NBE-E4250 - Mapping, Decoding and Modeling the Human Brain D, Lecture, 10.1.2023-22.2.2023

The goals of this week’s project work are 

• to continue from the last week's task (analyze fMRI data using general linear model)
• to make contrast images (e.g., find voxels with larger response to faces vs. houses), and
• to do region-of-interest analysis on the data.

(1) Finalize your design matrices and GLM fits (last week’s project). 

(2) Can you find voxels that show a larger response for faces compared to houses? Can you find voxels that show the opposite preference? What about the other comparisons, e.g., scissors vs. cats? What conclusions can you make based on the results? [See slides of Lecture 2 for hints on how to contrast two stimulus categories.]

(3) For the region-of-interest (ROI) analysis, you have in the Haxby dataset the mask*.nii.gz images (“Various masks in functional space provided by the original authors. "vt" refers to "ventral temporal", "face" and "house" masks are GLM contrast based localizer maps”). Load these to matlab and use as regions-of-interest. In addition, I suggest  that you make also “regions-of-interest” in a few random locations within and outside the brain to compare the results between the real and random regions-of-interest. When you have identified the voxels corresponding to the regions-of-interests: (i) Plot the average time-courses within the ROIs [Can you see the onsets and offsets of the stimulus blocks? Can you see from the raw data what stimulus evoked the largest response?], and (ii) Plot the average beta estimates as bar plots [Which stimulus evokes the largest response? Is this different for different ROIs?].

Start with the data of one subject and when you have a script ready, use the same analysis pipeline for the data of the other subjects.

Return the first version of your project work by Tuesday, January 24, at 10:00. Even if you have had trouble with the analysis, do return something that we can then discuss during the Tuesday's Q&A session. You will get 5 points for returning your preliminary results and actively contributing during the Q&A session.