The design matrix used for modelling the data consisted of onset vectors for the correct trials of each of the seven conditions (6 experimental, 1 control), with additional vectors for empty trials and cued response periods which included trial-by-trial RT as a parameter.
In order to detect potential ToM activation, we therefore applied this additional control task as opposed to the empty trials, which could mask actual ToM activity if contrasted with the experimental trials (Spreng, Mar, & Kim, 2009).
The remaining 30 trials were used to implement two independent control tasks: 18 empty trials with pink-noise playback, and 16 age task trials (with eight authentic and eight play-acted stimuli not used during experimental trials) in which participants had to determine the age of the speaker.
In addition, two control tasks were included: 16 word detection trials (utilizing the respective eight authentic and eight play-acted stimuli not used in the experimental task trials), in which participants had to count occurrences of the word und ("and"), and 18 empty trials with pink-noise playback.
Within 24 hours of the desk becoming empty, a trial screening by the BBFC of Baise-Moi (Rape Me), the most sexually-explicit mainstream film ever made, reminded us of the complexity of the decisions involved.
Delay testing and sample stimulus omission both resulted in a choose-short bias on filled interval trials, and a choose-long bias on empty interval trials.
This timing difference was not a result of pigeons timing marker duration on empty interval trials.
In Experiment 3, the results of varying the duration of the second marker on empty interval trials ruled out a reliance on timing and summing the duration of the markers themselves.
In Experiment 4, the sample durations were changed to 0.5 and 2 s in order to minimize the possibility that the birds were counting the markers, but a choose-long effect on empty interval trials was still evident.
