This repository contains a PsychoPy implementation of a dynamic localizer, facet inspired from the paradigm by Pitcher, D., Dilks, D. D., Saxe, R. R., Triantafyllou, C., & Kanwisher, N. (2011). Differential selectivity for dynamic versus static information in face-selective cortical regions. Neuroimage, 56(4), 2356-2363.

Each run contains five categories

1. faces
2. objects
3. bodies
4. scenes
5. scrambled objects

which are presented with the following paradigm:

• fixation (18s)
• randomized blocks of categories with no inter-trial interval (18s * 5)
• fixation (18s)
• reversed order of categories as in previous block (18s * 5)
• fixation (18s)

Each block contains six video clips of about 3s each. Each run lasts 234s.

Participants perform a 1-back repetition detection on the clip. In each run there is one such repetition once for every category. By default four runs are generated.

The stimuli are not shared with this repository because I don't have the license to release them. Please contact the authors of the original paper, or use your own clips. Simply store them in the stimuli directory, with a subdirectory for each category, e.g.

stimuli
├── bodies
├── faces
├── objects
├── scenes
└── scrambled_objects

Running the script without arguments will start a dialog where you can input the participant's information. Alternatively the script can be run from the command line with the following arguments

$ python run_localizer.py -h

usage: run_localizer.py [-h] [--subject SUBJECT] [--runnr {1,2,3,4}]
                        [--no-scanner] [--no-fullscreen]

Presentation script for a face/object/scene/bodies localizer, inspired by the
paradigm in Pitcher, D., Dilks, D. D., Saxe, R. R., Triantafyllou, C., &
Kanwisher, N. (2011). Differential selectivity for dynamic versus static
information in face-selective cortical regions. Neuroimage, 56(4), 2356-2363.

optional arguments:
  -h, --help            show this help message and exit
  --subject SUBJECT, -s SUBJECT
                        subject id
  --runnr {1,2,3,4}, -r {1,2,3,4}
                        run nr
  --no-scanner          do not listen to the serial port
  --no-fullscreen       do not run in fullscreen

The script will create a logfile for each subject and run under res/sub-id/. The log contains already all the information to create a BIDS compliant events.tsv files. You just need to grep BIDS, and that's it. For example:

$ grep BIDS res/test/sub-test_task-localizer_run-1_20171102T142349.txt | awk '{for (i=3; i<NF; i++) printf $i"\t";print $NF}' | head
onset   duration        stim_type       repetition
0.000   18.000  fixation        None    0
18.000  3.000   scrambled_objects       ./stimuli/scrambled_objects/scrambled_obj_17.mp4        0
20.491  0.000   button_press    null    0
21.002  3.000   scrambled_objects       ./stimuli/scrambled_objects/scrambled_flag.mp4  0
24.002  3.000   scrambled_objects       ./stimuli/scrambled_objects/scrambled_digital_mixer.mp4 0
27.003  3.000   scrambled_objects       ./stimuli/scrambled_objects/scrambled_wind_chimes.mp4   0
30.003  3.000   scrambled_objects       ./stimuli/scrambled_objects/scrambled_candle2.mp4       0
33.003  3.000   scrambled_objects       ./stimuli/scrambled_objects/scrambled_inside_piano.mp4  0
36.004  3.000   bodies  ./stimuli/bodies/body_17.mp4    0

Note that also button presses are recorded, so that they can be added as additional nuisance regressors in the GLM.

Thanks to Sarah Herald for sharing the initial implementation of the localizer in MATLAB/Psychtoolbox.