tag:blogger.com,1999:blog-21605329.post7080170989322123770..comments2024-03-14T23:52:09.893-07:00Comments on The Neurocritic: Should Policy Makers and Financial Institutions Have Access to Billions of Brain Scans?The Neurocritichttp://www.blogger.com/profile/08010555869208208621noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-21605329.post-25338390717901026712014-10-06T20:01:52.527-07:002014-10-06T20:01:52.527-07:00Dr. Gilaie-Dotan - Thank you for taking the time t...Dr. Gilaie-Dotan - Thank you for taking the time to explain. As they say, "more research is needed."The Neurocritichttps://www.blogger.com/profile/08010555869208208621noreply@blogger.comtag:blogger.com,1999:blog-21605329.post-15952119536201004912014-10-01T02:39:19.564-07:002014-10-01T02:39:19.564-07:00Thank you Neurocritic for adding the full figure l...Thank you Neurocritic for adding the full figure legends and pointing to my comment. <br /><br />And to your question - yes, we were expecting to find significant results in orbitofrontal/medial prefrontal cortex following the clear indications that exist from functional studies that the activation in this/these regions is associated with individual differences in risk attitudes. However, it is not always the case that functional and anatomical individual differences are co-localized in the brain. So co-localization could be function dependent. One possiblity is that when a network of regions is associated with some cognitive function (as is the case for most functions), then restriction or expansion of the anatomy in one subregion can affect the activation in the whole network (sort of like an anatomical bottleneck or the "weak link" of the net), and thus can eventually affect the outcome of the whole computation. So the computation of that function in individuals who have such anatomical "narrowing" can be significantly affected, but the activation differences might not be observed in that "anatomically restricted" area but in other regions in the network. <br /><br />Of course there could be many other explanations, this is just one alternative from many. Sharon Gilaie-Dotannoreply@blogger.comtag:blogger.com,1999:blog-21605329.post-91833250961230606292014-09-28T10:23:29.046-07:002014-09-28T10:23:29.046-07:00Dr. Gilaie-Dotan - Thank you for your comment that...Dr. Gilaie-Dotan - Thank you for your comment that clarifies the methods. I have added an addendum to the post pointing the readers to your comment, as well as putting in the legend to Fig. 2 (Bottom), which should not have been omitted from the original post.<br /><br />One question about your last statement: since the parietal cortex result was not surprising, were you at all surprised by the absence of an effect in the orbitofrontal cortex?The Neurocritichttps://www.blogger.com/profile/08010555869208208621noreply@blogger.comtag:blogger.com,1999:blog-21605329.post-1807238936086516402014-09-28T03:42:47.403-07:002014-09-28T03:42:47.403-07:00I am sorry if any information was misleading to pr...I am sorry if any information was misleading to provide an impression of voodoo correlations in our paper. This is definitely not the case and all the analyses we carried and that are provided in the paper are very cautious about and sensitive to circular inference. <br /><br />Since the full figure legends are missing from this blog, and all the relevant explanations are provided in the full text, let me fill you in on those:<br /><br />* Figure 2's legend says:<br />"Bottom, To demonstrate that the observed correlations were not driven by outliers, for each individual, GM volume of the PPC cluster (top) is plotted on the x-axis against risk attitude on the y-axis. <b>Note that this should not be used for inference as it is not independent of the whole-brain analysis and is presented for visualization purposes only.</b>" <br />I hope this clarifies the issue raised about Figure 2.<br /><br />* Figure 4 legend explains:<br />"Risk tolerance (alpha ) was estimated for each individual separately using the same procedure as in Study 1 (see Materials and Methods).<b> Importantly, the risk estimates are based on individual’s behavior only, and the GM volume was sampled independently using MarsBar toolbox for SPM.</b>" <br /><br />So note that the 2 measures in the scatter plot are independent of each other. <br /><br />With respect to your additional comments, we provide in Figure 4C the parametric relationship between financial decision making tendencies (as measured in our experiment, y axis), and brain structure in the right parietal cortex (R-PC) focus (parametically changing, x axis). The data (all from Study 2) are clear in showing that the gray matter of 15 and 20mm spheres around the center of the R-PC focus provide the best and significant prediction for behavior. This is not the case for the 10mm or the 25mm. There was no rational in choosing 15mm or 20mm spheres, it is what the data revealed. Since these results presented in Figure 4 (and in the Tables) are a replication with an independent second data set in our study, I don't second your conclusion. <br /><br />Even though the parietal cortex is not surprising with respect to decision making, as with any scientific finding, only future studies will be able to provide indications related to generalization of our finding to other populations and other decision related attitudes. Sharon Gilaie-Dotannoreply@blogger.comtag:blogger.com,1999:blog-21605329.post-82950916128820713892014-09-17T16:48:45.726-07:002014-09-17T16:48:45.726-07:00Both Figure 2 and (in my opinion) Figure 4C are vo...Both Figure 2 and (in my opinion) Figure 4C are voodoo correlations. I am not sure I would find any rational in selecting 15mm or 20mm as oppose to 10mm or 25 mm in Figure 4. The Z-scores just skim above 2 for the second data set. It is not much. <br /><br />Given the surprising area the researchers identify I would be cautions on any conclusions.Finn Årup Nielsenhttp://www.compute.dtu.dk/~faan/noreply@blogger.comtag:blogger.com,1999:blog-21605329.post-70094049706612411472014-09-16T10:36:42.536-07:002014-09-16T10:36:42.536-07:00I regard the unambiguous "risk tolerance"...I regard the unambiguous "risk tolerance" assessment as a simple matter of mathematics. A 50% chance of winning $5.00 means the average payout is $2.50. A 38% chance of winning $18 means an average payout of - well, I'm too lazy to do the math when it's clearly not even close, but certainly a lot more than $2.50. Given that the chances of winning something aren't drastically different (as in the case of a lottery with rare big payouts), the second choice is obviously better and the less you involve your feelings in it the better.Anonymousnoreply@blogger.com