A persistent accuracy ceiling in automated verbal deception detection
Automated methods have been proposed to overcome the limitations of human verbal deception detection, but evidence remains fragmented across disciplines.
ProofPaper ↗
Key points
- We systematically reviewed 25 years of research (289 reports, 6,136 classification models) and meta-analyzed 3,653 models nested within 97 datasets.
- Pooled accuracy was 74.4% (95% CI: 71.2%-77.4%) with substantial heterogeneity.
- Accuracy was driven by methodological quality (ground truth, data source, class balance, evaluation procedure) more than by model complexity: the adoption of embeddings and large language models has not translated into improved predictive performance.
- The pooled accuracy aligns with meta-analyses of manual approaches, suggesting a ceiling of 70-75%, unlikely to be lifted by current research conventions.
Sources (1)
- [1]A persistent accuracy ceiling in automated verbal deception detectionarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 03:15 PM
Automated methods have been proposed to overcome the limitations of human verbal deception detection, but evidence remains fragmented across disciplines.
We systematically reviewed 25 years of research (289 reports, 6,136 classification models) and meta-analyzed 3,653 models nested within 97 datasets.
Extractive summary: sentences quoted from the sources.