Modelling Exercises 6.5 

Introduction
This Section provides examples and tasks employing exponential functions and logarithmic functions,
such as growth and decay models which are important throughout science and engineering.

' $
• be familiar with the laws of logarithms

Prerequisites • have knowledge of logarithms to base 10

Before starting this Section you should . . . • be able to solve equations involving
logarithms and exponentials
&
 %


Learning Outcomes • develop exponential growth and decay models

On completion you should be able to . . .
 

38 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

1. Exponential increase

Task
(a) Look back at Section 6.2 to review the definitions of an exponential function
and the exponential function.

(b) List examples in this Workbook of contexts in which exponential functions

are appropriate.

Your solution

Answer
(a) An exponential function has the form y = ax where a > 0. T he exponential function has
the form y = ex where e = 2.718282......

(b) It is stated that exponential functions are useful when modelling the shape of a hanging chain
or rope under the effect of gravity or for modelling exponential growth or decay.

We will look at a specific example of the exponential function used to model a population increase.

Task
Given that
P = 12e0.1t (0 ≤ t ≤ 25)
where P is the number in the population of a city in millions at time t in years
answer these questions.

(a) What does this model imply about P when t = 0?

(b) What is the stated upper limit of validity of the model?

(c) What does the model imply about values of P over time?

(d) What does the model predict for P when t = 10? Comment on this.

(d) What does the model predict for P when t = 25? Comment on this.

HELM (2008): 39
Section 6.5: Modelling Exercises
Your solution

(a)

(b)

(c)

(d)

(e)

Answer
(a) At t = 0, P = 12 which represents the initial population of 12 million. (Recall that e0 = 1.)

(b) The time interval during which the model is valid is stated as (0 ≤ t ≤ 25) so the model is
intended to apply for 25 years.

(c) This is exponential growth so P will increase from 12 million at an accelerating rate.

(d) P (10) = 12e1 ≈ 33 million. This is getting very large for a city but might be attainable in
10 years and just about sustainable.

(e) P (25) = 12e2.5 ≈ 146 million. This is unrealistic for a city.

Note that exponential population growth of the form P = P0 ekt means that as t becomes large and
positive, P becomes very large. Normally such a population model would be used to predict values
of P for t > 0, where t = 0 represents the present or some fixed time when the population is known.
In Figure 6, values of P are shown for t < 0. These correspond to extrapolation of the model into
the past. Note that as t becomes increasingly negative, P becomes very small but is never zero or
negative because ekt is positive for all values of t. The parameter k is called the instantaneous
fractional growth rate.

P
30
P = 12e0.1 t
25

20

15

10

10 5 0 5 10 t

Figure 6: The function P = 12e0.01t

40 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

For the model P = 12ekt we see that k = 0.1 is unrealistic, and more realistic values would be
k = 0.01 or k = 0.02. These would be similar but k=0.02 implies a faster growth for t > 0 than
k = 0.01. This is clear in the graphs for k = 0.01 and k = 0.02 in Figure 7. The functions are
plotted up to 200 years to emphasize the increasing difference as t increases.

P
500 P = 12e0.02 t

250

P = 12e0.01 t

t
0 50 100 150 200

Figure 7: Comparison of the functions P = 12e0.01t and P = 12e0.02t

The exponential function may be used in models for other types of growth as well as population
growth. A general form may be written
y = aebx a > 0, b > 0, c≤x≤d
where a represents the value of y at x = 0. The value a is the intercept on the y-axis of a graphical
representation of the function. The value b controls the rate of growth and c and d represent limits
on x.
In the general form, a and b represent the parameters of the exponential function which can be
selected to fit any given modelling situation where an exponential function is appropriate.

2. Linearisation of exponential functions

This subsection relates to the description of log-linear plots covered in Section 6.6.
Frequently in engineering, the question arises of how the parameters of an exponential function
might be found from given data. The method follows from the fact that it is possible to ‘undo’
the exponential function and obtain a linear function by means of the logarithmic function. Before
showing the implications of this method, it may be necessary to remind you of some rules for
manipulating logarithms and exponentials. These are summarised in Table 1 on the next page, which
exactly matches the general list provided in Key Point 8 in Section 6.3 (page 22.)

HELM (2008): 41
Section 6.5: Modelling Exercises
 Table 1: Rules for manipulating base e logarithms and exponentials
Number Rule Number Rule
1a ln(xy) = ln(x) + ln(y) 1b e × ey = ex+y
x

2a ln(x/y) = ln(x) − ln(y) 2b ex /ey = ex−y

3a ln(xy ) = y ln(x) 3b (ex )y = exy
4a ln(ex ) = x 4b eln(x) = x
5a ln(e) = 1 5b e1 = e
6a ln(1) = 0 6b e0 = 1
We will try ‘undoing’ the exponential in the particular example
P = 12e0.1t
We take the natural logarithm (ln) of both sides, which means logarithm to the base e. So
ln(P ) = ln(12e0.1t )
The result of using Rule 1a in Table 1 is
ln(P ) = ln(12) + ln(e0.1t ).
The natural logarithmic functions ‘undoes’ the exponential function, so by Rule 4a,
ln(e0.1t ) = 0.1t
and the original equation for P becomes
ln(P ) = ln(12) + 0.1t.
Compare this with the general form of a linear function y = ax + b.
y = ax + b
↓ ↓ ↓
ln(P ) = 0.1t + ln(12)
If we regard ln(P ) as equivalent to y, 0.1 as equivalent to the constant a, t as equivalent to x, and
ln(12) as equivalent to the constant b, then we can identify a linear relationship between ln(P ) and
t. A plot of ln(P ) against t should result in a straight line, of slope 0.1, which crosses the ln(P )
axis at ln(12). (Such a plot is called a log-linear or log-lin plot.) This is not particularly interesting
here because we know the values 12 and 0.1 already.
Suppose, though, we want to try using the general form of the exponential function
P = aebt (c ≤ t ≤ d)
to create a continuous model for a population for which we have some discrete data. The first thing
to do is to take logarithms of both sides
ln(P ) = ln(aebt ) (c ≤ t ≤ d).
Rule 1 from Table 1 then gives
ln(P ) = ln(a) + ln(ebt ) (c ≤ t ≤ d).
But, by Rule 4a, ln(ebt ) = bt, so this means that
ln(P ) = ln(a) + bt (c ≤ t ≤ d).

42 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

So, given some ‘population versus time’ data, for which you believe can be modelled by some version
of the exponential function, plot the natural logarithm of population against time. If the exponential
function is appropriate, the resulting data points should lie on or near a straight line. The slope of
the straight line will give an estimate for b and the intercept with the ln(P ) axis will give an estimate
for ln(a). You will have carried out a logarithmic transformation of the original data for P . We
say the original variation has been linearised.
A similar procedure will work also if any exponential function rather than the base e exponential
function is used. For example, suppose that we try to use the function
P = A × 2Bt (C ≤ t ≤ D),
where A and B are constant parameters to be derived from the given data. We can take natural
logarithms again to give
ln(P ) = ln(A × 2Bt ) (C ≤ t ≤ D).
Rule 1a from Table 1 then gives
ln(P ) = ln(A) + ln(2Bt ) (C ≤ t ≤ D).
Rule 3a then gives
ln(2Bt ) = Bt ln(2) = B ln(2) t
and so
ln(P ) = ln(A) + B ln(2) t (C ≤ t ≤ D).
Again we have a straight line graph with the same intercept as before, ln A, but this time with slope
B ln(2).

Task
The amount of money £M to which £1 grows after earning interest of 5% p.a.
for N years is worked out as
M = 1.05N
Find a linearised form of this equation.

Your solution

Answer
Take natural logarithms of both sides.
ln(M ) = ln(1.05N ).
Rule 3b gives
ln(M ) = N ln(1.05).
So a plot of ln(M ) against N would be a straight line passing through (0, 0) with slope ln(1.05).

HELM (2008): 43
Section 6.5: Modelling Exercises
The linearisation procedure also works if logarithms other than natural logarithms are used. We start
again with
P = A × 2Bt (C ≤ t ≤ D).
and will take logarithms to base 10 instead of natural logarithms. Table 2 presents the laws of
logarithms and indices (based on Key Point 8 page 22) interpreted for log10 .
Table 2: Rules for manipulating base 10 logarithms and exponentials

Number Rule Number Rule

1a log10 (AB) = log10 A + log10 B 1b 10A 10B = 10A+B
2a log10 (A/B) = log10 A − log10 B 2b 10A /10B = 10A−B
3a log10 (Ak ) = k log10 A 3b (10A )k = 10kA
4a log10 (10A ) = A 4b 10log10 A = A
5a log10 10 = 1 5b 101 = 10
6a log10 1 = 0 6b 100 = 1
Taking logs of P = A × 2Bt gives:
log10 (P ) = log10 (A × 2Bt ) (C ≤ t ≤ D).
Rule 1a from Table 2 then gives
log10 (P ) = log10 (A) + log10 (2Bt ) (C ≤ t ≤ D).
Use of Rule 3a gives the result
log10 (P ) = log10 (A) + B log10 (2) t (C ≤ t ≤ D).

Task
(a) Write down the straight line function corresponding to taking logarithms of
the general exponential function
P = aebt (c ≤ t ≤ d)

by taking logarithms to base 10.

(b) Write down the slope of this line.

Your solution

Answer
(a) log10 (P ) = log10 (a) + (b log10 (e))t (c ≤ t ≤ d)

(b) b log10 (e)

It is not usually necessary to declare the subscript 10 when indicating logarithms to base 10. If you
meet the term ‘log’ it will probably imply “to the base 10”. In the remainder of this Section, the
subscript 10 is dropped where log10 is implied.

44 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

3. Exponential decrease
Consider the value, £D, of a car subject to depreciation, in terms of the age A years of the car. The
car was bought for £10500. The function
D = 10500e−0.25A (0 ≤ A ≤ 6)
could be considered appropriate on the ground that (a) D had a fixed value of £10500 when
A = 0, (b) D decreases as A increases and (c) D decreases faster when A is small than when A is
large. A plot of this function is shown in Figure 8.

12000

10000

8000
D pounds
6000

4000

2000 A years
0 1 2 3 4 5 6

Figure 8: Plot of car depreciation over 6 years

Task
Produce the linearised model of D = 10500e−0.25A .

Your solution

Answer
ln D = ln 10500 + ln(e−0.25A )
so ln D = ln 10500 − 0.25A

HELM (2008): 45
Section 6.5: Modelling Exercises
 Engineering Example 2

Exponential decay of sound intensity

Introduction
The rate at which a quantity decays is important in many branches of engineering and science. A
particular example of this is exponential decay. Ideally the sound level in a room where there are
substantial contributions from reflections at the walls, floor and ceiling will decay exponentially once
the source of sound is stopped. The decay in the sound intensity is due to absorbtion of sound at the
room surfaces and air absorption although the latter is significant only when the room is very large.
The contributions from reflection are known as reverberation. A measurement of reverberation in
a room of known volume and surface area can be used to indicate the amount of absorption.
Problem in words
As part of an emergency test of the acoustics of a concert hall during an orchestral rehearsal,
consultants asked the principal trombone to play a single note at maximum volume. Once the sound
had reached its maximum intensity the player stopped and the sound intensity was measured for the
next 0.2 seconds at regular intervals of 0.02 seconds. The initial maximum intensity at time 0 was
1. The readings were as follows:
time 0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20
intensity 1 0.63 0.35 0.22 0.13 0.08 0.05 0.03 0.02 0.01 0.005
Draw a graph of intensity against time and, assuming that the relationship is exponential, find a
function which expresses the relationship between intensity and time.
Mathematical statement of problem
If the relationship is exponential then it will be a function of the form
I = I0 10kt
and a log-linear graph of the values should lie on a straight line. Therefore we can plot the values
and find the gradient and the intercept of the resulting straight-line graph in order to find the values
for I0 and k.
k is the gradient of the log-linear graph i.e.
change in log10 (intensity)
k=
change in time
and I0 is found from where the graph crosses the vertical axis log10 (I0 ) = c
Mathematical analysis
Figure 9(a) shows the graph of intensity against time.

46 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

We calculate the log10 (intensity) to create the table below:

time 0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20
log10 (intensity) 0 -0.22 -0.46 -0.66 -0.89 -1.1 -1.3 -1.5 -1.7 -2.0 -2.2
Figure 9(b) shows the graph of log (intensity) against time.

Intensity Log(Intensity)
0 (0, 0)

−1

(0.2, −2.2)
−2

0 0.1 0.2 Time 0 0.1 0.2 Time

(a) (b)

Figure 9: (a) Graph of sound intensity against time (b) Graph of log10 (intensity) against time
and a line fitted by eye to the data. The line goes through the points (0, 0) and (0.2, −2.2).

We can see that the second graph is approximately a straight line and therefore we can assume that
the relationship between the intensity and time is exponential and can be expressed as
I = I0 10kt .
The log10 of this gives
log10 (I) = log10 (I0 ) + kt.
From the graph (b) we can measure the gradient, k using
change in log10 (intensity)
k=
change in time
−2.2 − 0
giving k = = −11
0.2 − 0
The point at which it crosses the vertical axis gives
log10 (I0 ) = 0 ⇒ I0 = 100 = 1
Therefore the expression I = I0 10kt becomes
I = 10−11t
Interpretation
The data recorded for the sound intensity fit exponential decaying with time. We have used a
log-linear plot to obtain the approximate function:
I = 10−11t

HELM (2008): 47
Section 6.5: Modelling Exercises
4. Growth and decay to a limit
Consider a function intended to represent the speed of a parachutist after the opening of the parachute
where v m s−1 is the instantanous speed at time t s. An appropriate function is
v = 12 − 8e−1.25t (t ≥ 0),
We will look at some of the properties and modelling implications of this function. Consider first the
value of v when t = 0:
v = 12 − 8e0 = 12 − 8 = 4
This means the function predicts that the parachutist is moving at 4 m s−1 when the parachute
opens. Consider next the value of v when t is arbitrarily large. For such a value of t, 8e−1.25t would
be arbitrarily small, so v would be very close to the value 12. The modelling interpretation of this is
that eventually the speed becomes very close to a constant value, 12 m s−1 which will be maintained
until the parachutist lands.
The steady speed which is approached by the parachutist (or anything else falling against air resis-
tance) is called the terminal velocity. The parachute, of course, is designed to ensure that the
terminal velocity is sufficiently low (12 m s−1 in the specific case we have looked at here) to give a
reasonably gentle landing and avoid injury.
Now consider what happens as t increases from near zero. When t is near zero, the speed will be
near 4 m s−1 . The amount being subtracted from 12, through the term 8e−1.25t , is close to 8 because
e0 = 1. As t increases the value of 8e−1.25t decreases fairly rapidly at first and then more gradually
until v is very nearly 12. This is sketched in Figure 10. In fact v is never equal to 12 but gets
imperceptibly close as anyone would like as t increases. The value shown as a horizontal broken line
in Figure 10 is called an asymptotic limit for v.

15

10
1
v (m s )

0
3 5
t (s)
0 1 2 4

Figure 10: Graph of a parachutist’s speed against time

The model concerned the approach of a parachutist’s velocity to terminal velocity but the kind of
behaviour portrayed by the resulting function is useful generally in modelling any growth to a limit.
A general form of this type of growth-to-a-limit function is
y = a − be−kx (C ≤ x ≤ D)
where a, b and k are positive constants (parameters) and C and D represent values of the independent
variable between which the function is valid. We will now check on the properties of this general
function. When x = 0, y = a − be0 = a − b. As x increases the exponential factor e−kx gets smaller,
so y will increase from the value a − b but at an ever-decreasing rate. As be−kx becomes very small,

48 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

y, approaches the value a. This value represents the limit, towards which y grows. If a function of
this general form was being used to create a model of population growth to a limit, then a would
represent the limiting population, and a − b would represent the starting population.
There are three parameters, a, b, and k in the general form. Knowledge of the initial and limiting
population only gives two pieces of information. A value for the population at some non-zero time is
needed also to evaluate the third parameter k.
As an example we will obtain a function to describe a food-limited bacterial culture that has 300
cells when first counted, has 600 cells after 30 minutes but seems to have approached a limit of 4000
cells after 18 hours.
We start by assuming the general form of growth-to-a-limit function for the bacteria population, with
time measured in hours
P = a − be−kt (0 ≤ t ≤ 18).
When t = 0 (the start of counting), P = 300. Since the general form gives P = a − b when t = 0,
this means that
a − b = 300.
The limit of P as t gets large, according to the general form P = a − b−kt , is a, so a = 4000. From
this and the value of a − b, we deduce that b = 3700. Finally, we use the information that P = 600
when t (measuring time in hours) = 0.5. Substitution in the general form gives
600 = 4000 − 3700e−0.5k

3400 = 3700e−0.5k
3400
= e−0.5k
3700
Taking natural logs of both sides:
 
3400 34
ln = −0.5k so k = −2 ln( ) = 0.1691
3700 37
Note, as a check, that k turns out to be positive as required for a growth-to-a-limit behaviour. Finally
the required function may be written
P = 4000 − 3700e−0.1691t (0 ≤ t ≤ 18).
As a check we should substitute t = 18 in this equation. The result is P = 3824 which is close to
the required value of 4000.

HELM (2008): 49
Section 6.5: Modelling Exercises
 Task
Find a function that could be used to model the growth of a population that
has a value of 3000 when counts start, reaches a value of 6000 after 1 year but
approaches a limit of 12000 after a period of 10 years.

(a) First find the modelling equation:

Your solution

Answer
Start with
P = a − be−kt (0 ≤ t ≤ 10).
where P is the number of members of the population at time t years. The given data requires that
a is 12000 and that a − b = 3000, so b = 9000.
The corresponding curve must pass through (t = 1, P = 6000) so
6000 = 12000 − 9000e−k
 t
−k 12000 − 6000 2 −kt −k t 2
e = = so e = (e ) = (using Rule 3b, Table 1, page 42)
9000 3 3
So the population function is
 t
2
P = 12000 − 9000 (0 ≤ t ≤ 10).
3
Note that P (10) according to this formula is approximately 11840, which is reasonably close to the
required value of 12000.

(b) Now sketch this function:

50 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

Your solution

Answer
P

12000
10000

5000

0 t (s)
0 2 4 6 8 10

5. Inverse square law decay

Engineering Example 3

Inverse square law decay of electromagnetic power

Introduction
Engineers are concerned with using and intercepting many kinds of wave forms including electromag-
netic, elastic and acoustic waves. In many situations the intensity of these signals decreases with
the square of the distance. This is known as the inverse square law. The power received from a
beacon antenna is expected to conform to the inverse square law with distance.
Problem in words
Check whether the data in the table below confirms that the measured power obeys this behaviour
with distance.
Power received, W 0.393 0.092 0.042 0.021 0.013 0.008
Distance from antenna, m 1 2 3 4 5 6

HELM (2008): 51
Section 6.5: Modelling Exercises
Mathematical statement of problem
A
Represent power by P and distance by r. To show that the data fit the function P = where
r2
A is a constant, plot log(P ) against log(r) (or plot the ‘raw’ data on log-log axes) and check

(a) how close the resulting graph is to that of a straight line

(b) how close the slope is to 2.

Mathematical analysis
The values corresponding to log(P ) and log(r) are
log(P ) -0.428 -1.041 -1.399 -1.653 -1.851 -2.012
log(r) 0 0.301 0.499 0.602 0.694 0.778
These are plotted in Figure 11 and it is clear that they lie close to a straight line.

− 0.5

−1
log(P ) − 1.5

−2
−2.5 log(r)
0 0.2 0.4 0.6 0.8

Figure 11
The slope of a line through the first and third points can be found from
−1.399 − (−0.428)
= −2.035
0.499 − 0
The negative value means that the line slopes downwards for increasing r. It would have been possible
to use any pair of points to obtain a suitable line but note that the last point is least ‘in line’ with
A
the others. Taking logarithms of the equation P = n gives log(P ) = log(A) − n log(r)
r
The inverse square law corresponds to n = 2. In this case the data yield n = 2.035 ≈ 2. Where
log(r) = 0, log(P ) = log(A). This means that the intercept of the line with the log(P ) axis gives
the value of log(A) = −0.428. So A = 10 − 0.428 = 0.393.
Interpretation
If the power decreases with distance according to the inverse square law, then the slope of the line
should be −2. The calculated value of n = 2.035 is sufficiently close to confirm the inverse square
law. The values of A and n calculated from the data imply that P varies with r according to
0.4
P =
r2
The slope of the line on a log-log plot is a little larger than −2. Moreover the points at 5 m and 6 m
range fall below the line so there may be additional attenuation of the power with distance compared
with predictions of the inverse square law.

52 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

Exercises
1. Sketch the graphs of (a) y = et (b) y = et + 3 (c) y = e−t (d) y = e−t − 1

2. The figure below shows the graphs of y = et , y = 2et and y = e2t .

16 y e2t
14 2et
12
10
8
et
6
4
2

2 1 0 1 2 t

State in words how the graphs of y = 2et and y = e2t relate to the graph of y = et .

3. The figures below show graphs of y = −e−t , y = 4 − e−t and y = 4 − 3e−t .

y
y y
1 2
−1 0 t 4
4
−1 y = −e−t y = 4 − e−t 2 y = 4 − 3e−t
3
2 −1 0 1 2 t
−2 −2
1
−4
−3
t
−1 0 1 2

Use the above graphs to help you to sketch graphs of (a) y = 5 − e−t (b) y = 5 − 2e−t

4. (a) The graph (a) in the figure below has an equation of the form

y = A + e−kt , where A and k are constants. What is the value of A?

(b) The graph (b) below has an equation of the form y = Aekt where A and k are constants.
What is the value of A?
(c) Write down a possible form of the equation of the exponential graph (c) giving numerical
values to as many constants as possible.
(d) Write down a possible form of the equation of the exponential graph (d) giving numerical
values to as many constants as possible.

HELM (2008): 53
Section 6.5: Modelling Exercises
 y y

2 -------------------------- 5

t t
(a) (b)
y
y

6 --------------------------------
3
2
1 ------------------------------
t t
(c) (d)

Answers

1.
y
et + 3

e−t et
e−t − 1
4
1

1
t
2 1 0 2

2. (a) y = 2et is the same shape as y = et but with all y values doubled.
(b) y = e2t is much steeper than y = et for t > 0 and much flatter for t < 0. Both pass
through (0, 1). Note that y = e2t = (et )2 so each value of y = e2t is the square of the
corresponding value of y = et .
y 5 − et y
6
4
4 − 3 et
2 3
3. (a) t (b) t

4. (a) 2 (b) 5 (c) y = 6 − 4e−kt (d) y = 1 + 2e−kt

54 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

6. Logarithmic relationships
Experimental psychology is concerned with observing and measuring human response to various
stimuli. In particular, sensations of light, colour, sound, taste, touch and muscular tension are
produced when an external stimulus acts on the associated sense. A nineteenth century German,
Ernst Weber, conducted experiments involving sensations of heat, light and sound and associated
stimuli. Weber measured the response of subjects, in a laboratory setting, to input stimuli measured
in terms of energy or some other physical attribute and discovered that:

(1) No sensation is felt until the stimulus reaches a certain value, known as the threshold value.

(2) After this threshold is reached an increase in stimulus produces an increase in sensation.

(3) This increase in sensation occurs at a diminishing rate as the stimulus is increased.

Task
(a) Do Weber’s results suggest a linear or non-linear relationship between sensa-
tion and stimulus? Sketch a graph of sensation against stimulus according
to Weber’s results.

(b) Consider whether an exponential function or a growth-to-a-limit function

might be an appropriate model.

Answer
(a) Non-linearity is required by observation (3).

10

S 5

0
0 2 4 6 8 10
P

(b) An exponential-type of growth is not appropriate for a model consistent with these experimen-
tal results, since we need a diminishing rate of growth in sensation as the stimulus increases.
A growth-to-a-limit type of function is not appropriate since the data, at least over the range
of Weber’s experiments, do not suggest that there is a limit to the sensation with continuing
increase in stimulus; only that the increase in sensation occurs more and more slowly.

A late nineteenth century German scientist, Gustav Fechner, studied Weber’s results. Fechner sug-
gested that an appropriate function modelling Weber’s findings would be logarithmic. He suggested
that the variation in sensation (S) with the stimulus input (P ) is modelled by

HELM (2008): 55
Section 6.5: Modelling Exercises
 S = A log(P/T ) (0 < T ≤ 1)
where T represents the threshold of stimulus input below which there is no sensation and A is a
constant. Note that when P = T, log(P/T ) = log(1) = 0, so this function is consistent with item
(1) of Weber’s results. Recall also that log means logarithm to base 10, so when P = 10T, S =
A log(10) = A. When P = 100T, S = A log(100) = 2A. The logarithmic function predicts that
a tenfold increase in the stimulus input from T to 10T will result in the same change in sensation
as a further tenfold increase in stimulus input to 100T . Each tenfold change is stimulus results in
a doubling of sensation. So, although sensation is predicted to increase with stimulus, the stimulus
has to increase at a faster and faster rate (i.e. exponentially) to achieve a given change in sensation.
These points are consistent with items (2) and (3) of Weber’s findings. Fechner’s suggestion, that
the logarithmic function is an appropriate one for a model of the relationship between sensation and
stimulus, seems reasonable. Note that the logarithmic function suggested by Weber is not defined
for zero stimulus but we are only interested in the model at and above the threshold stimulus, i.e.
for values of the logarithm equal to and above zero. Note also that the logarithmic function is useful
for looking at changes in sensation relative to stimulus values other than the threshold stimulus.
According to Rule 2a in Table 2 on page 42, Fechner’s sensation function may be written
S = A log(P/T ) = A[log(P ) − log(T )] (P ≥ T > 0).
Suppose that the sensation has the value S1 at P1 and S2 at P2 , so that
S1 = A[log(P1 ) − log(T )] (P1 ≥ T > 0),
and
S2 = A[log(P2 ) − log(T )] (P2 ≥ T > 0).
If we subtract the first of these two equations from the second, we get
S2 − S1 = A[log(P2 ) − log(P1 )] = A log(P2 /P1 ),
where Rule 2a of Table 2 has been used again for the last step. According to this form of equation,
the change in sensation between two stimuli values depends on the ratio of the stimuli values.
We start with
S = A log(P/T ) (1 ≥ T > 0).
Divide both sides by A:
S P
= log (1 ≥ T > 0).
A T
‘Undo’ the logarithm on both sides by raising 10 to the power of each side:
P
10S/A = 10log(P/T ) = (1 ≥ T > 0), using Rule 4b of Table 2.
T
So P = T × 10S/A (1 ≥ T > 0) which is an exponential relationship between stimulus and
sensation.
A logarithmic relationship between sensation and stimulus therefore implies an exponential rela-
tionship between stimulus and sensation. The relationship may be written in two different forms with
the variables playing opposite roles in the two functions.
The logarithmic relationship between sensation and stimulus is known as the Weber-Fechner Law of
Sensation. The idea that a mathematical function could describe our sensations was startling when

56 HELM (2008):
Workbook 6: Exponential and Logarithmic Functions
 ®

first propounded. Indeed it may seem quite amazing to you now. Moreover it doesn’t always work.
Nevertheless the idea has been quite fruitful. Out of it has come much quantitative experimental
psychology of interest to sound engineers. For example, it relates to the sensation of the loudness of
sound. Sound level is expressed on a logarithmic scale. At a frequency of 1 kHz an increase of 10
dB corresponds to a doubling of loudness.

Task
x
Given a relationship between y and x of the form y = 3 log( ) (x ≥ 4), find
4
the relationship between x and y.

Your solution

Answer
One way of answering is to compare with the example preceding this task. We have y in place of
S, x in place of P , 3 in place of A, 4 in place of T . So it is possible to write down immediately
x = 4 × 10y/3 (y ≥ 0)
Alternatively we can manipulate the given expression algebraically.
Starting with y = 3 log(x/4), divide both sides by 3 to give y/3 = log(x/4).
Raise 10 to the power of each side to eliminate the log, so that 10y/3 = x/4.
Multiply both sides by 4 and rearrange, to obtain x = 4 × 10y/3 , as before.
The associated range is the result of the fact that x ≥ 4, so 10y/3 ≥ 1, so y/3 > 0 which means
y > 0.

HELM (2008): 57
Section 6.5: Modelling Exercises