Dr Michael Graziano’s puzzling graphs?

I recently watched part of a presentation on Youtube by Dr Michael Graziano concerning his ideas about consciousness (below). In the video at 8:07 he shows a graph which I find interesting. You’ll need to watch from the start of the video to understand this experiment though.

On the graph within the video… …you can see at the top, three fMRI plots of aggregated trials, which show qualitative changes in blood oxygenation on the misaligned, ‘hard’ tasks. Graziano says these results imply increased brain activity for ‘hard’ tasks.

The interesting point for me is the bottom line, which shows the plot of aggregated trials on the aligned, ‘easy’ tasks. This too shows qualitative changes in blood oxygenation. Graziano says this implies decreased activity for ‘easy’ tasks, compared to the ‘hard’ tasks. OK, that sound reasonable, but why does the plot fall way  below the baseline? My assumption would be that Gaziano should interpret these results as indicating a reduction in activity for the ‘easy’ task compared to normal activity, not just a reduction compared to ‘hard’ tasks?

You can look at his actual paper here (which I’ve not read). The same data on the graph within the video is shown at Fig.4 in the paper, I’ve copied this below with some small alterations for clarity.

grazianoWhat seems clear is the aligned ‘Easy’ task trials seem to show a reduction in the fMRI signal, compared to baseline ‘normal’ activity.

Bear in mind that fMRI only shows qualitative changes – that is comparisons between one time period, compared to another. What the actual existing level of activity was before the trials is unknown, but the ‘Easy’ trial measurement indicate that normal baseline activity was somewhat higher, otherwise they should not have measured any reduction.

I have some issues with fMRI studies anyway, as to what interpretation should be attached to these measurements. I’m also not convinced by Graziano’s attempt to explain this reduction in his video presentation, I think it’s strange, and probably significant.

I’m now going to come back to the graph shown in the video, and compare it to the one provided in the published paper. The quality of the video graph is very poor, and I’ve adjusted both graphs so that their scales apparently match.

graziano2 Graziano has now confirmed that the video graph was preliminary data from 80 trials conducted with just one subject, whereas the graph in the paper contains data from all 50 subjects. Presumably this is not plotted from preliminary data, and thus has been adjusted.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s