Self-Monitoring Accuracy over Time for a Complex Diagnostic Skill.
Description
Roman Hari and Juliane E. Kämmer contributed equally to this work.
Publisher DOI
PubMed ID
42145916
Abstract
Background
Moment-to-moment self-monitoring correlates well with task performance. However, little is known about the development of self-monitoring accuracy over time. This retention study explores the long-term development of self-monitoring in the performance of a complex clinical task.Methods
Advanced medical students, without prior ultrasound skills, participated in an ultrasound course and were evaluated using OSCEs immediately after the course (T1) and after a six-month follow-up (T2). Assessment included expert evaluation of skills and self-reported confidence levels at each station. Two linear mixed models were used to track changes of performance and confidence over time, and assess the effect of additional training, demographics, and performance on confidence levels.Results
The study included 141 medical students (65% female, median age 22 years). Performance scores across six OSCE stations were significantly higher at T1 compared to T2, with median scores of 36.0 and 34.3 points (on a scale of 0-50), while confidence levels decreased from a median of 3 to 2.8 (on a scale of 1-5). On average, self-reported confidence decreased in some proportion to the skill decline over the follow-up period of six months, although individual students displayed various patterns of performance and confidence changes between T1 and T2. Male students tended to report higher confidence levels, while training-related factors positively influenced confidence and performance.Conclusion
A small but growing body of research suggests that self-monitoring accuracy is not a stable trait but changes over time. Our results suggest an improved self-monitoring accuracy at the delayed retest. Future studies should assess self-monitoring in a more fine-grained way and over a longer period of time, and systematically assess inter-individual differences.
Moment-to-moment self-monitoring correlates well with task performance. However, little is known about the development of self-monitoring accuracy over time. This retention study explores the long-term development of self-monitoring in the performance of a complex clinical task.Methods
Advanced medical students, without prior ultrasound skills, participated in an ultrasound course and were evaluated using OSCEs immediately after the course (T1) and after a six-month follow-up (T2). Assessment included expert evaluation of skills and self-reported confidence levels at each station. Two linear mixed models were used to track changes of performance and confidence over time, and assess the effect of additional training, demographics, and performance on confidence levels.Results
The study included 141 medical students (65% female, median age 22 years). Performance scores across six OSCE stations were significantly higher at T1 compared to T2, with median scores of 36.0 and 34.3 points (on a scale of 0-50), while confidence levels decreased from a median of 3 to 2.8 (on a scale of 1-5). On average, self-reported confidence decreased in some proportion to the skill decline over the follow-up period of six months, although individual students displayed various patterns of performance and confidence changes between T1 and T2. Male students tended to report higher confidence levels, while training-related factors positively influenced confidence and performance.Conclusion
A small but growing body of research suggests that self-monitoring accuracy is not a stable trait but changes over time. Our results suggest an improved self-monitoring accuracy at the delayed retest. Future studies should assess self-monitoring in a more fine-grained way and over a longer period of time, and systematically assess inter-individual differences.
Date Issued
2026-05-14
Publication Type
Article
Subject(s)
Language(s)
en
Author(s)
Schauber, Stefan K | |
Journal
Perspectives on Medical Education
Publisher
Ubiquity Press
ISSN
2212-277X
2212-2761
Funding(s)
Access(Rights)
open.access