Marimba Video Taxonomy

From KubovyLab

Characters 1 & 2) signify pitch level (ascending). 1 E is the lowest, 3 A the next, 5 D next, and 7 G the highest. The numbers are just there so that they get sorted in the appropraite order.

Character 3) Visual stroketype. S="short", L="long", N="normal", X= none (e.g. audio-alone token)

Character 4) Auditory stroketype. S="short", L="long", N="normal", P = "damPed", X= none (e.g. video-alone token)

So, a sample token of "1 E L L " is the lowest pitch level (E) with a long visual and long auditory stroke. A token of "5 D S P" is for the pitch level D with short visual and damped audio.

---

DEPENDENT VARIABLES

duration (0-100) the rated duration of the note (subjects are asked to base response on audio alone)

agreement (0-100) rating of how well the audio "agrees" with the video. If the stimuli "looks long" and "sounds long" then there is high agreement, if it "looks short" but "sounds long" there is low agreement.

INDEPENDENT VARIABLES

pitch (e,d,g) Pitch level of the marimba note

vis (s,l, x) Visual component of stroke type (see translation of coding scheme above)

aud (s,l,p x) Auditory component of stroke type (see translation of coding scheme above)

segment (w, f, b) Segment of video presented. This is generally w (whole) but in some experients we split the videos to create f (front) and b (back) options

Offset coding scheme:

     -21 = audio occurs 21 frames early
     -12 = audio gesture occurs 12 frames early
      0  = audio gesture occurs 0 frames early (e.g. a "normal" presentation)
      12 = audio gesture occurs 12 frames late
      21 = audio gesture occurs 21 frames late

OTHER VARIABLES (not generally important)

subj (1-n) Subject number

session (1-2) Session number

number (1-n) Trial number

block (1-n) Block number of a given set of trials

superBlock (1-n) Block of blocks. Used to control order and space out stimuli evenly

description (text) Text description of file name used for debugging purposes only