Response bias distorts threshold measurements in hearing experiments.

Evidence

Under the neutral criterion, a listener will have the highest number of correct detections. Note, however, that this listener will have some false positives due to the overlapping distributions. A listener adopting a conservative criterion will have fewer false positives (and possibly none) but also fewer true positives. The listener with a liberal criterion will have many false positives as well as a high proportion of true positives. In this way, both liberal and conservative criteria will lead to a lower percent correct, and therefore both types of criteria can cause an experimenter to misestimate the threshold. — Jennifer Lentz, Psychoacoustics- Perception of Normal…, 338

On the other hand, an extremely liberal responder produces the opposite result: many false positives. Both types of patients can lead to inaccurate audiograms unless audiologists are able to encourage their patients to adopt a neutral criterion. Patients with conservative criteria can produce audiograms with hearing thresholds that are higher (worse) than their true hearing abilities, and the reverse can be true for those adopting liberal criteria. — Jennifer Lentz, Psychoacoustics- Perception of Normal…, 344

We also refer to response proclivity as response bias, more explicitly defined as the listener’s tendency to say, “Yes I heard something.” Listeners may come to the laboratory or the clinic with different motivations for participation, and these factors must be considered in any psychoacoustic or audiological assessment. — Jennifer Lentz, Psychoacoustics- Perception of Normal…, 328

Categorization

• QUESTION: Methods like two-interval forced choice and yes-no procedures differ in whether criterion shift can bias threshold estimates — worth mapping.
• XREF: Connects to why clinical hearing tests use forced-choice procedures to minimize bias errors.
• DEFINE: Defines response bias as the listener’s tendency to say ‘yes I heard something,’ distinct from sensitivity.
• corroborated: 4 independent passages support this claim — Jennifer Lentz, Psychoacoustics- Perception of Normal…, 338; Jennifer Lentz, Psychoacoustics- Perception of Normal…, 344; Jennifer Lentz, Psychoacoustics- Perception of Normal…, 328