From Deficit to Dialogue on every level
Deficit vs. Dialogue in Practice
"Among science communication researchers and those concerned with the relationship between science and society, this model of communication is often considered essentially flawed, in that it affords politically problematic privileges to scientific knowledge and leaves scientists ignorant of the needs and knowledges their audiences, ignorant of the contexts in which decisions will be made, less likely to be viewed as trustworthy by those audiences, and less likely to do good science. The end result of such an approach is only likely to be a greater distrust between science and society, and flawed science.
Yet despite these critiques—and a near total absence of evidence in its favor—the model perseveres"
https://oxfordre.com/communication/display/10.1093/acrefore/9780190228613.001.0001/acrefore-9780190228613-e-1396
Reframing Common Statements
Moving from deficit to dialogue often requires just slight rewording—but the shift in perspective is profound. Here are practical transformations you can apply to your own communication.
❌ Deficit Model: "People don't understand that..."
✓ Dialogue Model: "Many people wonder whether... Here are different perspectives on this question..."
❌ Deficit Model: "If only the public knew X, they'd support Y"
✓ Dialogue Model: "Here's what research shows about X, AND here are the concerns communities have raised about Y..."
❌ Deficit Model: "I need to educate them about..."
✓ Dialogue Model: "Here's what we can explore together about..."
❌ Deficit Model: "They think X, but actually Y"
✓ Dialogue Model: "Many think X because [valid reason]. Scientists tend to see it as Y because [different valid reason]. Both views highlight important aspects..."
❌ Deficit Model: "The problem is public ignorance"
✓ Dialogue Model: "The challenge is that different communities value different aspects of this issue based on their experiences and priorities"

Key insight: The dialogue model doesn't abandon your expertise or avoid sharing knowledge. It positions that knowledge within a larger conversation that respects multiple forms of understanding and lived experience.
Step 2: Check Your Communication Model
Deficit Model Indicators

Warning signs you're using deficit model:
  • ❌ "Let me explain this to you"
  • ❌ "The public doesn't understand"
  • ❌ "If only they knew the facts"
  • ❌ One-way information flow
  • ❌ Assumes scientific knowledge is superior
The deficit model positions you as the expert dispensing knowledge to an ignorant audience. This approach often backfires, creating resistance rather than understanding.
Dialogue Model Indicators

Signs you're building dialogue:
  • ✓ "Here are multiple perspectives"
  • ✓ "What questions do you have?"
  • ✓ "Here's what we know and don't know"
  • ✓ Two-way exchange
  • ✓ Values different forms of knowledge
The dialogue model acknowledges that your audience brings valuable perspectives and experiences. It invites collaboration rather than demanding acceptance.
Reality Check for Monologue Formats
If you're creating a monologue format—an article, video, or presentation—you can still BUILD IN dialogue by deliberately incorporating these elements:
Acknowledge multiple perspectives
Present different viewpoints fairly, showing where reasonable people might disagree based on different values or priorities.
Explicitly state uncertainties
Be transparent about what you know, what you don't know, and why those boundaries matter.
Invite questions and responses
Create space for audience engagement, whether through comments, follow-up resources, or clear pathways for further discussion.
Show knowledge system intersections
Demonstrate where scientific knowledge meets local knowledge, lived experience, or practical wisdom.
Challenge: Balancing Precision and Accessibility
Structural Solution: Layered Structure
Provide multiple levels of depth and let readers choose their entry point. This respects different audience needs without compromising accuracy.
1
2
3
4
1
High Level
One-sentence accessible summary
2
Medium Level
Conceptual explanation with key mechanisms
3
Technical Level
Precise terminology and methodology
4
Expert Level
Full technical specifications and nuances
Example: Uncertainty Quantification in AI
HIGH LEVEL (1 sentence)
"We taught AI to express uncertainty—to say 'I'm not sure' when it doesn't know something."
MEDIUM LEVEL (Conceptual)
"By analyzing patterns in the model's internal representations—essentially watching which neural pathways activate—we can estimate confidence intervals for its predictions. The model learns to distinguish between answers based on strong training data versus extrapolations from limited information."
TECHNICAL LEVEL (Methods)
"Using vectorized backpropagation through probabilistic reasoning layers, we quantify both epistemic uncertainty (knowledge gaps) and aleatoric uncertainty (inherent randomness). The approach involves Bayesian neural networks with Monte Carlo dropout sampling across 1000 forward passes."
EXPERT LEVEL (Full specifications)
"Implementation details: We employ variational inference with reparameterization trick on weight posteriors, computing KL divergence between approximate and prior distributions. Calibration achieved through temperature scaling on validation set with ECE < 0.05 across benchmark datasets."
Provide all levels in your communication, but structure it so readers can stop at their comfort level. Each layer should be complete and accurate on its own terms.
Challenge: Finding Common Ground
Structural Solution: Feature-Style Opening
Start with SHARED EXPERIENCE, then connect to your research. This builds a bridge from the familiar to the technical, creating entry points for diverse audiences.
Shared Experience
Begin with something your audience has encountered in their daily lives
Raise the Stakes
Show how that familiar experience connects to larger consequences
Connect to Your Work
Introduce your research as addressing the stakes you've identified
Example Structure
"Think about the last time autocorrect changed 'I'm' to 'I'm' or turned 'ducking' into something else entirely. Frustrating, right? Maybe you laughed it off or had to send an embarrassing correction. [SHARED EXPERIENCE - everyone has autocorrect stories]
Now imagine that same technology, but making decisions about loan applications, medical diagnoses, or job candidates. Suddenly those errors aren't funny—they have real consequences for people's lives. [CONNECT TO STAKES]
That's why my research focuses on uncertainty quantification in AI systems—teaching these models to recognize when they're making educated guesses versus when they actually know something. [YOUR WORK - now clearly motivated]"
This approach works because it starts where your audience is, validates their experience, and shows them why they should care about your work. The research becomes the answer to a question they now have.
Final Checklist for Your Piece
Before you finalize your communication piece, conduct this comprehensive review to ensure your narrative choices serve understanding rather than convenience.
1
Structure Check
  • Have you chosen appropriate complexity for your target audience?
  • Does your structure serve understanding or narrative convenience?
2
Communication Model Check
  • Are you inviting dialogue or just delivering information?
  • Do you acknowledge multiple perspectives and knowledge systems?
  • Are you transparent about uncertainties and limitations?
  • Do you position yourself as colleague or expert-above?
  • Have you created space for audience response?
3
Honesty Check
  • Are you oversimplifying for narrative convenience?
  • Would your research colleagues recognize this story?
  • Are you hiding ongoing challenges or limitations?
  • Have you used past tense only for truly completed work?
4
Audience Check
  • Have you started with shared experience or common ground?
  • Is your technical depth appropriate for target audience?
  • Can non-experts find an entry point?
  • Does your structure help or hinder understanding?
  • Have you provided multiple levels of explanation?

Remember: A "yes" to these questions doesn't mean your piece is perfect—it means you've done the necessary self-examination to ensure your communication choices serve your audience and represent your research honestly.
Examples From Research Literature
Example 1: CRISPR Discovery (Well-Told Science Story)
1
Opening
Specific scene of Jennifer Doudna's nightmare about meeting Hitler—immediately stakes the ethical territory
2
Build
Decades of bacterial research across multiple labs worldwide—honors collaborative nature of science
3
Breakthrough
Realizing CRISPR's potential extends far beyond bacterial immunity systems
4
Complications
Ethical concerns, patent disputes, technical limitations emerge immediately
5
Current State
Revolutionary tool WITH significant unresolved questions and ongoing challenges
What Works in This Structure
  • Acknowledges multiple contributors rather than creating a "lone genius" narrative
  • Shows iterative discovery process with dead ends and collaborative breakthroughs
  • Balances excitement with genuine concerns without dampening the significance
  • Ends with open questions rather than false resolution or overpromising
  • Maintains scientific accuracy while remaining accessible to general readers
The article demonstrates how honesty about complexity and limitation actually enhances rather than diminishes the power of the story.
Additional Resources
Books
  • Vonnegut, K. (2005). A Man Without a Country - Original source for story shapes theory
  • Propp, V. (1968). Morphology of the Folk Tale - Mathematical structures in narratives
  • Olson, R. (2015). Houston, We Have a Narrative - Science communication through story
Academic Papers
  • Reincke, C.M., Bredenoord, A.L., & van Mil, M.H.W. (2020). From deficit to dialogue in science communication. EMBO Reports, 21(9), e51278.
  • Reagan, A.J., et al. (2016). The emotional arcs of stories are dominated by six basic shapes. EPJ Data Science, 5(1), 31.
  • Nisbet, M.C. & Scheufele, D.A. (2009). What's next for science communication? American Journal of Botany, 96(10), 1767-1778.
Videos
  • Kurt Vonnegut Lecture on Story Shapes (YouTube)
Remember
The goal isn't to:
  • ❌ Manipulate your audience with story tricks
  • ❌ Oversimplify for narrative convenience
  • ❌ Hide uncertainty behind confident narratives
  • ❌ Position yourself as expert-above
  • ❌ Create false resolutions or premature closure
The goal IS to:
  • ✓ Use structure to enhance understanding
  • ✓ Match your narrative shape to reality
  • ✓ Build trust through honest complexity
  • ✓ Invite dialogue, not just deliver information
  • ✓ Respect your audience's intelligence