2.2 Stimuli and Procedure
The experiment was programmed using E-Prime 3.0 (Psychology Software Tools, Pittsburgh, PA). All stimuli were pure sinusoidal tones, with seven different frequencies (500 Hz, 550 Hz, 600 Hz, 650 Hz, 700 Hz, 750 Hz, and 800 Hz). The duration of each tone was 70 ms (with 5 ms rise and fall times), and the stimulus onset asynchrony (SOA) was fixed at 650 ms. All tones were presented at a comfortable, fixed volume. As a roving paradigm was used, deviants (the 1st tone in each sequence) would become subsequent standards. Hence, the physical properties of all deviants and standards were controlled.
EEG data were recorded using NetStation, with a Net Amps 300 amplifier and 128-channel Electrical Geodesics Incorporated (EGI) nets against a Cz reference. The sampling rate was 500 Hz, and the online filter setting was 0.1 Hz high-pass. The impedance was kept below 50 kΩ for each electrode. The data recording took place in a sound-attenuated laboratory. Before the experiment began, participants were told to watch a silent movie while listening to some sounds, and they were instructed to limit body movements during the experiment to minimize muscle artifacts.
In the current experiment, the first tone was randomly played among the seven types of tones with different frequencies. After that, there was an 85% chance that the next tone would be the same as the first tone (standard) and a 15% chance that the next tone would be different from the first tone (deviant). This criterion was kept unchanged throughout the experiment. The only exception was that if 30 tones were played consecutively, the next tone must have been a deviant, which happened only 2.73 times on average (SD = 1.59 times; range : 0–5 times) among around 300 sequences for each participant. This restriction was to ensure that there were sufficient trials to be analyzed. The purpose of increasing the maximum stimulus number in each train (30) compared with previous studies (mostly 10 or fewer) and using a random stimulus arrangement was to minimize expectations of the stimulus arrangement by participants. Throughout the experiment, participants watched a silent movie, “Tom and Jerry”, via a laptop during the experiment, when they were listening to the tones played by speakers 80 cm away from them. The current experiment ended after 2,000 tones had been played, which were divided into two blocks with 1,000 tones each. The duration of the experiment was around 20 minutes. Figure 1 shows the experimental paradigm.
After the EEG experiment, a tone discrimination task run by E-Prime 3.0 was conducted to confirm that participants were able to discriminate tones with a frequency difference of 50 Hz. In each trial, participants listened to a pair of tones. The properties of the tones were the same as those in the EEG experiment, and they were played in a pseudorandomized sequence. All pairs of tones in the experimental trials were with 50 Hz discrepancy, the least frequency difference that appeared in the EEG task, but tone pairs with larger frequency differences appeared in practice trials. Participants were told to press “f” on the keyboard when the two tones were the same and press “j” when they were different. It was also emphasized to participants that they should make the judgment as accurately and quickly as possible. This task began with eight practice trials to ensure that participants understood the task. In each trial, a fixation cross first appeared for one second, followed by a pair of tones (70 ms each), with an inter-tone interval of 280 ms. Then, a question mark appeared on the screen, which remained until participants had judged whether the two tones were the same or different by pressing the button. After that, the fixation cross reappeared as the next trial began. Accuracy and reaction time were shown immediately after the response in practice trials, but not in the main experiment. The main experiment consisted of 104 trials in total, and this task lasted for around five minutes. The average accuracy was 94.49% (SD = 5.38%; d’ = 3.56), implying that participants were able to discriminate the tones used in the experiment.