In cardiac auscultation, physicians face the challenging task of distinguishing between normal heart sounds and pathological murmurs. Innovative technologies including AI offer potential clinical decision support (CDS). This study aims to investigate how different CDS-systems for digital auscultation affect physicians’ decision-making under different conditions. We aim to understand differences in terms of diagnostic accuracy, diagnostic confidence and diagnostic calibration between systems that provide pure information analysis (IA) communicated via data visualization and systems that provide additional decision recommendations (DR). Furthermore, we want to investigate the influence of data quality on subjective perception (e.g., trust, perceived usefulness) and reliance on digital systems. To answer these questions, we use a 2 between (degree of automation: IA vs. DR CDS) x 2 within (data quality: noisy audio vs. normal audio recordings) x 2 within (T1=audio only vs. T2=audio with CDS) mixed design with a target sample size of N=36 physicians. Among other things, we expect the DR group to improve more in performance and diagnostic calibration, while the IA group is expected to improve more in diagnostic confidence. We hypothesize that noisy audio data will negatively affect the performance of participants in the audio only condition. In the DR condition, we expect that noisy audio data will negatively affect confidence and positively affect performance. We expect that both systems will be perceived as more useful when the audio is noisy, but hypothesize that the DR system will be perceived as even more useful. We will present results from a currently ongoing preregistered study.