HOW-TO #5
Experiment Background
--------------
THE EXPERIMENT
--------------
This experiment used FMRI to examine the organization of brain responses to
different types of complex visual motion. Participants were presented with
visual displays depicting the movement of humans (e.g., doing jumping jacks) or
tools (e.g., a hammer swinging). The displays were either movies of real people
and tools engaged in motion, or point-light displays mimicking the movement of
people and tools. The point-light displays were created by identifying specific
points on the human body and tools, which moved with motion vectors equivalent
to the movie stimuli.
Design:
------
Event-related
Stimulus Conditions:
-------------------
1. Human Movie
2. Tool Movie
3. Human Point Light
4. Tool Point Light
Participants:
------------
The original sample size (n) for this study was 9 participants.
However, due to the tremendous amount of data collected for each
subject, we have included the data of only 3 participants that can be
downloaded from this how-to on the AFNI website. Participants are
identified by their two-letter ID codes ED, EE, and EF. Likewise, data
for each subject can be found in directories ED/, EE/, and EF/. Each
subject directory contains one anatomical volume, and 10 time-series
datasets (i.e., 10 runs).
--------
THE DATA
--------
Anatomical Data Collected:
-------------------------
The SPGR anatomical data consists of 124 sagittal slices, spanning from
67.6 mm Left to 80.0 mm Right. Each slice is a 256 x 256 voxel image,
with each voxel being a 0.938 mm square. The slice thickness is 1.2 mm
and the axis orientation is ASL.
The dataset was created using the AFNI program 'to3d', which has been
described in earlier how-to's (also see 'to3d -help'). Each subject's
anatomical volume is stored in their respective data directory:
ED/ EE/ EF/
ED_spgr+orig.HEAD EE_spgr+orig.HEAD EF_spgr+orig.HEAD
ED_spgr+orig.BRIK EE_spgr+orig.BRIK EF_spgr+orig.BRIK
Time-Series Data Collected:
--------------------------
Each EPI run consists of 138 volumes, 23 axial slices each. These
slices were taken from 68.3 mm Superior to 41.7 mm Inferior. Each slice
is a 64 x 64 voxel image, with a resolution of 3.75 mm on each side.
The thickness of each of the 23 slices is 5.0 mm, the axis orientation
is RAS, and the TR = 2 seconds.
The data were saved in a sequence of directories generated by the GE
scanner (003/, 023/, 043/... 623/). The AFNI program 'Imon' was used to
verify the integrity of the I-files (i.e., Are any files missing? Out of
sequence?). The '-GERT_Reco2' option was included on the command line
to create AFNI datasets for our 10 runs of data (this 'GERT_Reco2'
script is similar to the 'GERT_Reco' script found in the AFNI 'Ifile'
program. The 'Imon' command is shown below:
Imon -start_dir 003 -GERT_Reco2 -quit
An excellent description of the 'Imon' program can be found by typing
'Imon -help' on the command line.
The 10 AFNI datasets are stored in each subject's data directory:
ED/ EE/ EF/
ED_r1+orig EE_r1+orig EF_r1+orig
ED_r2+orig EE_r2+orig EF_r2+orig
. . .
. . .
ED_r10+orig EE_r10+orig EF_r10+orig
--------------
STUDY FINDINGS
--------------
Lateral Temporal Cortex:
-----------------------
The lateral temporal cortex showed strong responses to movement overall.
As such, this area of the brain seems to be the cortical locus for
processing complex visual motion.
Within the lateral temporal cortex:
----------------------------------
The superior temporal sulcus (STS) responded strongly to both
human movies and human point-light displays.
The middle temporal gyrus (MTG) and the inferior temporal sulcus
responded strongly to tool movies and tool point-light displays.
Ventral Temporal Cortex:
-----------------------
This area tends to respond better to movie displays than point-light
displays, suggesting that form, color, and texture (present in movies
but not point-light displays) are main contributors to ventral temporal
activity.
Within the Ventral Temporal Cortex:
----------------------------------
The lateral fusiform responded more to human movies than to any
other stimulus category, while the medial fusiform preferred
tool movies.
---------
REFERENCE
---------
The data used for this how-to was kindly donated by Michael Beauchamp. For a
full review of the experiment described in this how-to, see:
Beauchamp, M. S., Lee, K. E., Haxby, J. V., & Martin, A. (2003). FMRI
responses to video and point-light displays of moving humans and
manipulable objects. Journal of Cognitive Neuroscience, 15:7,
991-1001.
The data reported in this experiment have also been deposited with the fMRI Data
Center archive (http://www.fmridc.org). The ascension number is 2-2003-113QA.