SECTION A
1. Experimenters reported on the effect of a 24-week course of cognitivebehavioural therapy (CBT) as a treatment for anorexia. There were 17 girls in this
study, and they were weighed before and after treatment. The weights of the girls, in
pounds, are given below:
Subject 1 2 3 4 5 6 7 8 9 10
Before 83.8 83.3 86.0 82.5 86.7 79.6 76.9 94.2 73.4 80.5
After 95.2 94.3 91.5 91.9 100.3 76.7 76.8 101.6 94.9 75.2
Subject 11 12 13 14 15 16 17
Before 81.6 82.1 77.6 83.5 89.9 86.0 87.3
After 77.8 95.5 90.7 92.5 93.8 91.7 98.0
(a) Specify the null and experimental (alternative) hypotheses, and run the appropriate
t test, justifying your choice and stating your conclusions.
(b) The data above suggest that girls receiving CBT gained weight over the course of
the therapy. However, it is possible that they gained weight simply because they got
older. In order to control for this effect, the amount of weight gained (or lost) by the
CBT group (n = 17) was contrasted with the amount of weight gained (or lost) in a
Control group (n = 26), who received no therapy over the same time period. The data
on weight gain for the two groups is shown below:
Control CBT
-0.5 3.3 11.4 9.0
-9.3 11.3 11.0 3.9
-5.4 0.0 5.5 5.7
12.3 -1.0 9.4 10.7
-2.0 -10.6 13.6
-10.2 -4.6 -2.9
-12.2 -6.7 -0.1
11.6 2.8 7.4
-7.1 0.3 21.5
6.2 1.8 -5.3
-0.2 3.7 -3.8
-9.2 15.9 13.4
8.3 -10.2 13.1
As before, run the appropriate t test to compare the group means and state your
conclusions.
(c) Calculate the 95% confidence interval on µ1 – µ2 for the data from the CBT and
control groups, and interpret your results accordingly.
SECTION B
2. How would you determine if two cognitive processes were performed in a
serial manner in the human mind. Please give an example.
3. It is known that aspirin doses can lead to tinnitus (‘ringing’ sounds heard
when no such external sound is present) in people. Design an experiment
to assess whether rats experience tinnitus when given a large dose of
aspirin.
4. Design an experiment to assess why, after experience of correlated
presentations of a stimulus (e.g., a red triangle) and reward (e.g., an
increase in bonus score shown on a screen), subjects are able to give
accurate ratings reflecting the degree of correlation between the two
events.
E