tag:blogger.com,1999:blog-21605329.post8252405956832353897..comments2024-03-22T00:30:09.536-07:00Comments on The Neurocritic: Amygdala Stimulation in the Absence of Emotional Experience Enhances Memory for Neutral ObjectsThe Neurocritichttp://www.blogger.com/profile/08010555869208208621noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-21605329.post-57063852745081756902017-12-24T04:40:44.212-08:002017-12-24T04:40:44.212-08:00Neurocritic,
Been following your blog for quite a...Neurocritic,<br /><br />Been following your blog for quite a while too, and I've been waiting for you to take a look at the CurrBiol paper and then you leapfrog straight to Cory? So disappointed ;)Paul W.noreply@blogger.comtag:blogger.com,1999:blog-21605329.post-12822188170622233242017-12-24T01:38:21.660-08:002017-12-24T01:38:21.660-08:00Cory -- Thanks for your comments. The effect size ...Cory -- Thanks for your comments. The effect size for the one-day test was certainly large. For assessing the difference between the test delays, I was thinking about these papers: <a href="https://www.nature.com/articles/nn.2886" rel="nofollow"><i>Erroneous analyses of interactions in neuroscience: a problem of significance</i></a> and <a href="http://www.tandfonline.com/doi/abs/10.1198/000313006X152649" rel="nofollow"><i>The Difference Between “Significant” and “Not Significant” is not Itself Statistically Significant</i></a>.<br /><br />The RAM project is definitely ambitious (as one would expect with DARPA). I haven't yet read the 2017 paper in <i>Current Biology</i>, so my year-old post (<a href="http://neurocritic.blogspot.com/2016/12/bad-news-for-darpas-ram-program.html" rel="nofollow">Bad news for DARPA's RAM program: Electrical Stimulation of Entorhinal Region Impairs Memory</a>) is out of date now.<br /><br />For the neuropsych data, I did notice the large variation in memory and especially IQ. I'll take a closer look at that.The Neurocritichttps://www.blogger.com/profile/08010555869208208621noreply@blogger.comtag:blogger.com,1999:blog-21605329.post-71332144935085936722017-12-23T19:05:16.711-08:002017-12-23T19:05:16.711-08:00Sorry for another comment, but one other point is ...Sorry for another comment, but one other point is that if we control for the baseline variation in memory capacity (as measured by the Rey Auditory Verbal Learning test or other standard neuropsyschological tests of memory capacity) in our repeated measures ANOVA, we do see an interaction of the Stimulation condition and the Test Delay, with stimulation showing enhanced memory relative to no stimulation on the one-day test. This interaction is obscured by the broad variation in baseline memory capacity across our sample, thus we have to covary out that variability. Anonymoushttps://www.blogger.com/profile/15674363610221457809noreply@blogger.comtag:blogger.com,1999:blog-21605329.post-60254518866686153732017-12-23T18:52:51.193-08:002017-12-23T18:52:51.193-08:00Also, thank you for making the point that a sample...Also, thank you for making the point that a sample of 14 is a lot for this rare patient population. I also work on the current DARPA Restoring Active Memory project and through that project we've produced some large sample sizes (>100 patients) and very interesting data, but that is certainly the exception rather than the rule given that that project combines data across 8 epilepsy centers and has cost around 30 million dollars over the past 4 years. Thanks again!Anonymoushttps://www.blogger.com/profile/15674363610221457809noreply@blogger.comtag:blogger.com,1999:blog-21605329.post-52023471971209317962017-12-23T16:51:14.562-08:002017-12-23T16:51:14.562-08:00Hi Neurocritic! Thank you very much for reviewing ...Hi Neurocritic! Thank you very much for reviewing our work. I've been a fan of your blog for a long time! In our paper we did both a repeated measures ANOVA to determine whether there was an omnibus effect. Our main statistical test, as done in the previous rat studies, was a planned paired t-test between the stim and no stim conditions for each of test delays (immediate and one-day test). In your opinion, is the planned t-test between each condition during the one-day test sufficient to show this memory enhancement effect specifically for the one-day test? It's a very strong effect according to Cohen's D. Also since we have a relatively small N, we also ran the Non-parametric version of the paired t-test, the Wilcoxon Sign-Rank test with monte carlo simulation (10000 iterations). We didn't report this test in the paper to keep it simple since the paired t-test was sufficient, but it also showed a significant difference between stim and no stim on the one-day test (p<0.003). I'm happy to discuss this more and would appreciate your opinion. Thanks again for reviewing our work on your blog!Anonymoushttps://www.blogger.com/profile/15674363610221457809noreply@blogger.com