Olfactory measurement

We measured olfactory detection thresholds using a pulse ejection system called the “Fragrance Jet for Medical Checkup” (Fig. 2) (Keio University) [20]. The stimuli used were isoamyl acetate and allyl caproate, which were simple chemicals and, unlike many natural fragrances, were not affected by the preservation method and had the added advantage of being easy to reproduce.

The display emitted scents from multiple holes at the same time, allowing the number of simultaneous ejections (NSE) to be set in the range of 0 to 255. Before the experiment, we confirmed that all participants understood the rules of the experiment. We began with an NSE of 80 and an ET of 200 ms based on the results of preliminary experiments and our earlier study [20]. The touch panel system displayed three boxes, each of which contained a stimulus (Fig. 3), and the olfactory detection threshold was assessed using a triple forced-choice procedure where three stimuli were presented at random (one stimulus was scented while the other two were odorless). All children were fascinated by our touch panel display. The presentation was as simple as possible so that all children, including those with ASD, could concentrate during the experiment. The subject’s task was to identify the box that contained the scented stimulus. When the participants pushed one of the three boxes, an odor was given off 3.0 s later. The participant was allowed to push each box up to two times. After two successful trials, the EQ fell by 50 %. This was continued until the participant made a mistake and selected the box that contained an odorless stimulus or until the participant cleared an NSE of 10. We administered the odors from strongest to weakest to maintain the motivation and concentration of all children. The detection threshold was generated after the procedure was completed. Measurements were finished approximately 5 min after the start of the procedure and were influenced very little by lingering scent trails. The olfactory detection threshold was defined using a logarithmic function as the NSE of the last trial. The mathematical process used to determine the olfactory detection threshold is as follows:

Touch panel display. The touch panel is comprised of three boxes. Three stimuli are presented at random, of which one is scented and the other two are odorless. When participants push a box, odors are given off

If participants were not able to identify an NSE of 80, their score was 5. Participants that were able to identify an NSE of 80, but could not identify an NSE of 40, were given a score of 4. Finally, if a participant was able to identify an NSE of 40, 20, and 10, their score was 3, 2, and 1, respectively. A subject’s scores could range from 5 (unable to detect even the highest concentration) to 1 (able to detect the lowest concentration) for each odor.

Participants neither ate nor drank anything but water for 30 min prior to testing. Because temperature can influence odor, the temperature of the experimental room was maintained at (21–23 °C) during the study [34]. Experiments were conducted under conditions in which the participant and experimenter were diagonally opposite. A board was placed between them to prevent participants from being able to see the procedures and results. Each participant was required to sit in front of the olfactory display and position their chin on the chin rest such that the distance from the olfactory ejection point to the nose was fixed.