DETECTION AND DISCRIMINATION OF MONAURAL AND BINAURAL INCREMENTS AND DECREMENTS OF INTENSITY WITH A

1
DETECTION AND DISCRIMINATION OF MONAURAL AND
BINAURAL

INCREMENTS AND DECREMENTS OF INTENSITY WITH A
TEMPORAL FRINGE Mark A. Stellmack, Neal F.
Viemeister, Andrew J. Byrne - Department of
Psychology, University of Minnesota
Increment Detection - Probe is incremented in the
signal interval (or ear, in the interaural
conditions), and there is no change in the
non-signal interval/ear. Decrement Detection -
Probe is decremented in the signal
interval/ear. Discrimination - One interval/ear
has an incremented probe, the other has a
decremented probe. Decrement w/Contra - Same as
2-interval monaural Decrement Detection, except
with a fixed (N030 dB) noise presented to the
right ear in both intervals. Carriers were
identical in both ears. Forward-fringe duration -
0, 0.5, 1, 2.5, 5, 7.5, 10, 15, 20, 25, 30, 35,
40, and 45 ms. No ramps were applied to the onset
or offset of the 5-ms probe
2aPP12. Detection thresholds are larger for
brief monaural intensity increments in
long-duration stimuli than for decrements. This
experiment assessed the contribution of a
decrement in one ear to direction discrimination
of an interaural intensity difference (IID),
where a brief intensity increment was presented
at one ear and an equal-magnitude intensity
decrement at the other. In two-interval monaural
tasks, thresholds were measured for detection of
an increment, a decrement, and discrimination of
the two. In single-interval binaural tasks,
thresholds were measured for identifying ear of
presentation of an increment, a decrement, and
for identification of direction of IID achieved
by presenting an increment and decrement to
opposite ears. In all cases, thresholds were
measured for a 5-ms probe segment as a function
of its temporal position in a 50-ms stimulus
(broadband noise). The major results (1)
Thresholds for discriminating the ear of
presentation of a decrement were much larger than
those for simply detecting the decrement
monaurally (2) interaural thresholds were
slightly better when both increment and decrement
were present relative to increment alone. These
results indicate that the decrements contributed
to binaural discrimination although by themselves
the same decrements were detectable but, at best,
weakly lateralizable.
Fig. 3. Data replotted from Fig. 2. From left to
right Increment detection, decrement detection,
and discrimination data, with monaural and
interaural plotted together.
Fig. 1. A schematic illustration of the envelopes
of the stimuli (noise bursts) presented in the
single-interval interaural and two-interval
monaural conditions. Only one trial type is
illustrated. In the experiment, the ear/interval
containing the increment/decrement was chosen
randomly from trial to trial. Stimuli are not
drawn to scale.
Left panel of Fig. 3 Monaural and interaural
increment detection thresholds are essentially
equal across forward-fringe duration. Middle
panel of Fig. 3 Interaural decrement detection
thresholds (circles) are much larger than
monaural decrement detection thresholds
(squares). More importantly, monaural
decrement-with-contra thresholds (triangles) are
nearly equal to the pure monaural
decrement-detection thresholds (squares). This
indicates that decrements are equally detectable
whether or not there is a stimulus present in the
contralateral ear. Furthermore, comparison of the
interaural decrement-detection thresholds
(circles) and monaural decrement-with-contra
thresholds (triangles) suggests that there is a
range of interaural decrements that is detectable
but listeners cannot identify which ear carries
the decrement. Right panel of Fig. 3
Interaural discrimination thresholds are lower
than monaural discrimination thresholds for
longer forward-fringe durations. This indicates
that decrements contribute more to interaural
discrimination than to monaural discrimination
for longer forward-fringe durations.
probe segment (unless those coincided with ramps
at the onset or offset of the entire 50-ms
signal). Procedure Threshold differences of
intensity in dB 10 log(?I/I) between the probe
segments were measured adaptively. Each threshold
is based on the mean threshold measured in four
adaptive runs. Data have been replotted in units
of 10 log(?I I)/I. Before each block of
interaural trials each listener was allowed to
listen repeatedly to a diotic broadband noise
burst and was instructed to adjust the headphones
so that the stimulus produced an intracranial
image at the center of his or her head.
INTRODUCTION In a number of studies, temporal
resolution of the binaural system has been
assessed by measuring thresholds for detection
and discrimination of non-zero interaural
intensity differences (IIDs) of a probe segment
of noise embedded in an otherwise-diotic stimulus
(e.g. Zurek, 1980 Bernstein et al., 2001
Akeroyd and Bernstein, 2001 Stellmack et al.,
2005). In the cited studies, the non-zero IID was
produced by incrementing the intensity of the
stimulus (relative to the fringe) in one ear
while simultaneously decrementing the intensity
in the other ear. The increments and decrements
were equal in magnitude (in dB) relative to the
fringe intensity. Because monaural increments are
more detectable than monaural decrements when
generated in this way (e.g. Oxenham, 1997), it is
possible that these interaural tasks were
performed by identifying the ear to which the
increment was presented while the decrement was
undetectable. The present experiment assesses
the contribution of the decremented stimulus to
the interaural tasks. Monaural and interaural
thresholds are measured for the detection and
discrimination of increments and decrements in
the intensity of a brief probe segment of an
otherwise-diotic stimulus. 
RESULTS AND DISCUSSION Left panel of Fig. 2
Monaural discrimination thresholds appear to be
driven primarily by detection of increments,
particularly at short forward-fringe durations.
For longer forward-fringe durations, monaural
discrimination thresholds are slightly larger
than increment-detection thresholds. If
increments and decrements are detected as
unsigned changes (Macmillan, 1971),
discrimination will be more difficult when they
are nearly equally detectable, as they are for
longer forward fringe durations. Right panel of
Fig. 2 Thresholds for interaural decrement
detection are much larger than those for
interaural increment detection. However,
decrements that are well below their interaural
detection thresholds appear to contribute to
discrimination such that discrimination
thresholds are slightly lower than interaural
increment-detection thresholds.
CONCLUSIONS 1) Monaural and interaural
decrement-detection thresholds are larger than
monaural increment detection thresholds. The
difference is much larger for interaural
thresholds. 2) Thresholds for identifying the ear
in which a decrement occurs are much higher than
thresholds for simply detecting a decrement in
one ear with a fixed level stimulus in the other
ear. This indicates that there is a range of
intensity changes for which listeners can detect
a decrement in one ear but they cannot identify
the ear in which it occurred. 3) Decrements that
are well below the threshold for identifying the
ear of presentation and slightly below detection
threshold contribute to lower interaural
discrimination thresholds.
METHODS Subjects 4 adult listeners with normal
hearing. Stimulus (See Fig. 1.) Carrier -
50-ms broadband noise (1Hz-8kHz), fringe N030
dB, 1-ms cosine-squared onset/offset
ramps. Conditions Interaural (Single-Interval)
- Diotic noise, except for a 5-ms probe segment
with a non-zero IID. Listeners indicated
whether the IID of the probe on each trial
favored the left or right ear. The probe IID was
produced by incrementing or decrementing the
intensity of the signal in one ear (chosen
randomly) during the probe segment (relative to
the fringe). Monaural (2-Interval) - The monaural
conditions can be considered as situations in
which the left- and right-ear stimuli of the
binaural condition were presented to only the
left ear in two separate temporal intervals of
each trial. The interval containing the increment
or decrement was chosen randomly from trial to
trial.
References Akeroyd, M. A., and Bernstein, L. R.
(2001). The variation across time of sensitivity
to interaural disparities Behavioral
measurements and quantitative analyses, J.
Acoust. Soc. Am. 110, 2516-2526. Bernstein, L.
R., Trahiotis, C., Akeroyd, M. A., and Hartung,
K. (2001). Sensitivity to brief changes of
interaural time and interaural intensity, J.
Acoust. Soc. Am. 109, 1604-1615. Macmillan, N.A.
(1971). Detection and recognition of increments
and decrements in auditory intensity, Percept.
Psychophys. 10, 233-238. Oxenham, A. J. (1997).
Increment and decrement detection in sinusoids
as a measure of temporal resolution, J. Acoust.
Soc. Am. 102, 1779-1790. Stellmack, M. A.,
Viemeister, N. F., and Byrne, A. J. (2005).
Comparing monaural and interaural temporal
windows Effects of a temporal fringe on
sensitivity to intensity differences, J. Acoust.
Soc. Am., (In press). Zurek, P. M. (1980). The
precedence effect and its possible role in the
avoidance of interaural ambiguities, J. Acoust.
Soc. Am. 67, 952-964.
Fig. 2. Threshold intensity differences as a
function of forward fringe duration. Left panel
monaural conditions, right panel interaural. All
thresholds are the means of 4 listeners. Error
bars are standard errors of the mean.
Work supported by NIH-NIDCD grants R01-DC-00683
and R03-DC-05343.