Audience-Specific Health Communication: Mixed Methods Evaluation of the Maria Ciência AI-Assisted Knowledge Translation Tool
Bibliographic record
Abstract
Background: Scientific misinformation remains a major barrier to effective health communication. Bridging the gap between academic research and public understanding requires tools that simplify scientific language and adapt content to diverse audiences. Objective: This study presents Maria Ciência (LPCT-IGM), a specialized GPT-based assistant for science communication. The tool supports researchers in translating peer-reviewed scientific findings through simple prompts into accessible, ethically appropriate materials tailored for children, the general public, health professionals, and policymakers. Methods: The tool was configured using prompt engineering techniques and guided by curated reference materials on inclusive and nonstigmatizing scientific language. Materials derived from 47 public health papers resulted in 188 outputs, which were assessed by 121 evaluators using 4 criteria: clarity, level of detail, language suitability, and content quality. In addition, outputs generated by Maria Ciência were compared with those produced by a base large language model and with human-written science communication materials. Readability and linguistic accessibility were assessed using multiple established metrics. Results: Worldwide, mean scores were high: clarity (4.90), language suitability (4.78), content quality (4.72), and level of detail (4.56), on a 5-point scale. Materials for children and the general public consistently achieved the highest ratings across all criteria. A targeted comparison with the base large language model demonstrated superior performance of Maria Ciência in contextual stability. Readability analyses indicated that Maria Ciência's outputs were significantly more accessible than human-written texts, while maintaining high legibility classifications. Conclusions: Maria Ciência demonstrates the potential of artificial intelligence-assisted tools to enhance knowledge translation and counter scientific misinformation by producing scalable, audience-specific content that balances accessibility and informational integrity.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.107 | 0.174 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.004 | 0.002 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".