80 subjects were used. These were gathered opportunistically from within and around the Psychology Department at the University of Warwick. Equal numbers of male and female subjects were tested, all of whom were aged between 18 and 24 years, most of whom were undergraduates at Warwick. Procedure ConditionsEach subject was tested under a combination of two conditions, both of which had two levels. One condition was the presence of an audience (either there or not) and the other condition was the ambient presence of androstanone (either present or not). Thus, there were four possible treatments under which any subject could be tested.

  1. No androstanone and no audience
  2. No androstanone and an audience
  3. Androstanone and no audience
  4. Androstanone and an audience

Subjects were tested one at a time. They were led into the testing room and given an instruction sheet and a mood questionnaire, designed after Mayer, Allen and Beauregard (1995) so as to appear as normal as possible to the subjects, the majority of which would be Psychology undergraduates. The room was an anosmic chamber: a room with a constant flow of air through it to ensure that extraneous smells (such as those of the previous subjects) did not linger and contaminate the environment for later subjects. It contained the computer and printer on a desk, and a closed-circuit camera connected to a monitor outside the room. The monitor remained hidden by the door to the chamber whenever any subject might have been in a position to see it.

The subjects were asked to read the instructions and were left alone for two minutes in order to do so. After this time, subjects were instructed not to correct any errors they made and instructed to begin the first task. As this happened so the clock was started. If the subject was being tested under an audience condition, then a confederate observer was called into the testing room by the experimenter, who left. If the subject was being tested under a no audience condition then the experimenter left and surreptitiously watched the subject from outside on the CCTV system. After five minutes, the subject was instructed to stop. The subject then filled out the second part of the mood questionnaire and was instructed to read the second instruct ion sheet, which contained the other task. The experimental procedure for the second task was the same as that for the first. If the subject was taking part in an audience condition, then the observer stood behind the chair during the trials. This position was close to the outflow vent of the anosmic chamber, so that the observer's own bodily odours did not interfere with the experimental control.

During testing, the observer did not communicate with the subject. Once testing had been completed, subjects were then tested for androstanone sensitivity, by presenting them with three apparently identical perfumer's smelling strips, one of which was impregnated with androstanone, the other two of which were not. Subje cts had to correctly identify which strip smelled different for their results to be included. In this way androstanone non-reactive subjects could be excluded from the results. TasksEach subject had to perform two tasks, one easy and one hard. The order was randomised between subjects to avoid any kind of order effects. For the easy task, the subject had to type a familiar version his or her name as many times as possible in five minutes. For the hard task, the subject had to type the same name backwards, inserting a number between each letter, starting at 2 and rising by 2 with each letter, throughout the task. In addition to these tasks, each subject had to complete a mood ques tionnaire at before and after each task.


The computer was used for the typing tasks. The mood questionnaire which was used to administer the androstanone to the subjects. For the conditions which involved exposing the subject to androstanone, the mood questionnaire was impregnated with 100ul of 0.8mg/ml androstanone solution, using the fixed dose pipette. The ethanol base quickly evaporated, leaving the androstanone invisible on the questionnaires. The questionnaires were chosen for impregnation because only one A5 questionnaire was needed per subject, as compared with 2 A4 instruction sheets. A separate set of instruction sheets, and mood questionnaire, was used for each subject, so as to maintain as consistent an environment as possible between subjects. Pilot StudyA pilot study was conducted with 10 subjects to establish that


The number of characters typed in each task was counted and coded according to the experimental treatment of the subject. The presence or not of androstanone and the presence or not of an audience (i.e. the treatments) were the independent variables. The number of characters typed by each subject (referred to as the score) for each task were the dependent variables. The subjects performed both tasks each, but performed their tasks in only one combination of treatments, so the experiment was a 2 x 2 x 2 within-between design. The within factor (the task difficulty) which had two levels was crossed with the between factors (the treatments) which 2 x 2 levels, made up of the two audience conditions (an observer present, or not) and the two semiochemical conditions (androstanone present, or not). A repeated measures within-between 2 x2 x2 ANOVA was conducted to test for main effects and also task-treatment interactions.

Turn to the Results...