Psy 123_Lec 6

C d the 18 million movie clips were then ranked

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ct exactly what is going on. C & D. The 18 million movie clips were then ranked according to how well their predicted HRF matched the observed HRF of the unknown movie. A cumulative average of the top 30 highest posterior clips were then used to reconstruct Goal of this is to take the hemodynamic response the movie.! while the subject is watching a movie they don't know. The question is can the observed signal be deconstructed to predict what the subject was watching. So took training subjects, and have movie clips and broke down different features of the image and this went into the encoding model, and then decoding the voxel from the different features in the clips… Went to the internet and down-loaded 18 mil clips, 0.01 s each, and for each one were able to come up with predicted BOLD signal. Then can compare the clip with what they watched. B. During training sessions, subjects viewed 7200 1-sec movie clips in the fMRI scanner. Mathematical models of the HRF were created at each voxel based on the dynamic spatial and temporal character...
View Full Document

This note was uploaded on 01/18/2014 for the course PSY 123 taught by Professor Staff during the Fall '11 term at UCSB.

Ask a homework question - tutors are online