Sequential dependencies in magnitude scaling of loudness

Ten normally hearing listeners used a programmable sone-potentiometer knob to adjust the level of a 1000-Hz sinusoid to match the loudness of numbers presented to them in a magnitude production task. Three different power-law exponents (0.15, 0.30, and 0.60) and a log-law with equal steps in dB were...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of the Acoustical Society of America 2013-05, Vol.133 (5_Supplement), p.3284-3284
Hauptverfasser: Joshi, Suyash N., Jesteadt, Walt
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Ten normally hearing listeners used a programmable sone-potentiometer knob to adjust the level of a 1000-Hz sinusoid to match the loudness of numbers presented to them in a magnitude production task. Three different power-law exponents (0.15, 0.30, and 0.60) and a log-law with equal steps in dB were used to program the sone-potentiometer. The knob settings systematically influenced the form of the loudness function. Time series analysis was used to assess the sequential dependencies in the data, which increased with increasing exponent and were greatest for the log-law. It would be possible, therefore, to choose knob properties that minimized these dependencies. When the sequential dependencies were removed from the data, the slope of the loudness functions did not change, but the variability decreased. Sequential dependencies were only present when the level of the tone on the previous trial was higher than on the current trial. According to the attention band hypothesis [Green and Luce, Perception Psychophys., 1974] these dependencies arise from a process similar to selective attention, but observations of rapid adaptation of neurons in the inferior colliculus based on stimulus level statistics [Dean et al., Nature Neurosci. (2005)] would also account for the data. [Work supported by NIH.]
ISSN:0001-4966
1520-8524
DOI:10.1121/1.4805381