51
Visual and auditory localizers
In separate scan series, subjects performed different auditory and visual
localizers (see Table 1 for a summary). In the first localizer, subjects viewed low-contrast
random moving dots presented in the left or right hemifields alternating with stationary
dots. In the second localizer, subjects viewed real photographs of objects and scrambled
photographs, alternating with fixation baseline. In the third localizer, subjects heard
brief (1-2 s) recordings of a variety of non-linguistic stimuli, including recordings of
animal calls, recordings of man-made objects (both manual and powered), scrambled
versions of these recordings, and pure tones (Beauchamp, et al. 2004b). Subjects
performed a simple detection task during each Iocalizer to ensure attention to the
stimulus.
Experiment 1 and 2: fMRI experimental design and data analysis
fMRI data was analyzed using AFNI (Cox 1996). Individual cortical surface models
were created with FreeSurfer (Fischl, et al. 1999) and visualized in SUMA (Argali, et al.
2006). Localizer experiments were performed with a block design and analyzed with the
general linear model by convolving the timing of each type of stimulation block with a
gamma-variate function. Tactile experiments were conducted using a rapid event-
related design, and analyzed with finite impulse response deconvolution. This allows
estimation of the hemodynamic response to each trial type as if it had been presented
in isolation in a slow event-related design.