panel attrition and ethical risk when testing high-arousal clips. | Compliance Manager | Low | Med |
Initiatives (30–90 days)
| # | Initiative | Description | Owner | Timeline | Dependencies |
|---|---|---|---|---|---|
| 1 | Violence Tone Calibration Framework | Create a taxonomy and 1–7 scales for casualness of violence, graphicness, authority/impunity cues, and moral distance. Train raters, codify thresholds for brand-safety tiers, and publish placement guidance for clients. | Head of Insights | 4–6 weeks | Archive of diverse clips, Rater training set, Coding guidelines, QA review loop |
| 2 | High-Arousal Creative Testing Protocol | Standardize stimuli requirements: mandatory CTA end-card, context slide ('You are viewing an ad concept'), A/B tone-variant edits, and a readiness checklist. Include forced-choice CTA recall and Tone Discomfort Index in every run. | Product (Testing) Manager | 6–8 weeks | Creative variant production, Survey instrument updates, Client buy-in, Panel scripting |
| 3 | Contextual Adjacency Experiment | Test identical creatives adjacent to different program genres/dayparts to model how context moderates action intent and discomfort. Deliver a placement optimizer for buyers/publishers. | Data Science Lead | 8 weeks | Simulated or partnered inventory contexts, Sample expansion, Modeling framework, Reporting templates |
| 4 | Panel Expansion + Psychometrics | Increase N and diversity; add scales (moral foundations, authority trust, sensation seeking). Establish normative baselines and test-retest reliability for tone-sensitive content. | Panel Operations Manager | 6 weeks | Recruitment budget, Screeners/consent updates, Incentive plan, IRB/ethics review |
| 5 | Segment Playbooks: Messaging Around Safety and Order | Produce concise client playbooks indicating
|
Client Strategy Director | 4 weeks | Calibrated taxonomy outputs, Creative examples, Design/resources |
KPIs to Track
| # | KPI | Definition | Target | Frequency |
|---|---|---|---|---|
| 1 | CTA Comprehension Rate | % of respondents who correctly state the CTA unaided/forced-choice after exposure. | ≥ 80% per concept test | Per study |
| 2 | Action Intent Lift vs. Neutral Baseline | Difference in stated likelihood to act vs. a matched non-violent neutral creative baseline. | ≥ +10 points (5-pt or 7-pt scale equivalent) | Per study |
| 3 | Tone Discomfort Index (TDI) | Mean 1–7 rating on discomfort specifically attributed to matter-of-fact violence. | ≤ 3.5 for broad-reach ads (or segment-targeted thresholds) | Per study and monthly rollup |
| 4 | Ad-vs-Film Misclassification Rate | % of viewers who categorize the stimulus as non-ad content. | ≤ 10% | Per study |
| 5 | Brand-Safety Flag Rate | % of stimuli triggering red/amber flags under the new calibration framework. | Downward trend post-protocol rollout | Monthly |
| 6 | Segment Divergence Spread | Variance in TDI across key segments (older/rural vs younger/urban, income cohorts). | Within predefined bands or accompanied by tailored playbook guidance | Quarterly |
Risks & Mitigations
| # | Risk | Mitigation | Owner |
|---|---|---|---|
| 1 | Overgeneralizing from a single public-domain clip to broad media trends. | Expand stimuli set across genres and tones; use the clip only as a calibration anchor. | Head of Insights |
| 2 | Panel fatigue or attrition due to repeated exposure to violent content. | Pre-screening, content warnings, cooldown periods, and enhanced incentives. | Panel Operations Manager |
| 3 | Ethical and reputational concerns testing high-arousal material. | Ethics review, opt-outs, anonymized reporting, and strict stimulus suitability criteria. | Compliance Manager |
| 4 | Client resistance to constraints on violent tone in creatives. | Demonstrate ROI with A/B tests, quantify risk via TDI and misclassification metrics, provide alternative creative routes. | Client Strategy Director |
| 5 | Measurement confounds when creatives lack CTA or context. | Enforce a creative readiness checklist and reject non-compliant stimuli for persuasion testing. | Product (Testing) Manager |
Timeline
Weeks 2–6: Build and validate Violence Tone Calibration Framework; begin panel expansion.
Weeks 4–10: Roll out High-Arousal Testing Protocol; enforce readiness checklist; initial client pilots.
Weeks 6–12: Run Contextual Adjacency Experiment; model placement effects; draft optimizer.
Weeks 10–12: Publish segment playbooks and initial benchmarks; socialize with clients and programming partners.
Objective and Context
A+E Global screened a public-domain excerpt from Night of the Living Dead (1968) to the standardized 12-agent calibration panel to capture immediate viewer response to the clip’s visual and audio material. Findings from this calibration run align with a broader set of 36 viewers: the footage’s matter-of-fact treatment of violence consistently generated negative affect, clear comprehension of on-screen intent, and rejection of any implied persuasion or call-to-action (CTA).
What We Learned (Audience Response Patterns)
- Tone, not gore, drives disturbance. Across segments, viewers pointed to the casual, matter-of-fact presentation of violence as the unsettling core-described as “cold,” evoking a “pit in my stomach”-more than any graphic detail. This tone reliably depressed willingness to act.
- High comprehension, low persuasion. Most participants accurately identified the on-screen action as armed men hunting/disposing of perceived threats (“plain as day,” “blunt”), yet they were not motivated to act. This underscores that message clarity ≠ persuasion when tone elicits moral aversion.
- Perceived as film, not advertising. Viewers commonly categorized the excerpt as movie content. The lack of explicit context and CTA suppressed action intent and introduced measurement confounds.
- Outlier awareness. One respondent (Mario Bockus) read the scene as “not all that intense or graphic,” highlighting a minority, more neutral stance amid broad discomfort.
Persona Correlations and Demographic Nuance
- Older, rural participants (55+): Reported the most visceral reactions (“gut-level unease”) and could articulate procedural intent most clearly (men hunting/disposing of threats). Despite clarity, they rejected any advertised action.
- Younger, urban/suburban women (late 20s–early 30s): Equally uncomfortable but framed responses through social/civic lenses (e.g., “collapse of order,” “cold”), focusing less on procedural detail and more on implications for norms and safety; no action intent.
- Lower-income care/support and service workers: Expressed strong aversion and distrust tied to the tone (“off‑putting,” “cold”). Attempts to use such matter-of-fact violent imagery persuasively would likely backfire with this cohort.
- Higher-income/professional respondents: Emphasized believability and clarity (“plain as day”) yet still declined to act-clarity could not overcome moral aversion.
Implications and Recommendations
For US media and advertising contexts, depictions of casualized violence reliably depress stated action intent across segments. Interpretive lenses differ (visceral/practical vs. societal/ethical), but avoidance is shared.
- Quick wins
- Add an explicit context slide and CTA end-card to stimuli to reduce ad-vs-film misclassification.
- Insert two items into surveys: CTA recognition and a Tone Discomfort attribution measure to separate comprehension from acceptability.
- Tag clips with a “Casualized Violence” metadata flag for brand-safety and placement guidance.
- Segment dashboards by Older/Rural vs. Younger/Urban to reflect lens differences alongside common avoidance.
- Update moderator guides to probe “tone vs. gore” and “order/collapse” explicitly.
- Initiatives
- Build a Violence Tone Calibration Framework (casualness, graphicness, authority/impunity cues, moral distance) with tiered brand-safety thresholds.
- Launch a High-Arousal Creative Testing Protocol requiring context slides, CTA end-cards, tone-variant edits, forced-choice CTA recall, and a Tone Discomfort Index.
- Run a Contextual Adjacency Experiment to quantify how program context moderates intent and discomfort.
- Expand the panel and add psychometrics (moral foundations, authority trust, sensation seeking) to establish normative baselines.
- Publish Segment Playbooks on messaging around safety/order: avoid casualized killing/impunity cues; emphasize agency and restoration of safety; tailor CTAs by segment.
Risks and Measurement Guardrails
- Risks: Overgeneralizing from a single clip; panel fatigue with violent content; ethical/reputational concerns; client resistance to tone constraints; measurement confounds without CTA/context.
- Mitigations: Broaden stimuli set; content warnings/cooldowns; ethics review and opt-outs; A/B demonstrations of ROI; enforce creative readiness checklists.
- KPIs:
- CTA Comprehension Rate: ≥ 80%.
- Action Intent Lift vs. neutral baseline: ≥ +10 points.
- Tone Discomfort Index (matter-of-fact violence): ≤ 3.5 for broad-reach ads.
- Ad‑vs‑Film Misclassification: ≤ 10%.
- Brand‑Safety Flag Rate: downward trend post‑rollout.
Next Steps
- Weeks 0–2: Deploy context/CTA cards; add survey items; apply metadata flags; update moderator guide.
- Weeks 2–6: Build and validate the Violence Tone Calibration Framework; begin panel expansion.
- Weeks 4–10: Roll out the High-Arousal Testing Protocol and enforce the readiness checklist with pilot clients.
- Weeks 6–12: Conduct the Contextual Adjacency Experiment; develop a placement optimizer.
- Weeks 10–12: Publish segment playbooks and initial benchmarks; brief clients and programming partners.
Decision point: approve protocol updates and taxonomy build this sprint; confirm KPI targets and readiness criteria for all upcoming tone-sensitive tests.
-
What kind of content did you think this clip was?single select Clarifies perceived content type to guide framing and disclosure tactics for future cuts.
-
What, if any, specific action were you being asked to take?open text Diagnoses CTA recognition to inform end-card/overlay design and copy.
-
How uncomfortable did the tone of this clip make you feel?semantic differential Measures tone discomfort to adjust tone or add context without changing core visuals.
-
Which component most influenced your reaction to the clip?single select Guides whether to prioritize audio or visual edits for maximum impact.
-
How appropriate would this clip be for use in a general‑audience advertisement?likert Evaluates brand safety and placement suitability across channels.
-
What, if anything, kept you from wanting to take any action after viewing?multi select Identifies fixable barriers (clarity, relevance, ethics, tone) to improve conversion potential.
Research group: 12 calibrated U.S. viewers (ages 29–62), majority rural with mixed hands-on/technical roles, plus younger urban/suburban women and both lower-income care/service and higher-income professional cohorts.
What they said: Viewers reported strong disturbance driven by the clip’s casual, matter-of-fact violence (e.g., “good shot”), describing visceral unease or a collapse of order.
Most accurately identified the on-screen intent (armed men hunting/disposing of threats) yet rejected it as advertising due to no clear CTA and reported very low likelihood to take any action; one outlier found it less intense. Main insights: Tone-not gore-drove negative affect; clarity of what was happening did not translate into persuasion; segments differed in interpretive lens (older rural visceral/procedural vs younger urban societal/ethical) but converged in refusing action.
Takeaways:
- Add explicit context and a CTA end-card to stimuli; track CTA recognition and a Tone Discomfort Index to separate comprehension from acceptability.
- Avoid “casualized violence” when seeking persuasion; if used, enforce brand-safety flags and restrict to context-appropriate placements.
- Implement content warnings/opt-outs and segment reporting to manage ethics and show shared rejection despite lens differences.
| Participant | Response | Actions |
|---|