envirosound: Auditory Unpleasantness of Environmental Sound

In eba: Elimination-by-Aspects Models

Description

Zimmer et al. (2004) investigated the auditory unpleasantness of twelve short binaural recordings (Johannsen and Prante, 2001); recordings were presented via headphones to 74 participants.

Usage

data(envirosound)

Format

A data frame containing 74 observations on 2 variables:

unpleasantness

paired comparison of class paircomp; judgments for all 66 paired comparisons from 12 recordings: circular saw, stadium, dentist's drill, waterfall, ship's horn, stone in well, typewriter, hooves, fan, howling wind, tyre on gravel, wasp.

rt

median response time.

Details

Details of the recordings, including psychoacoustic metrics, are available as an attribute of the unpleasantness variable (see Examples).

Source

Zimmer, K., Ellermeier, W., & Schmid, C. (2004). Using probabilistic choice models to investigate auditory unpleasantness. Acta Acustica united with Acustica, 90(6), 1019–1028.

References

Johannsen, K., & Prante, H.U. (2001). Environmental sounds for psychoacoustic testing. Acta Acustica united with Acustica, 87(2), 290–293.

See Also

eba, strans, paircomp.

Examples

requireNamespace("psychotools")
data(envirosound)
set.seed(1019)

## Choice-model representation of unpleasantness
mat <- summary(envirosound$unpleasantness, pcmatrix = TRUE)
strans(mat)
btl1 <- eba(mat)
eba1 <- eba(mat, A = list(c(1, 13), c(2, 13), c(3, 13), c(4, 13),
                          c(5, 13), c(6, 13), c(7, 13), c(8, 13),
                          c(9, 13), c(10, 13), c(11, 13), 12))
eba2 <- eba(mat, A = list(c(1, 13), c(2, 13), c(3, 13), c(4, 13),
                          c(5, 13), c(6, 13), c(7, 13, 14), c(8, 13, 14),
                          c(9, 13, 14), c(10, 13, 14), c(11, 13, 14), 12),
            s = runif(14))
anova(btl1, eba1, eba2)

sounds <- psychotools::covariates(envirosound$unpleasantness)
sounds$u <- 10 * uscale(eba2, norm = 9)  # u(fan) := 10

plot(magnitude ~ u, sounds, log = "x", type = "n",
     xlab = "Indirect scaling (EBA model)",
     ylab = "Direct magnitude estimation",
     main = "Auditory unpleasantness of environmental sound")
mtext("(Zimmer et al., 2004)", line = 0.5)
abline(lm(magnitude ~ log10(u), sounds))
text(magnitude ~ u, sounds, labels = abbreviate(rownames(sounds), 4))

## Predicting unpleasantness from psychoacoustic metrics
summary(
  lm(log(u) ~ scale(sharpness, scale = FALSE) +
              scale(roughness, scale = FALSE):I(loudness.5 > 27),
     sounds[-12, ])  # w/o wasp
)

eba documentation built on Jan. 13, 2021, 10:12 a.m.