Meeting 04-02-2022

[gjheij]

In this meeting, we discussed preliminary analyses following the scanning session on 01-02-2022, where we performed 1 3D-EPI run, 2 pRF-line runs, and 3 Size-Response (SR) runs. Overall, the session was pretty successful, with here and there opportunities to improve:

pRF-experiment:

• longer breaks in between sweeps > now "only" 20 seconds. Make this maybe 30-40 seconds?
• bar should span more of the visual field. The timecourse suggests that the bars never really leave the pRF. Now it extends -3*bar_width to 3*bar_width around the center of the pRF. I'll experiment with increasing the span of the bar sweep by creating design matrices for each alteration and observe the predicted response of the target pRF.

SR-experiment:

• Instead of diving straight into deconvolution, do a GLM first to see what to expect from the HRF. Generally, we wouldn't expect differences in HRF-shape (maybe amplitude) in response to different stimulus sizes. We would expect, however, to see changes in HRF-shape across depth. We will have a clearer understanding of what to expect of these HRF-changes once we've explored the GLM with a canonical HRF

So, to do:

• Adapt pRF-experiment; test which type of design results in more pronounced responses
• Do GLM of SR-experiment

[gjheij]

specify stimulation-settings file to play with duration in between sweeps and span across visual field. Currently have screenshots for*:

• sim1: 30s baseline (-); span [-3,3] (-); rest duration 30s (+10s); no inter-sweep blank (-)
• sim2: 30s baseline (-); span [-4,4] (+1); rest duration 20s (-); no inter-sweep blank (-)
• sim3: 30s baseline (-); span [-5,5] (+2); rest duration 20s (-); no inter-sweep blank (-)
• sim4: 30s baseline (-); span [-5,5] (+2); rest duration 30s (+10); no inter-sweep blank (-)
• sim5: 30s baseline (-); span [-5,5] (+2); rest duration 30s (+10); 3s inter-sweep blank (+3)
• sim6: 30s baseline (-); span [-5,5] (+2); rest duration 20s (-); 5s inter-sweep blank (+5)
• sim7: 20s baseline (-10); span [-5,5] (+2); rest duration 15s (-5); 10s inter-sweep blank (+10)
• sim8: 20s baseline (-10); span [-7,7] (+4); rest duration 15s (-5); 15s inter-sweep blank (+15)

These will form the foundations for the simulations.

*(-) means same as we're currently using

[gjheij]

pRF-analysis:

• savitzky-golay low-pass filter; 5 (poly 2) tot 10 TRs (poly 3) windowlength
• quantify removed frequencies
• simulate pRF-experiment with blanks inbetween sweeps too

SR-analysis:

• 3 runs, t-stat/percent-signal change > plot SR-curve (for each run)

[gjheij]

See this pdf for the predictions for each of the above specified design matrices