Lecture Notes for 04/26/00

Sensitivity

I. Overview

A. The ear has an extremely wide range of sensitivity

1. Dynamic range

a. Range of hearing from just detecting a sound

(absolute threshold) to the point of pain (pain threshold)

b. From approximately 0 dB SPL to 140 dB SPL

2. Audible frequency range

a. From approximately 20 Hz to 20,000 Hz for humans

b. From 2 Hz to 19 Hz can be heard by some people but

they report a lack of tonality

B. Changes in sensitivity due to internal and external factors

1. Aging (presbycusis) – increases in high frequency

thresholds

2. Noise – increases in thresholds centered around 4000 Hz

II. Absolute sensitivity (threshold of audibility)

A. Definition = the sound intensity necessary to just detect the

presence of a sound

B. Methods used:

1. Minimum audible field

a. Listener’s threshold measured in sound field

b. Sound pressure level is measured at place where

listener’s head was positioned during testing

2. Minimum audible pressure

a. Listener’s threshold measured under headphones

b. Sound pressure level delivered by the headphones is

measured in a coupler

III. International standards

A. Prior to 1964, different reference levels for normal hearing

were used in different countries

B. It was decided that there needed to be an international

standard for normal hearing levels

1. 1^st standards were published in 1969

2. They were updated in 1989 and again in 1996

3. Referred to as ANSI S3.6 – 1969, ANSI S3.6 – 1989, or

ANSI S3.6 - 1996

4. Give reference levels in dB SPL for several types of

headphones

a. Table 9.1 p.286

b. Know values for TDH-49/50 headphones

IV. Hearing level

A. Up to this point we have been talking about absolute

sensitivity in terms of dB SPL

B. A second way to look at absolute sensitivity is in dB HL

1. The dB HL scale uses the ANSI S3.6 standards for the

0 dB HL value

a. For TDH – 49 headphones:

at 125 Hz, 0 dB HL = 47.5 dB SPL

at 1000 Hz, 0 dB HL = 7.5 dB SPL

at 8000 Hz, o dB HL = 13.0 dB SPL

b. For ER- 3A insert earphones:

at 125 Hz, 0 dB HL = dB SPL

at 1000 Hz, 0 dB HL = dB SPL

at 8000 Hz, o dB HL = dB SPL

2. The dB HL scale calls each zero reference SPL value 0 dB

HL so that threshold can be measured in comparison to a

straight line rather than a curved one

a. dB HL is the norm according to the ASHA standards for

audiology in the US

b. dB SPL is the norm in many European countries

c. Figure 9.2

1) a = dB SPL

2) b = dB HL

C. A third way to look at sensitivity (relative to absolute) is in

dB SL

1. The dB SL scale uses the person’s own threshold as the

reference level

2. If a sound is presented at 10 dB SL, then it is presented at

10 dB above his/her absolute threshold à can be dB SPL

OR dB HL

V. Duration Effects

A. The duration of the sound will have an effect on sensitivity

1. At very brief durations, you hear a click rather than a tone

2. Also depends on frequency of sound

a. 15 ms at 500 Hz to hear tonal quality

b. 10 ms at 1000 Hz to hear tonal quality

B. Temporal integration

1. Ear acts as an energy detector which samples energy

present within a certain time window

2. Up until between 200 ms to 300 ms, you get a 10 dB

decrease in threshold for every tenfold increase in duration

3. Figures 9.3, 9.4

VI. Differential sensitivity

A. Definition = smallest perceivable difference between 2

sounds

B. Also see referred to as DL (difference limen) or jnd (just

noticeable difference)

C. Weber fraction

1. DL over starting (baseline) level

DL DS

--- - or ------

starting level S

D. Weber’s law

1. The value of Weber’s fraction is constant regardless of

stimulus

------ = k

3. E.g.

If you have 10 candles, only need 1 more to see difference

in light level

If you have 100 candles, need 10 more to see difference in

light level

If you have 1000 candles, need 100 more to see difference

in light level

1/10 = 10/100 = 100/1000 = 0.1

4. This is a nice idea but we will see when we get to

intensity/loudness that it doesn’t always quite happen this

way

Masking

I. Definition = the threshold of audibility for one sound is raised by

the presence of another sound

A. The target sound is the probe or signal

B. The other sound is the masker

C. E.g.

Tone A (signal) alone à threshold = 10 dB SPL

Tone A (signal) + Tone B (masker) à threshold =

15 dB SPL

Tone B is said to provide 5 dB SPL masking to Tone A

II. The frequency of the masker and the signal affects the amount

of masking obtained

A. Low frequency sounds provide more masking than higher

frequency sounds

B. This is referred to as the upward spread of masking

III. Critical Band concept

A. The ear is made up of many filters (think back to the BM

physiology lecture)

B. This was shown in Fletcher’s band-widening experiment

1. Task was to discriminate the signal from the masker (noise

in this case)

2. Fletcher increased the bandwidth of the noise masker and

measured its effect on the signal threshold

C. The width of the auditory filter is referred to as the critical

band

1. Because the ear is a very good narrowband filter, anything

outside of the auditory filter affects hearing minimally

2. Only low bandwidths (~ 100 Hz) are effective in masking

(i.e. only low bandwidths are effective in raising tone

threshold)

3. At bandwidths greater than critical bandwidth, only

masking effect of energy inside the critical band is felt (rest

is wasted – i.e. it doesn’t add to the effective masking)

IV. Types of masking

A. Simultaneous

1. Figure on overhead

2. The two sounds (signal and masker) are presented at the

same time

B. Forward

1. Figure on overhead

2. The masker is presented just before the test tone

C. Backward

1. Figure on overhead

2. The masker is presented just after the test tone

3. Tallal (1973) and recently Wright et al. (1997) found that

this type of testing differentiated children with SLI from

normal language children

D. These will all give slightly different amounts of masking

V. Experiments with masking (handout)