Subjects bias adjustment experiments by setting sounds louder for good audibility

Evidence

Inspecting the two data sets more carefully, it becomes clear that there is a systematic difference between the results. It seems that at least for sounds that are not extremely loud, i.e. have an SPL smaller than about 80 dB, the subjects have the tendency to set the variable sound louder than expected. This seems to be reasonable for faint and intermediate loud sounds. In this case, the subjects want to have the sound in a loudness range that corresponds to good audibility, i.e. fairly loud. — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 1193

Categorization

Category: Causal Relationships

Related

• Auditory assessment must account for the listener’s response bias, not just sensitivity (similarity 0.92)
• Adjustment methods bias moderate sounds toward louder settings for good audibility. (similarity 0.92)
• Response bias shifts audiometric thresholds away from a patient’s true hearing abilities (similarity 0.91)
• Active stimulus control by subjects can bias psychophysical measurements (similarity 0.83)
• Non-auditory factors strongly bias loudspeaker purchase preferences in sighted listening (similarity 0.83)
• Spectrally shaped targets can equalize audibility across listeners with differing audiograms. (similarity 0.82)
• Presentation method, room acoustics, and listener state are independent variables in listening experiments (similarity 0.82)
• Visual and contextual cues bias human judgments of sound quality (similarity 0.81)
• Adjustment steps sized near the JND prevent subjects overestimating imperceptible changes (similarity 0.81)
• Uncontrolled loudness differences confound audio quality listening-test results (similarity 0.81)
• Randomizing identifiers, positions, and music prevents experimenter bias in listening tests (similarity 0.81)
• Louder playback is unconsciously rated as better quality in audio listening tests (similarity 0.80)
• Standard loudness analysis discards response variability that probabilistic models can exploit (similarity 0.80)
• Testing hearing-impaired listeners could reveal how loudness distortion maps to brain activity (similarity 0.80)
• Roving sound level prevents listeners from using absolute intensity as a cue (similarity 0.80)

Broader Connections

• Controlled experiments trace sound-feature variables against listeners’ behavioral responses #expands (0.90)
• Experimental method determines what discrepancies listeners can perceive #expands (0.85)
• The method of adjustment lacks a fixed response criterion across listeners #qualifies (0.85)
• Visual and contextual cues bias human judgments of sound quality #expands (0.78)