# Clever Hans Effect Found in Automatic Detection of Alzheimer's Disease through Speech

arXiv: https://arxiv.org/abs/2406.07410 · HTML: https://mlsanity.com/paper/2406.07410/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Jun 2024.

## Built on by (most cited first, 9 of 9)

- [Shortcut Learning in Binary Classifier Black Boxes: Applications to Voice Anti-Spoofing and Biometrics](https://mlsanity.com/paper/2601.17782/) (2026)
- [SpurAudio: A Benchmark for Studying Shortcut Learning in Few-Shot Audio Classification](https://mlsanity.com/paper/2605.13672/) (2026)
- [Detecting Neurocognitive Disorders through Analyses of Topic Evolution and Cross-modal Consistency in Visual-Stimulated Narratives](https://mlsanity.com/paper/2501.03727/) (2025)
- [Unveil Multi-Picture Descriptions for Multilingual Mild Cognitive Impairment Detection via Contrastive Learning](https://mlsanity.com/paper/2505.17067/) (2025)
- [Agentic Cognitive Profiling: Realigning Automated Alzheimer's Disease Detection with Clinical Construct Validity](https://mlsanity.com/paper/2603.17392/) (2026)
- [A Toolkit for Detecting Spurious Correlations in Speech Datasets](https://mlsanity.com/paper/2604.26676/) (2026)
- [Towards Quantifying Benchmark Optimization in ASR Models](https://mlsanity.com/paper/2608.19936/) (2026)
- [Behind the [MASK]: Disentangling Representation and Faithfulness in DAPF-Based Dementia Detection](https://mlsanity.com/paper/2608.25028/) (2026)
- [Beyond Decodability: Do Acoustic Factors Drive Predictions in Speech-Based Alzheimer's Assessment?](https://mlsanity.com/paper/2610.01846/) (2026)

## Abstract (arXiv)

We uncover an underlying bias present in the audio recordings produced from the picture description task of the Pitt corpus, the largest publicly accessible database for Alzheimer's Disease (AD) detection research. Even by solely utilizing the silent segments of these audio recordings, we achieve nearly 100% accuracy in AD detection. However, employing the same methods to other datasets and preprocessed Pitt recordings results in typical levels (approximately 80%) of AD detection accuracy. These results demonstrate a Clever Hans effect in AD detection on the Pitt corpus. Our findings emphasize the crucial importance of maintaining vigilance regarding inherent biases in datasets utilized for training deep learning models, and highlight the necessity for a better understanding of the models' performance.