Signal Detection Theory’s Learning Problem
Author: Calvin Nguyen | Psychological Sciences
Abstract:
Signal Detection Theory (SDT) has proven to be a successful explanation for several observed regularities in the experimental data. The basic framework of signal detection theory is that people have the cognitive ability to detect a signal from background noise. However, in most classical SDT experiments, researchers analyze data that have been averaged across trials. The problem with this approach is that it ignores the process of learning that must occur for an organism to detect a signal over time inductively through experience, so that it can be positively identified from the background noise. A better approach would be to analyze performance on a trial-by-trial basis and determine whether the ability to identify a signal improves over time with experience. Unfortunately, we had difficulty locating publicly available trial-level datasets that could be analyzed in such a manner.
To address this limitation of the existing literature, we designed two experiments (N = 100 and 108) in which participants engaged in a binary judgment task on a computer and were asked to indicate the longer of two lines. In this situation, the signal is the longer line, while the noise is the shorter line, making this an ideally simplified design. On every trial, there is a line on the left of the screen and another on the right. In one between-subjects treatment, the longer line is on the left with a probability of 0.8, and in the other treatment, the longer line is on the right with a probability of 0.8. SDT clearly predicts that participants will learn the distribution and incorporate this information to improve their judgments over time, as they realize that the longer of the pair of lines more often appears on a certain side of the screen. Specifically, SDT predicts that participants in the longer-left treatment will have a bias toward lines on the left, and participants in the longer-right treatment will have a bias toward the right. However, we do not find evidence of this in our learning-by-experience experiment. To our surprise, the literature contains results analogous to ours (e.g., Parducci & Sandusky, 1965).
In our second learning-by-description experiment, we explicitly told participants which side, on average, the longer line appears, as well as provided feedback on each trial as to which side contained the longer line. In this experiment, we find that aggregate behavior is consistent with SDT. However, the effect increases across trials, despite no new information being learned. We hope that our results prompt renewed attention toward these apparently still unresolved problems with SDT and that researchers consider the dynamic effects in their data.
Graduate Poster Exhibition at SPARK!
The Graduate Poster Exhibition celebrates the research and creative work of the graduate community, showcasing everything from prose and code to original research and artistic expression. As part of SPARK! (Showcase of Projects, Art, Research, and Knowledge), a reimagining of Research Week, the exhibition highlights the depth, range, and impact of graduate scholarship and invites the campus community to engage with ideas taking shape across disciplines.
Understanding the Mind: The Psychological Sciences Program
The Psychological Sciences program at Rutgers–Camden offers a comprehensive curriculum that prepares students for advanced research and practice in various psychological disciplines. The program provides a foundation in research design and data analysis, with opportunities for individually supervised research apprenticeships. Students can choose between a thesis track, ideal for those aiming for doctoral studies, and a content track, suited for careers in non-academic research settings. The curriculum emphasizes scientific research methods applicable to real-world problems, equipping graduates with critical thinking and analytical skills essential for diverse career paths in psychology.
Spotlights & Stories
Want more stories like this? Spotlights and Stories highlights alumni, faculty, staff, and student experiences through stories, video, and voice. Learn more
Explore Graduate School Programs
Rutgers Graduate School-Camden offers 20+ graduate certificates, master’s, and doctoral programs across various disciplines, including biology, data science, creative writing, and psychology. We take pride in our academic diversity. Learn more
Former Students, Get Involved
Stay connected with Rutgers-Camden! Engage with former students, attend events, and support current students. Your involvement strengthens our community and helps shape future success. Learn more