How different mental models of AI-based writing assistants impact writers’ interactions with them
Bibliographic record
Abstract
AI-based writing assistants are being integrated into a range of products and platforms. While direct users—who use AI writing assistants for various writing tasks—often get to decide how they integrate these systems in their writing process, the final outcomes (e.g., the writing artifact) will be influenced by the users’ understanding of how these systems work, and by whether their generated output (e.g., suggestions) is accurate and matches their expectations. In this write-up, we examine the types of controls users believe they have—based on their understanding of the system—when using AI-based writing assistants (e.g., getting personalized suggestions) and whether they can effectively use these controls to minimize possible negative outcomes (e.g., poor writing quality). To do so, we examine users’ mental models of writing assistants and how these models affect users’ ability to intervene appropriately. We argue that more work is needed to examine the connection between different mental models and a user’s ability to control these systems to minimize potential negative outcomes. To this end, we discuss an illustrative case study design where participants are asked to use an AI-based writing assistant to write a cover letter. We discuss how the results from this study could help us understand the impact that different mental models have on user reliance on AI-based writing assistants.
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 | 0.017 | 0.121 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.000 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.014 | 0.007 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 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".