2021-05-14 03:17:22 @Raamana_ fingers crossed for you! 2021-05-14 00:35:42 @JhendersonIMB Just use your hands, it's nature's gasoline cup! 2021-05-14 00:21:39 RT @hakwanlau: the official ad is out https://t.co/a5vtGjIg3F pls RT 2021-05-13 17:00:53 @graemedmoffat @tyrell_turing Hah, fat chance. We are super good at this. 2021-05-13 16:55:04 @tyrell_turing @graemedmoffat I think the variance is on how the fitting-in pressure manifests But it's always there, a product of our hardwired in-group morality. 2021-05-13 15:50:51 @graemedmoffat @tyrell_turing Mood = mode.. 2021-05-13 15:50:16 @graemedmoffat @tyrell_turing Yeah I think only the mood of expression has changed. In fact it was perhaps easier for men to fit in when everybody just wore suits and ties. Contrast the difficulty of renting tuxes to choosing bridesmaid dresses at a wedding. 2021-05-13 15:33:30 @tyrell_turing @graemedmoffat Do you seriously think that is not true now? 2021-05-13 15:30:42 It's great that pharmacies have decided for us that date of birth is a secure password for picking up prescriptions. 2021-05-13 14:07:48 @rudzinskimaciej @AcademicEEG @IntuitMachine Thanks Maciej, here's a thread on the paper: https://t.co/En68zS86LI And the model is discussed in about the last third of this WWN talk: https://t.co/5e4iwdG7V5 2021-05-13 12:31:17 RT @bradleyvoytek: Wow this new @biorxivpreprint is compelling. Like we've been saying with EEG / LFP oscillations, it argues that apparent… 2021-05-13 11:25:49 RT @CathrynTownsend: Listen to Richard Dawkins flounder on eugenics. 2021-05-13 04:37:59 @rudzinskimaciej @AcademicEEG @IntuitMachine Thanks, I'll watch it! 2021-05-12 22:57:13 RT @olivermb: Proud of my faculty senate today at Penn State. They passed 2 resolutions: a) to mandate vaccines 2021-05-12 19:49:22 RT @TaraViegen: My first first author paper (with equal con.) is out and freely accessible until 30th of June: https://t.co/BNB8SjK6a3 Tha… 2021-05-12 14:03:48 RT @syndrome22q1: I get sometimes people calling me names making fun of me because of my disability it makes me feel sad sometimes and ma… 2021-05-12 13:30:28 @aeryn_thrace @neurograce @ceptional I'm not saying we shouldn't document our work. I'm saying that we shouldn't have the same federal publication guidelines apply to every possible kind of scientific work. We need degrees of freedom to accommodate the many varieties of endeavors 2021-05-12 13:00:39 @aeryn_thrace @neurograce @ceptional Agreed! And I don't think that a blanket recommendation by the GAO about how all research should be published is going to push us in that direction. Some work is very exploratory and doesn't need to be as carefully processed, there should be a diversity of publication pipelines. 2021-05-12 04:43:21 @tage_rai So glad that you've been doing such good advocacy as an editor. 2021-05-12 04:40:44 RT @NunezKant: Paisanos, como le hacemos para subir a México en esta lista ? Quieren aprender neurociencias computacionales y deep learn… 2021-05-12 04:37:57 RT @SBMost: Very sad news. A loss to psychology and of someone who had a huge impact on me. Jerry took a chance on me, first as an inexperi… 2021-05-12 01:24:19 RT @CharlieLudowici: My PI is looking for a postdoc with expertise in at least two of the following: eye movements, psychophysics, retinal… 2021-05-12 00:53:26 @AndyPerfors Is that a done deal? 2021-05-11 23:17:55 @neurograce @ceptional Yes! and also sustainable funding for development of tools and infrastructure (that aren't 10 million $ center grants) 2021-05-11 23:05:48 @ceptional That is terrifying TBH. The last thing we need is the GAO providing additional regulations on how science should be published because they are likely to overreach. If anyone does get involved here, please don't push hard for big changes. 2021-05-11 20:52:27 @GunnarBlohm @queensu I think universities are concerned that parents and students will tolerate a 1 year virtual experience but two is pushing it. 2021-05-11 19:09:56 RT @KordingLab: Incidentally, Jon, with his industry background, convinced me that we need to make sure students see deployment. We are wor… 2021-05-11 18:10:19 @neuralreckoning Ms paint 2021-05-11 16:55:52 @itjohnstone @samuelmehr @wgervais @PsychScientists @OHBM Yeah the poster sessions at SFN with the same deal for me 2021-05-11 14:41:06 @PsychScientists @samuelmehr @wgervais In my experience the small ones start off inclusive but then end becoming exclusive. That may not be generally true though 2021-05-11 14:37:51 @samuelmehr @wgervais @PsychScientists Yea, for sure some areas won't be well represented at all big conferences, but I also think that a lot of the important career development work happens off stage at big conferences (just like at small conferences where networking also happens over beers) 2021-05-11 14:05:33 @samuelmehr @wgervais @PsychScientists The small invite lists these events are often reserved for people who are on on an inside track either because of prestige or who they know. They often seem like much more of a status clique than what big conferences offer. 2021-05-11 14:01:58 @samuelmehr @wgervais @PsychScientists Let's remember that the small conferences don't work for everybody. They're often very exclusive and don't leave room for the many grads and ECRs that use big conferences to make their own path. 2021-05-11 12:30:07 @IrisVanRooij Yes, that was my first thought. This is related to failure of conditional reasoning 2021-05-11 11:30:41 RT @neuro_data: #LearningSalon this week has a special treat! @DLBarack will be presenting "Two views on the cognitive brain", a paper co-a… 2021-05-10 22:20:30 @lukesjulson @Apple No, 300 ms please 2021-05-10 22:19:52 RT @tage_rai: Announcement! Job Opportunity! Please RT!! I’m leaving my post as editor for social and behavioral sciences @ScienceMagazine… 2021-05-10 19:04:57 RT @meganakpeters: this will be a super amazing program — very pleased to be working on organizing this. apply to lead a GAC workshop, or… 2021-05-10 13:34:03 RT @katestorrs: Jumping off from this incredibly generous tweet (...please lower your expectations before reading the paper), I do think we… 2021-05-10 13:21:22 RT @amazngspidermom: Now that Mother’s Day is over, I guess it’s time to get back to my futile attempts to simultaneously hide my children… 2021-05-10 11:49:38 This massive thread is a historical perspective on the science of infectious diseases and illustrates how the CDC became confused about aerosol transmission years ago and that error is still affecting the worldwide response to COVID. Fascinating and scary. Buckle up! https://t.co/4Wvu3XOU0U 2021-05-10 00:10:29 RT @SBMost: My best #AcademicTwitter tip for students & Twitter's great for telling others about your work, but self-promotion feels… 2021-05-09 14:58:16 RT @ashishkjha: India reports another 400,000+ cases, 4000+ death day A sustained level of horribleness And its not correct True number… 2021-05-09 12:09:25 @todd_gureckis I think the most accessible mystery is how perception is constructed, in particular size constancy illusions. A close second would be how the brain manages to solve problems. 2021-05-09 04:36:02 @DCameron84 Awesome! Was very glad ( and unsurprised ) to hear this 2021-05-09 04:32:10 RT @meganakpeters: we have extended the #NeuromatchAcademy application deadline! one more whole week to submit your application to be an in… 2021-05-08 21:17:44 @hannah_merseal Thanks for this timely and informative source of anxiety 2021-05-08 19:08:46 Fives of protestors have turned out for these rallies. Truly a corner has been turned. https://t.co/51HgCIunZI 2021-05-07 17:32:07 So glad for this bit of good news for my wonderful colleague. https://t.co/4rOU3qVbgH 2021-05-07 12:28:55 @andpru But this time we'll get it right 2021-05-07 12:05:04 @KordingLab @skornblith @jgyanar @danilobzdok @criticalneuro @tyrell_turing @neuro_data And a very weird human at that. 2021-05-07 04:27:51 @leavittron @tdverstynen @spiantado Yea definitely sarcasm. 2021-05-07 04:24:57 @Dirk67985531 @DrGBuckingham @LindaKKaye @TimKietzmann There's a horizontal wall of stone that is in front of her 2021-05-07 00:03:32 @DrGBuckingham @LindaKKaye @TimKietzmann Ahhh. Nice one 2021-05-06 22:17:56 @jslbutler @ebellm That is amateur hour 2021-05-06 15:52:20 @ira_hyman @JhendersonIMB @SBMost @fchollet @PessoaBrain @o_guest And fire them. 2021-05-06 15:25:37 @JhendersonIMB @ira_hyman @SBMost @fchollet @PessoaBrain @o_guest But I don't think they do see the light. This is why companies keep claiming that they can decode emotions, or optimize worker performance, etc. 2021-05-06 11:58:18 RT @venpopov: "Dear grant review panel, I don't know who you are. I don't know what you want. I can tell you I don't have money but what I… 2021-05-06 00:38:53 @SBMost @fchollet @PessoaBrain @o_guest @JhendersonIMB It's not really about who gets credit, it's about building links between disciplines. And I agree a reaction from you would be different because you have a very different audience than @fchollet. 2021-05-06 00:36:28 @fchollet @SBMost @PessoaBrain In my view, many in AI (even if not you) don't have a good sense of perspective on how these rapidly emerging ideas and successes have built on the tapestry of other sciences. This is ultimately counterproductive because it leads AI to think it can solve all the problems 2021-05-06 00:33:52 @fchollet @SBMost @PessoaBrain For me, the reaction is related to the fact that your audience, largely AI folks who admire you (as do I), would not see in your tweet a link back to the rich traditions of psychology that provided the foundation of this idea. Why is this a problem? > 2021-05-05 20:14:47 @JhendersonIMB @PessoaBrain @fchollet @SBMost Yes, definitely true that attention is many things. But I think the key aspect underlying the basic cognitive concept of attention is the ability to alter how processing occurs based on an interaction of stimulus input and intrinsic information/goals/biases/etc. 2021-05-05 18:09:34 @sarabwtroutman You will be missed! 2021-05-05 13:03:18 @fchollet @SBMost @PessoaBrain Solving that problem would be a massive step forward for image recognition. 2021-05-05 13:02:52 @fchollet @SBMost @PessoaBrain Regarding active perception in AI, I think it's been ignored because it's (currently) inconsistent with batch processing & 2021-05-05 13:00:53 @fchollet @SBMost @PessoaBrain Attention allows transformers to change the way that they interpret part of the image based on other aspects of the pipeline/image. This is exactly what cognitive visual attention does: gives us "on-the-fly" flexibility in how we link parts together to form interpretations. 2021-05-05 11:51:13 @fchollet @SBMost @PessoaBrain Moreover, building bridges between AI and cognitive is going to be crucial for making real progress towards something like AGI but it is rare that people on the AI side make genuine efforts to reach across the divide. Therefore I should have reacted better myself. 2021-05-05 11:38:42 @fchollet @SBMost @PessoaBrain Psychology has had to defend itself as a legitimate science for a long time and it is exhausting. It is also common for basic insights from our field to be reinvented or restated without attribution. Is not fair to expect credit? 2021-05-05 00:50:41 @SBMost @PessoaBrain This is the smoothest "hey, everyone over in psychology already knew this" I've seen in awhile. 2021-05-04 21:20:46 This is my favorite conclusion from Iris's recent paper. It means that we should stop trying to prescribe how others do science. https://t.co/PKSXCR27Ps 2021-05-04 03:18:17 RT @AriBerman: Bruce Bartman, white guy in PA who illegally voted for Trump on behalf of his dead mother, gets 5 years probation Crystal M… 2021-05-03 15:10:11 @neurograce I think the argument for gravity is pretty weak as it requires that there is no sensitivity to luminance of the full moon, which is 1000x brighter than starlight. They base this assertion on replication within Seattle students, but even in a city, moonlight is perceptible. 2021-05-03 13:01:42 @stevejohnryan @KordingLab @neurograce That's a voice to text error, I can't take credit 2021-05-03 12:37:35 @KordingLab @neurograce Yeah this effect as being caused by gravity is only believable for the sunset of people who live exclusively where there's massive light pollution. I haven't read it yet, but the gravity hypothesis all hinges on that statistic. 2021-05-03 11:33:25 @GunnarBlohm @phant0msp1k3 @neuromatch @meganakpeters Yep, you can be considered for both roles at the same time. 2021-05-02 19:58:58 @PessoaBrain @neuralreckoning I agree with Dan 2021-05-02 13:51:38 @timmytimmytim What happens in slack, stays in slack 2021-05-02 10:46:32 @timmytimmytim You should see what I say to my lab when they mention this. 2021-05-01 16:07:18 RT @SabiVM: Many people have asked me how & 2021-05-01 14:45:19 @o_guest WOW I'm especially impressed by the mattress, which I assume isn't in two parts 2021-05-01 12:43:38 @o_guest Omg, How did you get it in there?? 2021-05-01 12:36:18 RT @nschawor: eastern european friends, what are you doing this summer? one possibility: these 2 excellent online summer schools! in all… 2021-05-01 11:38:47 @roydanroy I think they're talking about the vaccines immunity weakening after 6 months 2021-05-01 03:53:28 @roydanroy Natural immunity is probably weakening by now if it was November. 2021-05-01 03:43:30 @zerdeve Omg pixels photo system is such a dream. I have not ever run out of space and yet I can scroll back to any picture or video I have taken in the last 5 years 2021-04-30 23:14:57 RT @mingyussong: Being a TA last year was one of the best things I did during this pandemic. Learned so much from interacting with my stude… 2021-04-30 15:56:47 Neuromatch will be paying TAs USD $1500 for both of these courses. If you're a programmer, you're probably qualified to serve as a TA one of them. The lectures and codebooks are provided, you just need to guide the students in the learning process. It was loads of fun last year. https://t.co/BuFThc33wO 2021-04-30 13:27:40 Does Rudy know that Apple's servers are not literally up in the clouds. I guess he thinks they're suspended from blimps? Or maybe he thinks that Helecarriers are a real thing. https://t.co/JtvkoOORe1 2021-04-30 12:35:47 @mark_riedl It would be easier for Tesla if everyone's eyes were surgically replaced with GPS and LIDAR. Why don't we just do this?? 2021-04-30 12:06:28 Counterpoint: Cognitive Vision is a good tool for navigation at fast speeds and you haven't solved it yet. Roads are designed for vision because vision has evolved for navigation (among other things). https://t.co/2SOE6jdqtK 2021-04-29 23:27:10 @DrYohanJohn Elon Musk. 2021-04-29 21:42:35 RT @FT__Dan: And yes, the decision to yank J& Daily vaccinations were rapidly growing for every demographic un… 2021-04-29 20:44:17 @zerdeve @IrisVanRooij I also self censor more often these days, but the Twitter barrage must be so much harder in your shoes 2021-04-29 19:34:56 Like many of my colleagues at PSU, I'm concerned about the lack of faculty representation and oversight on the presidential search process. If you're a member of the Penn State community please sign this petition in support of a better search. https://t.co/pqMk3Nyc2n via @Change 2021-04-29 15:13:51 @fedeadolfi One modality at a time! (thanks, I may do that, and I'll let you know if I do) Consider sending this to Michael Bach, he runs a great website on illusions and is always looking for cool demos. https://t.co/chajppBeKK 2021-04-29 15:02:19 @fedeadolfi Your illusion has been acknowledged -Vision science (seriously though, this is really cool, and I'm tempted to ask you for permission to put it into a grant I'm writing) 2021-04-29 12:12:52 RT @PamKreeger: @ScienceMagazine - I fixed your crappy article title. You are welcome. https://t.co/SkYsTloLNd 2021-04-29 11:27:27 RT @fedeadolfi: Flip back and forth between these two photos. They were taken 1 day apart at different times of the day from my living ro… 2021-04-29 11:25:44 RT @m_wall: A colleague sent me this - she is seething, and rightly so. I am too. Arrant fucking nonsense, and extremely sexist. https://t.… 2021-04-29 04:46:37 RT @tdverstynen: @primalpoly @tage_rai @ScienceMagazine You’d prefer that it be run by people who use BMI as an evaluation criteria? https:… 2021-04-29 00:25:14 RT @DirkBWalther: Satellite Event on Teaching Vision @VSSMtg with Jessica Witt, @MGreenePhD , @bjbalas and @m_the_cohen! May 24th 4:15 pm… 2021-04-29 00:24:36 RT @meganakpeters: attn: cogneuro tweeps! still looking for a postdoc, w/ me, @UCR_LANIlab, Aaron Seitz, & got fMRI experien… 2021-04-28 20:09:16 @danielgott @KevinZollman Is that a realistic portrait of academics? I suspect the modal academic is accepting < 2021-04-28 18:39:47 @PsychScientists A massive religious ceremony from February to April. 2021-04-28 18:19:42 RT @neuro_data: #LearningSalon (w/me, @criticalneuro & 2021-04-28 13:29:37 RT @KordingLab: Another @neuromatch DL day lead is He He from NYU. She builds cool NLP models. I really like her recent paper on overcoming… 2021-04-27 19:07:53 @talyarkoni Are you subtweeting yourself here? 2021-04-27 17:33:17 @b_d_evans @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data hah, yea, that I am :) 2021-04-27 17:29:12 @b_d_evans @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data Like this? https://t.co/UYYNcqbT8X 2021-04-27 17:07:55 @neuralreckoning @b_d_evans @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data I think we just have different definitions and what constitutes a specific task. If we zoom in far enough, my point is correct if we zoom out far enough your point is correct 2021-04-27 17:04:33 @b_d_evans @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data Believe me, I am the first to sputter with indignation when Ng talks about vision being a solved problem. I am right there with you on the importance of bio vision and how poorly DNNs measure up to the general problem. 2021-04-27 17:03:28 @b_d_evans @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data It depends on what you mean by a specific task. I don't think "driving" or "image classification" is sufficiently specific. But if you mean a specific benchmark on a specific set of limited data, I think engineering will always beat people, that's what it's good at. 2021-04-27 17:01:38 @b_d_evans @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data @TonyZador Right, I agree with all of this. My entire research program is based on how restricting information in smart ways can lead to better learning and better generalization. Also, bio algorithms are better suited to out-of-sample generalization. 2021-04-27 16:55:42 @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data @b_d_evans Evolution has found a great solution for the general problem of vision (or driving or language), yes I agree. But if you're talking about a highly specific task, engineering will beat us every time. And yes, Bio is a great source of inspiration. Inspiration != implementation 2021-04-27 16:52:47 @b_d_evans @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data What examples do you have in mind? 2021-04-27 16:51:30 @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data No that's exactly the point. On any specific task with a specific benchmark, neuroscience is going to lose vs engineering. Has always been this way. Bio wins in the general case. That's why I study Bio. 2021-04-27 16:23:47 @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data So I guess what I mean is that if you started to try to make convnets more like biological vision in various ways I think you'd be paying a price in image classification performance 2021-04-27 16:22:11 @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data The initial idea of convolution may have been bio inspired put the actual implementation in 2021 is not constrained by specific neuroscience principles. For example we don't find feature dictionaries iterated across IT cortex. 2021-04-27 14:19:42 @neuralreckoning @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data Hey c'mon now :) I'm a computational neuroscientist so obviously I believe that the brain is a great source of ideas. But the question was about whether ANNs will do better on a specific task with or without the additional constraint of biological features. 2021-04-27 14:16:26 RT @AlonZivony: This interaction (read the whole thing) is the best thing I saw on academic twitter today https://t.co/pmbfCOz2vu 2021-04-27 13:24:21 @neurograce @AlGoulas @KordingLab @KriegeskorteLab @skornblith @neuro_data Yes, definitely interesting! But I think any such list would be short lived. Any algo that showed up on that list could be quickly improved on within the context of a specific task by removing the constraints of realism. 2021-04-27 12:45:14 @AlGoulas @KordingLab @neurograce @KriegeskorteLab @skornblith @neuro_data I don't think this is a fair standard to expect realistic ANNs to do better on a given task. A specialized ANN that is tailored to the task and away from biology is always going to beat an network that has extra constraints of realism. 2021-04-27 12:31:32 RT @tdverstynen: When I was younger (and poorer), I was always baffled at how efficient wealthier people seemed to be with their time. I me… 2021-04-27 11:26:06 RT @databoydg: Super proud of @lynetcha and everyone at my alma mater for leading this Black women centered-initiative We're actively recr… 2021-04-27 04:35:11 @adenaschachner @timothyfbrady https://t.co/oa3F3zqOed 2021-04-27 01:59:33 @yael_niv I told my first postdoc to start thinking about applying for jobs the first year he started with me. And he looked at me like I was completely insane. 2 years later he was saying that he should have followed my advice 2021-04-26 19:16:28 @JoshuaGrubbsPhD Maybe the real Josh Grubbs was the Josh Grubbs's we met along the way 2021-04-26 19:10:00 @tsawallis @TimKietzmann @Zoom @eric_mazur Curious what you used as a question to get such a large group of students engaged. 2021-04-26 18:27:25 @tsawallis Reviewer 3: Clearly an imposter. 2021-04-26 16:56:53 "We are aware of these concerns and are taking steps to address the problem in future versions of our product" https://t.co/FAaUuN1GyJ 2021-04-26 16:54:16 @cian_neuro We had the stripes and everything. 2021-04-26 16:53:27 Hey @CNN, get this white supremacist whackadoodle off your program please. You cannot possibly be intending to support these views right? He is using you as a platform to normalize the hateful rhetoric of the GOP. Please stop letting him use you to hurt our country. https://t.co/GTc6VWnPli 2021-04-26 16:32:37 Live footage of my lab's first submission to COSYNE a few years ago https://t.co/f02hnHpg8l 2021-04-26 13:11:15 Don't worry, this isn't true everywhere in science, only in the field of *checks notes* equitable and fair use of machine learning. https://t.co/UQj1vr0FIe 2021-04-26 12:09:48 RT @PSUPsychology: Dr. Kristin Buss, Head of the Psychology Department receives the President's Award for Academic Integration. To read mor… 2021-04-26 12:09:37 Great thread on expertise in deep learning. https://t.co/OLe43754fg 2021-04-26 12:02:36 RT @KordingLab: Long wanted to get into Deep Learning? Got 3 weeks in August? Learn it in a supportive intensive group environment @neuroma… 2021-04-26 12:01:56 This is a great thread on two papers about how NN models can help explain the brain. I agree with the idea that we need varying levels of abstraction for different problems. https://t.co/mGlbTurToZ 2021-04-26 11:59:08 @GunnarBlohm In this case, I don't see how the government could have done better. In the US, the federal agencies have done nothing to encourage the development and deployment of the vaccine. And Here in the us, for-profit companies are doing the community distribution legwork. 2021-04-26 11:54:57 @GunnarBlohm @dr_ppetrov E.g. https://t.co/SkaKFB1gM4 2021-04-26 11:52:06 @GunnarBlohm @dr_ppetrov I don't think china is really a communist country in this respect, it is their authoritarianism that played a hand here and I think there are questions about that vaccine's efficacy. 2021-04-25 16:39:09 RT @ira_hyman: Disturbing story of internet fraud. A very disturbing aspect to me is the role of Google. Another way in which algorithms ar… 2021-04-25 12:40:19 @azhir_io Yep, that too. 2021-04-25 12:06:49 @azhir_io Also, I would guess that patents can be a signal that the company is looking to find a way to create IP that can be sold to a bigger company rather than focusing on the product. 2021-04-25 12:04:45 @pbaylies Yea, that's pretty sobering, another example of corporate greed and shortsightedness trying to push forward when there's no feasible profit model. 2021-04-25 04:36:29 RT @Holly_4Congress: Hi, I’m Holly McCormack and I’m running against Marjorie Taylor Greene. I’m a band mom, small business owner, and *a… 2021-04-24 16:08:01 @IamSubhayu That sounds terrible to live with. You have my deepest sympathies and condolences. 2021-04-24 12:38:49 RT @seanjtaylor: Here’s my trick. https://t.co/1e3UTvtLf4 2021-04-24 12:34:13 RT @PSUPsychology: Congratulations to Dr. Koraly Pérez-Edgar for receiving the 2021 Graduate Faculty Teaching Award. See https://t.co/vI2Dt… 2021-04-24 12:29:07 @JoshuaGrubbsPhD You monster 2021-04-24 00:09:07 RT @lfrayer: Amid ghastly scenes of India's health system collapsing, the country's biggest vaccine maker @SerumInstIndia asked President B… 2021-04-23 23:39:17 @bjbalas I made the neuromatch websites with it and it was dead easy. 2021-04-23 22:54:44 @bjbalas Gsites, works and it's free. 2021-04-23 21:25:06 The hypothetical dialog at the end of this paper is absolutely wonderful for clarifying the complexity of cognitive science and also the crucial importance of a diversity of approaches. If you don't have time to read the whole paper just skip to that part at least. https://t.co/6tMHtt9uhL 2021-04-23 20:26:04 @OffTargetEffect How did you remain calm in the face of such absurdity? 2021-04-23 20:24:17 RT @NeuroNaud: Hey comp-neuro twitter, I am looking for a post-doctoral fellow to join our team. If you (or know someone who) are intereste… 2021-04-23 20:05:21 @ira_hyman Definitely sounds like "you're wrong" to me, but the proof we be to play the video to someone who hasn't read the prompt. 2021-04-23 04:46:27 RT @SECNAV75: #HarveyMilk was a Navy diver (1 of toughest jobs there is)who was kicked out only for being gay. Made our Navy weaker. I name… 2021-04-23 04:38:02 RT @theNowhereLab: Hi Twitter, we’re the Nowhere Lab, a lab for people that don’t have a lab founded by @priyasilverst. We run weekly lab-m… 2021-04-23 04:37:00 @neuro_data Channel: I think it is anything that can change state in a persistent way in response to an external event. Would include an ion. 2021-04-23 02:57:33 @samuelmehr @JoshuaGrubbsPhD I can't tell if he can't stop, or doesn't want to. 2021-04-23 01:23:56 RT @LarsPenke: Wir suchen derzeit eine(n) deutschsprachige(n) Postdoc, die/der unsere Abteilung Biologische Persönlichkeitspsychologie an d… 2021-04-23 01:21:58 @fernandaedi @timmytimmytim @PessoaBrain I suppose that's probably more likely to work. 2021-04-23 00:41:49 @graemedmoffat @KordingLab @yasmeenbrain There's that, but also, it can't require 3 GPU hours to transact 30 cents either. (or whatever the current rate is) I guess I'm annoyed that people ever sold this as currency. Who could ever have thought that was viable? 2021-04-23 00:29:06 @graemedmoffat @KordingLab @yasmeenbrain Sure call it whatever you want but it was definitely sold as a way to make transactions easier. I'm looking forward to this thread being overrun by crypto Bros tomorrow morning. 2021-04-22 23:29:49 @KordingLab @yasmeenbrain Yea I don't see the advantage and lots of drawbacks. You need to burn down a few trees to perform the smallest transaction and I think it just gets worse as the transactions accumulate. 2021-04-22 21:18:19 @hakwanlau Fantastic! (for you, not so much for the US) I wish you well on the new adventures 2021-04-22 21:17:14 RT @meganakpeters: the US is losing a great scientist, but congrats to @RIKEN_CBS! and, great opportunities here to work with this truly s… 2021-04-22 18:53:03 Your AirTag has moved more than three feet away from another Apple Product. Please relogin to your Apple account. For our amusement, your password will be randomly rejected 40% of the time and your account will be locked if you fail to login successfully twice in a row. https://t.co/FTbj3ZS0bL 2021-04-22 12:02:43 @TuckerCarlson has always had these horrible views deep down in his bones. It's not just an act. https://t.co/VhdTdsP6IE 2021-04-21 21:38:38 RT @neuro_data: #LearningSalon (w/@criticalneuro+@blamlab) is thrilled to announce that this week we will be spending time with our loved o… 2021-04-21 19:52:13 @tuennermann @JeffreyFDunn @m_wall Or to put it more succinctly: https://t.co/sZvsBNWS3n 2021-04-21 19:49:37 @tuennermann @JeffreyFDunn @m_wall The answer is here: https://t.co/a6LH1fEEne 2021-04-21 17:31:23 @timmytimmytim https://t.co/PUhqKYAKbC 2021-04-21 14:29:00 @JeffreyFDunn @m_wall Agreed, this isn't being caused by the arrows, so it's a bit misleading as presented. 2021-04-21 14:27:40 @hisotalus @m_wall @DrGBuckingham There's something amiss here. If you cover the arrows with your fingers the circles still seem to move. 2021-04-21 13:26:22 Debate me! I can take it. ..Not Like that. https://t.co/1Zj7WXJ3bH 2021-04-21 13:22:00 RT @PSUCogAgingLab: Always great to get work out hard work published. Even better when it includes greats grads and collaborators. https://… 2021-04-21 13:06:32 RT @rashadrobinson: We know police unions, white nationalists & 2021-04-21 00:38:54 @tomstafford @timmytimmytim @tsawallis But this is a very sensory specific view of cognitive. There are so many other aspects for which SDT has no special relevance. I think students should get. A broader perspective. 2021-04-20 23:51:21 @shengokai That was painful and revealing :( 2021-04-20 21:22:07 RT @lornaquandt: At the risk of putting myself out there... I could use more invited talks on my CV. I really love giving talks and I pull… 2021-04-20 16:19:49 @blamlab @behrenstimb @dileeplearning @KennethHayworth @KordingLab @neuro_data @chchatham @talyarkoni @tdverstynen @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest @DLBarack I would qualify that as low-level vision. Much of what is done by vision I would consider somewhat cognitive/abstract. 2021-04-20 13:47:48 @KordingLab @neuro_data @katecrawford That is indeed a great idea! 2021-04-20 13:47:29 RT @KordingLab: Another one of our leaders at @neuromatch DL school is Tim Lillicrap (@countzerozzz ) from deepmind. He is one of the autho… 2021-04-20 12:52:07 @emilymbender What a bunch of horseshit that was. Equating working for NGOs with speaking to VCs tells you where his values are. 2021-04-20 01:50:49 @KennethHayworth @neurosutras @KordingLab @blamlab @neuro_data @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest Don't get me wrong, I wish a connection map would be sufficient. I just don't think neural dynamics are so straightforward. Thank you as well! 2021-04-20 00:45:40 @KennethHayworth @neurosutras @KordingLab @blamlab @neuro_data @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest All I really wanted to know was whether connectomes can give us molecules. I wasn't expecting to be able to convince you to change your beliefs. 2021-04-20 00:40:37 @KennethHayworth @neurosutras @KordingLab @blamlab @neuro_data @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest I'm going to rely instead on my intuition from decades of work which has taught me that when building recurrent networks even small changes in parameters can have dramatic effects. That's not something I need papers to convince myself of. 2021-04-20 00:27:53 RT @KordingLab: I want to talk about a few of the lecturers at the @neuromatch deep learning school that taught me so much. e.g. @SaxeLab h… 2021-04-20 00:26:08 @KennethHayworth @neurosutras @KordingLab @blamlab @neuro_data @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest I'm in the camp that there are a lot of crucial, nonlinear molecular mechanisms within neurons and even synapses. So unless we can drill down to the molecules I just don't see how the connectome is going to give us enough information. Ergo @blamlab is right IMO. 2021-04-20 00:17:28 @neuro_data @KennethHayworth @KordingLab @blamlab @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest No, the ENIAC was reserved for SPACEWAR 2021-04-20 00:11:51 @KennethHayworth @neurosutras @KordingLab @blamlab @neuro_data @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest I think the question being asked here is whether you can predict the functional dynamics of a recurrent network if you only know the sign of connections but not the values. It's hard to believe that the answer is yes but I admit to not having proof one way or the other. 2021-04-19 23:48:02 @KennethHayworth @KordingLab @blamlab @neuro_data @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest My first paper in comp neuro was published in 1997, maybe I can help 2021-04-19 23:47:24 @KennethHayworth @KordingLab @blamlab @neuro_data @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest It's the connectivity and the parameters. Connectivity by itself is insufficient. 2021-04-19 23:30:37 @tdverstynen @SaraASolla We'd be all https://t.co/4twQMLZuKE 2021-04-19 23:28:29 @KennethHayworth @KordingLab @blamlab @neuro_data @chchatham @talyarkoni @tdverstynen @behrenstimb @dileeplearning @jjodx @stochastician @theamygdaloid @IrisVanRooij @o_guest At what level is the mouse connectome going to deliver? Just synapses? If not all the way down to molecules, I don't understand how is it going to define which attractors exist. 2021-04-19 19:23:12 @PsychScientists @DrGBuckingham Yea I don't really get a depth percept here. It just feels like my visual system is shorting out as it tries in vain to figure out what direction each contour is going. 2021-04-19 19:21:43 @jioornavsiri Well at least you get to live rent-free, for what that's worth. Unless you count the many hours you spent writing those papers as rent... 2021-04-19 18:35:03 RT @KordingLab: 3 week @neuromatch DL Summer course - a truly amazing faculty. We will teach everything from the ground up. https://t.co/aj… 2021-04-19 18:31:03 RT @viajake: See mice smell! Multi-lab collaboration seeks talented research tech for a project focused on understanding how the sense o… 2021-04-19 17:24:03 @o_guest Well it is revealing (of my inaccurate understanding) 2021-04-19 15:35:19 @o_guest Guilty as charged, sorry about that. 2021-04-19 14:10:05 @sharoz You can't even change your skin color to blend in with your surroundings! 2021-04-19 12:55:54 Today humanity flew a powered aircraft on another planet for the first time. Your move, cephalopods. https://t.co/T0mtZSbOcq 2021-04-19 02:44:00 RT @KordingLab: So you are not a computer scientist. Maybe you do Neuroscience. Psychology. Materials. You name it. But you are good with d… 2021-04-18 23:27:32 RT @TheTrafton: New job posting! I am searching for a lab manager to help with projects involving eye-tracking, EEG, navigation and attenti… 2021-04-18 20:30:07 @o_guest Black flowers, I love it. How are you liking the new place? 2021-04-18 19:29:07 Mice neurons can detect blockchain activity. This is huge. All of physics is now meaningless. https://t.co/Q83W9BHSRu 2021-04-18 15:19:47 @BoringDystop1a @Tweetermeyer https://t.co/sguPNIdCP7 2021-04-18 15:15:23 RT @Tweetermeyer: Two people dead, nobody in the driver seat. 6,000 lb experiments in half-baked, camera-only autonomy, capable of doing 0-… 2021-04-18 13:37:47 @DrYohanJohn Where's the RCT on whether we need to sleep? Or eat? It's all up in the air now. Is it even good to learn a language? 2021-04-18 12:03:08 RT @AthenaAkrami: 1 year ago, first seeds of our Patient-Led Research group were planted out of frustration & 2021-04-17 22:08:52 This effect is a little stronger if I play this noise in the background pretty loudly: https://t.co/DsgwqKa4j1 https://t.co/PrzfD3eJsz 2021-04-17 16:30:55 @ira_hyman To me it seems like it's a combination of the words in the screen priming the perception and then also the lip movements that match those words 2021-04-17 16:30:21 @ira_hyman You could test this by playing two sounds one with the very faint volume one without and seeing if people could tell the difference. 2021-04-17 14:51:51 Does anyone else hear the words (very faintly) when viewing this silent GIF? This might be a linguistic version of the bouncing electric suspension towers. https://t.co/DA4IIaDS27 2021-04-17 14:39:27 @neuromusic Apparently they changed their minds after this announcement but the new cap is not yet announced. 2021-04-17 03:39:03 @PsychScientists How is it not a sensible question to quantify the information available in visual cortex? 2021-04-17 03:32:29 @BestmannLab @neuromatch We try not to teach them so much but they just don't listen. 2021-04-16 23:40:59 @yoavgo I think what you are asking for is not possible, superordinate is only a relative designation as the semantic network is too complex to follow a single gradient. But @glupyan may know 2021-04-16 23:25:04 @iamknighton Heh, good point. 2021-04-16 14:00:15 @JoshuaGrubbsPhD https://t.co/vuPHrok8AA 2021-04-16 13:50:43 RT @KordingLab: so, we are thinking of teaching @neuromatch summer school materials (both for DL and for computational neuroscience) on @ti… 2021-04-16 13:16:42 @shengokai https://t.co/g8pUu6m95i 2021-04-16 12:22:08 The NMA is now accepting applications for students, TAs and Mentors for its 2021 courses: Comp-Neuro and Deep learning. Deadline for applications is May 7. https://t.co/VOtNmwcTSm 2021-04-16 12:10:55 @hisotalus @JimGrange @rj_redden Agreed, I wouldn't ask basic questions that make it sound as if you hadn't done your homework, or are just filling time. 2021-04-16 11:56:32 @hisotalus @JimGrange @rj_redden I disagree with that a bit. If a candidate doesn't ask at least some penetrating questions they might come off as disinterested or aloof. Asking good questions is both informative and conveys an impression of being sharp and motivated to succeed. 2021-04-15 23:32:02 @Merz Probably a 30 or a 40 2021-04-15 23:00:19 @chazfirestone @SBMost @carolinemyers0 @Pascallisch @carrasco_lab https://t.co/wCRsJGD4tl 2021-04-15 22:19:15 RT @JohnSakaluk: In the off-chance you’re following me and not Sam, we’re looking for postdoctoral fellow (2 years) starting in the fall! C… 2021-04-15 22:17:53 @nickgaspelin Congrats!!! 2021-04-15 22:11:45 @nilosarraf @KordingLab @neuromatch And fixed! 2021-04-15 20:36:35 NMA is once again planning to provide a stipend to TAs for their time (about 1 month of full time work). We can't guarantee the exact amount yet because we don't know the #'s but it was USD1500 last year and that is baseline goal for this year https://t.co/yRSwAXMVzz 2021-04-15 19:55:01 RT @jslbutler: A world class line up of content creators for the deep learning @neuromatch summer school in August. And the phenomenal com… 2021-04-15 18:32:53 RT @INCForg: 5 days away: the virtual 2021 INCF Assembly! April 19-28! Exciting program covering different practical aspects of #FAIR ne… 2021-04-15 18:06:33 @KordingLab @nilosarraf @neuromatch Yep, it's already been sent to the dev team 2021-04-15 18:05:54 @nilosarraf @KordingLab @neuromatch It makes sense! You're just the first to find this bug and it only seems to exist for that field, which is dynamically populated. We may get a fix in a few days. 2021-04-15 17:45:23 @nilosarraf @KordingLab @neuromatch Thanks for the feedback! It looks like you're using a mobile browser, but we haven't tested it yet on those platforms. Could you try from a desktop instead and let us know if that resolves it? 2021-04-15 16:47:21 @MillerLabMIT Just doing a little bit of red-teaming. Although I guess it's a bit too late in this case. 2021-04-15 16:46:31 If you missed out on NMA last year, or want to take our new Deep Learning course, applications just went live for students, TAs and Mentors. https://t.co/VOtNmwcTSm 2021-04-15 16:35:22 What a great place to put my chewed gum https://t.co/VHiyifOcRt 2021-04-15 15:56:28 @talyarkoni @paulbloomatyale Sorry mine got flagged for vulgar language. 2021-04-14 20:18:32 @drkjjeffery @criticalneuro @tyrell_turing I think the skin texture too, it's not sufficiently transparent, more like a rubber sheet. 2021-04-14 03:29:41 @ira_hyman Hey but at least screenshare works sometimes. It doesn't *always* fail. So there's room for being worse still. 2021-04-14 02:23:46 @ira_hyman Teams is the worst, so much worse than zoom, and zoom was already the worst! 2021-04-13 15:16:01 RT @_joaogui1: Hi folks! Me and my friends from @data_icmc are organising a seminar series themed Understanding Deep Learning. https://t.co… 2021-04-13 14:27:13 @timmytimmytim @Nate__Haines Agreed, we went through this paper in lab meeting and it was very clarifying 2021-04-13 13:42:33 RT @WiringTheBrain: This poor Irish girl... "I'm six and my own mother won't let me go to the pub!" #callchildline https://t.co/HD8sGLqWzo 2021-04-13 13:23:26 @PaulEDux @roydanroy Well if the rest of the world can get part of its act together that will at least help keep things under control in Australia 2021-04-13 13:19:00 @PaulEDux @roydanroy Oh boy that really sucks I'm so sorry 2021-04-13 12:45:36 @roydanroy Or even easy ones. :( 2021-04-12 18:07:40 @jenlhoy @atypical_me @neuroamyo No argument there, I'm just defending credit assignment. 2021-04-12 18:06:52 @atypical_me @neuroamyo Those aren't granular and they're not public. 2021-04-12 18:05:01 @atypical_me @neuroamyo @TheHandLab Sure, there's always room for improvement. I'm just saying that there is clear utility in granular credit assignment, which is where we started this discussion. Citations are imperfect, but are better than any alternatives I've seen. 2021-04-12 17:45:41 @jenlhoy @atypical_me @neuroamyo I'm not saying that we have the problem solved, just that there is some utility in assigning credit. 2021-04-12 16:55:47 @atypical_me @neuroamyo But that's also my point, citations provides a method for a wide network (yes biased, but still a network) to be given some credit, rather than just a few historical figures, which is even more biased and ignores most of the work. 2021-04-12 16:53:41 @neuroamyo @atypical_me That's my point, apart from patents, or making the news, how is credit assigned in industry? 2021-04-12 13:52:26 @neuroamyo @atypical_me Yes, this is definitely true, but I'd argue it is still better than the alternatives we've seen on the industry side. 2021-04-12 13:24:15 @atypical_me @neuroamyo I would argue that despite its faults this is much better than the method of heaping credit on just one or two people at the very top, or patents which convert credit into a sellable good. 2021-04-12 13:22:42 @atypical_me @neuroamyo But science has always been this way, as has the realm of invention/industry. Edison, Ford, Tesla, Wright brothers. This is just human nature To you want to be remembered. The only difference is that academia has created a granular citation-based attribution system. 2021-04-12 12:25:23 @neuralreckoning The fact that we have a patent system argues that credit attribution is even more important in industry than it is in academia 2021-04-12 11:54:34 @atypical_me @neuroamyo E.g. look at this. https://t.co/vahlQaNZQ1 2021-04-12 11:42:34 @Raamana_ @uhaul I've never had to answer that question... 2021-04-12 11:40:49 @atypical_me @neuroamyo Who's to say that they're not worried about securing their place in history? Working in that environment might not give them a lot of latitude for expressing how they would like to be credited 2021-04-12 01:39:20 @tage_rai @EmilyGorcenski Tage I've been hitting for the cycle so long that I can't even remember why I'm doing it. 2021-04-12 01:29:14 @JoshuaGrubbsPhD Just 53 more signatures on my petition to remove Josh from the internet, then the white house legally has to respond. 2021-04-12 01:27:35 @atypical_me @neuroamyo And finally it provides an intrinsic source of motivation that works for a lot of people. I don't know what science would look like if we could just write whatever we wanted without thinking about where any of the ideas came from, but I suspect it would suck. 2021-04-12 01:25:56 @atypical_me @neuroamyo There are several reasons why it's helpful to keep track of who came up with what idea. First it's a very cheap form of allocating credit in a way that can have useful professional consequences. Second it's a way to index a network of who has expertise in what areas. > 2021-04-12 01:15:38 This whole thread is terrifying and infuriating. https://t.co/i09FWhXaK1 2021-04-12 01:09:27 @talyarkoni Sounds like a great time, good luck! 2021-04-12 01:07:39 @CrystalNSYoung @tage_rai 2 associate editors go in, one editor comes out 2021-04-11 22:58:57 RT @analog_ashley: Just preparing for lecture tomorrow https://t.co/Ho2YMnLuwq 2021-04-11 19:17:56 @bmwiernik Throwing things at me is also acceptable as a form of notification. 2021-04-11 12:22:29 RT @JeffMYau: As anti-Asian sentiment grows, I’ve been reflecting on my own experiences as a Chinese American. This highlights how “norma… 2021-04-11 02:04:14 @RichardDawkins This is a wildly false equivalence. There's nothing to discuss. Please go back to your book tours or whatever keeps you off twitter. 2021-04-11 00:04:46 @JoeHilgard Bottle of whiskey will do that.. 2021-04-11 00:02:26 @neurobongo https://t.co/1OGwudFZZI 2021-04-10 23:40:40 @datingdecisions That doesn't look Ill advised. Ill advised would be "digging a really deep hole for no reason" 2021-04-10 23:25:31 @neurograce https://t.co/mNz3nOCZ8f 2021-04-10 22:55:57 @iandanforth @neurobongo @ArloSmartHome @ring You can roll your own system with something like ispy. Lotta work but you control the data. 2021-04-10 20:57:14 @vineettiruvadi @neuroamyo There's also the fact that outlets like CNN just repeat his press releases despite a general lack of transparency while scientists that are transparent about methods/results would never get such coverage. He's doing the opposite of what we consider good practice and gets lauded 2021-04-10 18:41:02 @neuralreckoning Thoughts and prayers 2021-04-10 16:02:24 @neuroamyo Maybe partially causal though. I completely agree that if politicians see this and send more money to the NSF, that's great. 2021-04-10 15:33:37 @neuroamyo I guess people just react to it differently, which was your original point. So yea, I guess that mystery is not solved. But I think it's understandable why people are upset and we don't need a belief in zero-sum to explain it. 2021-04-10 15:02:41 @neuroamyo There's a lot of work he's not citing, goes back much longer and includes core neuroscience. 2021-04-10 14:50:05 @neuroamyo I'm not concerned because of the zero sum game I'm concerned because he basically steals credit for previous work without clear attribution. And that's why you observe #2. 2021-04-10 14:48:01 @WorldImagining @AthenaAkrami @neurograce This sounds like a job for ...@Neuro_Skeptic ! *Turns on the Skeptic signal* 2021-04-10 14:33:58 @andpru @aina_puce @KordingLab @elonmusk Plausible based on what? 2021-04-10 14:31:30 @AthenaAkrami @neurograce Yep. If Musk put out a video showing wired recordings of neurons for 40 years ago, he would still be front-page news. He has made the media a part of his comms, and he no longer needs to earn coverage, he just gets it for free. 2021-04-10 14:26:17 @andpru @aina_puce @KordingLab @elonmusk Are we confident that there are thousands of isolated single units? 2021-04-10 14:25:07 @bjbalas @neurobongo Just a few more years! 2021-04-10 02:39:38 RT @neuro_data: Some highlights from #LearningSalon with @perryzurn & My favor… 2021-04-09 18:02:06 RT @neuro_data: Super excited to be hosting twins @DaniSBassett and @perryzurn on #LearningSalon today @ 4PM ET on https://t.co/FOjF3pC4bf… 2021-04-09 13:59:40 @Raamana_ I have no problem with him being a businessman, but he sells himself as a scientist and people perceive him as such. 2021-04-09 12:17:16 This is what bugs me about Elon's approach to science. https://t.co/3C7C9PeIIB 2021-04-09 02:11:09 @neuroMDL @Zoom You get me. 2021-04-08 23:41:28 @tsawallis Zoom doesn't make it easy with their mighty morphin power windows that reconfigure themselves when you share screen. 2021-04-08 16:16:27 @SBMost @PessoaBrain @chazfirestone Exactly, we often frame these as "whether" x affects y, but it's typically helpful as a way of studying "how" x affects y. 2021-04-08 12:38:42 @o_guest Ugh, I'm sorry that this is your experience and thank you for explaining. 2021-04-08 12:02:08 @azhir_io Got it. FWIW, It's a real loss to academia that you didn't get a grad school spot this year. We need more people like you. 2021-04-08 11:00:51 @azhir_io I hope it went better than you thought. What kind of positions were these for? 2021-04-08 01:51:37 RT @RainaBrands: While it is becoming normative to acknowledge maternity and its impact on work, what about #miscarriage? I have decided t… 2021-04-08 01:50:25 @RuthRosenholtz I'm frankly in awe that you have so quickly identified the point of domestic dysfunction during my grad school days. 2021-04-08 01:24:13 @RuthRosenholtz I hate to discredit your fledgling theory but I once poured frozen peas into my coffee. 2021-04-07 18:43:01 @JoshuaGrubbsPhD That's why it makes a nice bite-sized alternative to an actual debate. 2021-04-07 15:52:38 @tdverstynen Show me the evidence where this happens Tim 2021-04-07 15:48:50 @lioninawhat @distractal @AlexeiAravin @UnBette @Abebab @tage_rai Who decides who is smart? 2021-04-07 15:47:40 @AlexeiAravin @UnBette @distractal @Abebab @tage_rai I'm not the one who introduced the concept of voluntary into this discussion. I've been arguing instead that the distinction between voluntary and involuntary is unclear. 2021-04-07 15:44:40 @JoshuaGrubbsPhD May I interest you in some eugenics instead? 2021-04-07 13:08:15 @hannah_merseal What is Bill Gates telling you to do?? 2021-04-07 12:47:07 @AlexeiAravin @UnBette @distractal @Abebab @tage_rai There is plenty written about the implications of eugenics so no need for me to summarize. I will add though that the distinction between voluntary and involuntary is rarely crystal clear which is why I put it in quotes. 2021-04-07 04:28:45 @UnBette @AlexeiAravin @distractal @Abebab @tage_rai No the toxic part is not just the involuntary part. There are a lot of reasons that even "voluntary" eugenics has problematic implications for society and particularly marginalized groups. 2021-04-07 04:12:35 RT @djnavarro: For my academic colleagues: please pay attention. Trans people have been telling this exact story for years about this perso… 2021-04-07 04:11:44 @AlexeiAravin @distractal @Abebab @tage_rai If there's no difference, what are you objecting to about the original tweet? 2021-04-07 01:17:43 RT @GunnarBlohm: @EllaBatty is amazing to work with! If you have some time to help a great cause and benefit from the @neuromatch community… 2021-04-06 23:46:43 @AlexeiAravin @distractal @Abebab @tage_rai Do you perceive an important ethical difference In using selective breeding to change intelligence versus doing it with biological tools? 2021-04-06 23:32:20 She is the absolute best. Another person we do not deserve. https://t.co/EqM2j5Bt5M 2021-04-06 22:05:58 RT @meganakpeters: NEUROIMAGING POSTDOC JOB ALERT! i'm collabing w/ Aaron Seitz, Xiaoping Hu, @UCR_LANIlab & 2021-04-06 20:11:01 @AlonZivony I often end up having new ideas while preparing a presentation. 2021-04-06 13:02:58 @neuroecology I'm sorry Adam :( good luck in figuring out what to do next. 2021-04-05 18:02:26 @PsychScientists Yes, but they can also be the result of a predictive mechanism that has just received new information. 2021-04-05 17:30:33 @PsychScientists Not if that was representative of the likely trajectory no? e.g. if I watch someone bowling, one can predict if the ball path is going to curve based on a variety of factors. 2021-04-05 17:13:19 @PsychScientists Isn't this an issue of how narrowly you define prediction? I am also not thrilled with prediction as a universal explanation, but I don't see how you can argue that prediction can't produce reversals. 2021-04-05 16:00:23 @glupyan @ka_lindquist @PaulaNiedenthal @adriennerwood Yes, there is a huge difference between inferring emotions from facial expressions, and studying how people infer emotions form facial expressions. They sound similar but are entirely distinct. 2021-04-05 15:27:57 This is a great point. Calling these systems "emotion recognition" is a category error even if you qualify them as "inaccurate". They are not measuring emotions, at all. They are measuring facial expressions, face shape, shadows, skin color, etc. https://t.co/0TkM5Sx6HH 2021-04-05 13:56:09 This is disappointing. And it's not just Facebook that's donating if you read down the thread. https://t.co/8oP5E0H8dj 2021-04-05 12:04:41 @PsychScientists E.g. check this out. https://t.co/Sj3Axc52zF 2021-04-05 11:48:25 @tdverstynen @PessoaBrain Rarely if ever.. 2021-04-05 11:43:49 @tdverstynen @PessoaBrain Also those discoveries are built on the work of many others. 2021-04-04 14:16:39 @Raamana_ Yea, it's possible that those specific amish don't use them even though kids in their local church do. I've also heard that some homes have electric lighting in their house using fiber optic bundles thay they can turn them off when the deacon comes to visit.. 2021-04-04 14:12:12 @Raamana_ Think of them kind of like an IRB, making decisions about what to permit and sometimes they don't fully understand the implications of new technology. (But the teenagers sure do :)) 2021-04-04 14:10:50 @Raamana_ Well it varies by region and age group but also don't be surprised if they are reluctant to tell you everything about their culture. As I understand the decision makers in the community didn't see phones as a problem because there is no physical connection to the modern world. 2021-04-04 14:02:11 @Raamana_ They also use cellphones and Facebook. So it's fair to say that they do use computers. 2021-04-04 06:20:39 RT @womenpostingws: huge fucking W https://t.co/nljVjhuPFn 2021-04-03 21:44:02 RT @meganakpeters: https://t.co/4KWwCDCzqy 2021-04-03 15:38:48 Good thoughts about the impossible standards of K99 grants https://t.co/ymKeysMCtP 2021-04-03 13:45:17 RT @AndyPerfors: Now I really, really wish professors got our own classrooms. I'd put all the digits of pi going around the whole room, and… 2021-04-03 01:42:38 @ptheodoni Nice artwork! 2021-04-02 21:29:42 This position is still accepting applications. This is a great opportunity for someone to get some experience in a wonderful neuroimaging lab. https://t.co/JGJAbCDNfx 2021-04-02 20:14:59 Also, please use Chrome if you can. Crowdcast is picky. 2021-04-02 20:11:28 This is starting in a few minutes. If you join us, please make sure you come to the right room (Simon Kornblith) using the SCHEDULE dropdown at the top of the crowdcast window https://t.co/GSuPwPta5c 2021-04-02 13:57:09 @azhir_io @tdverstynen @KevinZollman This is true but it is possible to appreciate that complexity exists without understanding it. For example I am aware that there are layers of subtext in discourse of other cultures that I am unable to even perceive, let alone decipher. 2021-04-02 13:41:35 @tdverstynen @KevinZollman Yeah I see the oversimplification on the left as well 2021-04-02 02:35:16 RT @neuro_data: Tomorrow (Friday) at 4PM EDT @skornblith with me, @criticalneuro @blamlab and YOU on https://t.co/PgvNxujuid with help from… 2021-04-01 19:01:40 @vineettiruvadi @tyrell_turing @_rdgao Now you're talking. Let's throw some MLE at this problem... Oh, it can't converge 2021-04-01 18:52:27 @vineettiruvadi @tyrell_turing @_rdgao Now it's publish and/or perish 2021-04-01 17:41:04 @tyrell_turing @_rdgao Publish! But not too much. Wtf seriously. 2021-03-31 20:08:06 @andpru @graemedmoffat @tyrell_turing @AdamHantman But I think even if you can demonstrate that it's possible to control individual motor units, that does not mean that we can control multiple ones at the same time Independently. That's the likely cognitive limitation one would encounter. 2021-03-31 13:00:31 RT @patrickmineault: Tutorial on RNNs in neuroscience in Python, Julia & 2021-03-31 04:17:11 @NelsonCowan1 Yes, it's very complicated to consider what is truly bad faith, or in what reference frame. i.e. if someone is delusional, maybe their arguments do not seem bad-faith to themself, but are essentially bad-faith for the rest of the world in terms of their impact. 2021-03-31 02:07:29 RT @meganakpeters: at risk of tooting my own horn, here is a soundbite of me talking about how @neuromatch Academy is going beyond passivel… 2021-03-31 02:07:22 RT @KordingLab: 3 week deep learning summer school (Aug 2-20) by at @neuromatch is coming. Will have projects. And obviously tutorials wher… 2021-03-30 17:43:41 So I'm wondering if anyone has run into this in the context of instruction. Are there tools we can rely on? Would be helpful for mentorship as well. 3/fin 2021-03-30 17:42:39 Critical thinking typically focuses on the argument itself, But thinking about good-faith requires us to think about the history/motivation/bias of the person. It’s a much more difficult kind of reasoning and we often spend years doing it before we realize it has a name. 2/ 2021-03-30 17:41:35 People advocate for critical thinking education, but what about good-faith thinking (conversely detecting bad faith)? In my view these are different very thinking styles of evaluation, and students have even less exposure to the latter compared to the former. 1/ 2021-03-30 04:00:29 RT @meganakpeters: I am extremely pleased to share with you all that our paper led by Brian Maniscalco is now available online at @PLOSComp… 2021-03-30 03:58:39 @neuro_data Isn't it also impossible to explain the behavior of a human without explaining society/culture? Practically zero of the behaviors I do in a daily basis make any sense outside of this space 2021-03-28 21:12:09 RT @ButNotTheCity: Cis people will literally call a car 'she' and then claim pronouns are about biological reality. 2021-03-28 21:09:59 RT @GaryMarcus: AI can be part of solution. Or it can exacerbate the problem. As long the models du jour do little more than parrot histo… 2021-03-28 15:29:53 @syndrome22q1 @LiamWBZ @FaZeRug @dylanblackradio @Reflog_18 @KevinBoston25 @PaulaEbben @KurtSchlichter @NewSpirit2020 @RexChapman @TankSchottle Hello Justin! It's great to meet you. 2021-03-28 12:32:25 @jmourabarbosa @JulioMTNeuro Some people have much stricter definitions of working memory though. For example in Oberauer's 2009 paper he defines 6 criteria for human WM. I agree with that perspetive. 2021-03-27 23:34:28 RT @meganakpeters: the countdown begins!! 2021-03-27 19:57:38 RT @SuezDiggerGuy: THE SHIP MOVED!! Although just 17 metres but this is good indication. The tugboats are all sounding their foghorns in ju… 2021-03-27 19:52:15 RT @BlackinPsych: Today, on our FINAL day of #BlackInPsychWeek, we want to highlight your resiliency! 1⃣ Tell us what brings you joy/… 2021-03-27 18:51:23 @criticalneuro This worked pretty well for some people here in State College PA 2021-03-27 13:38:54 @JoshuaGrubbsPhD @samuelmehr You are either our Chuck Yeager, or maybe Evil Knievel... I can't decide which 2021-03-27 13:25:33 @graemedmoffat I disagree, it will be both disappointing and not surprising when they fail to follow suit. 2021-03-27 13:24:00 @samuelmehr @JoshuaGrubbsPhD So glad to have Josh helping us to firmly explore the boundaries of Twitter discourse https://t.co/Ka0ysXyT6k 2021-03-27 09:20:21 RT @MrGeorgeWallace: I'll bring voters all the damn water I want. I'm gonna have a cart out there like a motherfletchin' flight attendant.… 2021-03-26 23:39:50 @fedeadolfi Always do X but sometimes don't. You're welcome. 2021-03-26 12:43:23 @SMBrocklehurst @IntuitMachine @tyrell_turing @theBjornErik @KordingLab @WiringTheBrain Yep, I have the same opinion. There is a conflation in this thread between the fact that there is hemispheric localization and that these drive personality traits. These are two separate concepts. 2021-03-26 12:15:21 @IntuitMachine @tyrell_turing @theBjornErik @KordingLab @WiringTheBrain Also, as Blake pointed out these differences are not entirely consistent though someone needs to approach with caution. 2021-03-26 12:14:27 @IntuitMachine @tyrell_turing @theBjornErik @KordingLab @WiringTheBrain I agree that the left and right halves of the brain process information differently. Lesions make this abundantly clear. I was only objecting to the notion that people can be "left brained" based on your eye dominance thread. Apologies for the misunderstanding 2021-03-26 02:00:12 @IntuitMachine @tyrell_turing @theBjornErik @KordingLab @WiringTheBrain I mean, there is no doubt that function is highly localized. It's also differentiated front-back, but noone talks about being a back brain thinker. Also, only our legs are able to run but noone talks about being a "leg mover" 2021-03-26 01:48:10 @IntuitMachine @tyrell_turing @theBjornErik @KordingLab @WiringTheBrain But the fact that function is lateralized (even if it were 100% consistent) does not mean that there is such a thing as a left-brained thinker. These are completely different concepts that are erroneously conflated. 2021-03-26 00:07:16 @IntuitMachine @tyrell_turing @theBjornErik @KordingLab @WiringTheBrain There are indeed differences in what the left and right brain halves do but this does not mean that people can half left vs right brained thinking styles. E.g. a car has different functions in the front and back but they function collectively. 2021-03-25 21:36:36 RT @patrickmineault: How do you build big things from scratch? My long overdue post on building Neuromatch Academy 2020 https://t.co/YO8k6k… 2021-03-25 19:06:09 RT @roger_p_levy: The MIT Computational Psycholinguistics Lab seeks to fill an open postdoc position for an @MITIBMLab supported multi-PI p… 2021-03-25 16:29:51 GPT-0.1 https://t.co/qMLCC4lgRW 2021-03-25 16:13:58 @IntuitMachine Again, this is about where it sits in the visual field, not which eye is carrying the information or is dominant. But the information in the fovea (from both eyes) seems to go to both hemispheres. 2021-03-25 15:31:27 @IntuitMachine There's no debate about which hemisphere processes information outside of the fovea, it's clearly split down the middle. There is a lot of debate about the foveated information is processed. I don't know if this interacts with eye dominance though. It's completely different 2021-03-25 14:20:07 @IntuitMachine Well, each hemisphere interprets one half of the input from both eyes. So the hemispheres literally see different (almost entirely disjoint) information. Ocular dominance has little to do with this, and is more about how early visual cortex deals with conflicting information. 2021-03-25 14:16:50 @AmandaGefter @PsychScientists @NeuroYogacara Which basically means that we need to appeal to cognitive psych explanations for behavior, as I interpret it. 2021-03-25 14:15:43 @IntuitMachine In this experiment, they didn't present the information to just one eye, they presented it to both eyes, but only to one half of the field in each eye. Also, the neuroanatomy of the visual system is indisputable with regard to this fact. 2021-03-25 14:10:25 @IntuitMachine @tdverstynen Yes, this is known to psychology professors. 2021-03-25 13:51:29 @IntuitMachine Remember that both eyes go to both hemispheres. So being left eye dominant shouldn't tell you much about which hemisphere is doing the processing. 2021-03-25 13:23:12 RT @BlackinPsych: THANK YOU to everyone who was transparent and described how they coped through 2020!! Tomorrow we are in for another tre… 2021-03-25 11:44:49 RT @bremen79: For the future PhD students that are deciding in these days which offer to accept, let me give you a very simple advice: Avo… 2021-03-24 16:03:03 RT @BlackinPsych: It’s Day 3 of #BlackinPsychWeek and as we prepare for our wonderful panel today we want to hear from YOU! 1⃣ Tell is a… 2021-03-24 14:20:10 @yael_niv Attention human! the system acknowledges your biological substrate's needs and allots you 5 minutes for relief. Return immediately afterwards or you will be derezzed. 2021-03-23 14:53:07 About to use the old @60Minutes episode on eyewitness testimony with Elizabeth Loftus to teach about memory and I'm having to skip huge chunks because it only has POC in the eyewitness lineups throughout the entire episode. 2021-03-22 18:32:37 This is a good mini thread by Crystal on the inherent resistance withing the science reform community for addressing problematic attitudes. We need to listen to voices like hers and believe that these warning signs have been there for years. Too few people have been speaking up. https://t.co/54vpYcWwoo 2021-03-22 17:15:25 @tyrell_turing Yep, we did too, right after SimCLR. Fully agree. 2021-03-22 15:45:18 RT @BlackinPsych: We're kicking off #BlackinPsychWeek with #BlackinPsychRollCall!! Introduce yourself: 1⃣ Tell us your name, sub-field,… 2021-03-22 13:53:23 RT @JenessaNHolder: Hey y’all My name is Jenessa and I’m a 2nd year PhD candidate. I’m originally from Guyana but currently in NYC. M… 2021-03-22 11:39:32 RT @AlonZivony: Tomorrow! Want to impress you cog sci mates with nuanced criticism about the field of attention? Well, friend, I've got jus… 2021-03-22 04:27:33 @leavittron @KordingLab This might be helpful: https://t.co/EYnqhhPIBj 2021-03-22 03:28:59 @KordingLab @dlevenstein I assume this is part of what research mentorship is about. We have discussions in which we practice the art of creating testable hypotheses for a given theory, eliminating down the bad ideas, refining the good ones, etc. 2021-03-22 00:06:40 RT @_danilo: Sometimes I worry that instead of building truly transformational technology, Silicon Valley will endlessly innovate new ways… 2021-03-21 23:53:37 @azhir_io Congrats! First step in a long journey 2021-03-21 15:40:43 @rudzinskimaciej @tyrell_turing @AdamHantman I wasn't suggesting that we don't have fine control. we can control each finger with incredible precision as long as that's the only thing we're doing. The problem arises when you try to control multiple muscle groups to do different things at the same time 2021-03-21 15:38:32 RT @RuthRosenholtz: Good thread. (My first name also has a space. I eventually dropped the second half for just this sort of reason.) 2021-03-21 15:13:10 @rudzinskimaciej @tyrell_turing @AdamHantman But again, extra dimensions don't help if you can't exert independent control simultaneously. You can experiment on yourself to demonstrate. Tap one finger at 2bps, tap another at 3bps. Easy. Now try to do both at the same time. 2021-03-21 13:53:05 @rudzinskimaciej @tyrell_turing @AdamHantman But again where's the evidence that you can achieve higher bandwidth? I can manipulate each of my hand joints independently (both physically and mentally), but I cannot manipulate them independently at the same time. That's the cognitive bottleneck. It's not about tissue. 2021-03-21 13:51:37 @rudzinskimaciej @tyrell_turing @AdamHantman Sure, but the same could be done with real motor movements. i.e. run mouse/keyboard/joystick through an ML. 2021-03-21 13:42:01 @rudzinskimaciej @tyrell_turing @AdamHantman The question (in this thread at least) is not whether it will work, but whether this will be better than using our hands and feet and physical controls like joysticks, mice and keyboards. 2021-03-21 13:36:42 @rudzinskimaciej @tyrell_turing @AdamHantman If we could get that out of one wrist, I would be very impressed. Try to control three fingers independently while also flexing the wrist. I can't and I've been practicing with this hand for 40+ years. 2021-03-21 11:43:05 RT @wesbonifay: IMO the best way to “curate science” is by creating a height leaderboard. I already gave myself a hexagonal badge for being… 2021-03-21 04:28:37 This is awesome to hear https://t.co/T02p8rNVMX 2021-03-21 04:28:03 @CousinAmygdala @andpru Well it's easier to *implant* them... 2021-03-21 03:28:45 @tyrell_turing @AdamHantman Yep, the brain controls many muscle groups at the same time and the bandwidth is impressive. But those groups are typically engaged to do one or two high level actions, they're not all doing independent things. It's hard to just draw a square and a circle at the same time. 2021-03-20 22:11:39 @DirkBWalther @J_D_Crawford With the caveat that if you're operating in a regime where 1 vs 2 tailed tests makes a difference, you're probably on thin ice statistically and should probably consider the results not very definitive. 2021-03-20 21:53:41 @tyrell_turing @AdamHantman Yea I hear you but I just don't know if I buy that. If someone is an expert at e.g. SC2, I think the interface has become exceedingly intuitive. What kinds of commands are you thinking of that might be accelerated with a BMI? 2021-03-20 21:36:44 @tyrell_turing @AdamHantman Oh lordy, Do you think I don't game? The mouse already allows an insane number of actions per minute, I think our brain is completely taxed keeping track of world state and isn't output limited in games either. 2021-03-20 17:15:09 @tyrell_turing @AdamHantman But the content generation is the bottleneck. You're suggesting that a skilled touch typist is rate limited by their typing speed when writing a paper but I really doubt it. I can certainly type faster than I can think. 2021-03-20 16:28:05 @GunnarBlohm @KordingLab I'm going to need you to turn in your causal reasoning badge! 2021-03-20 15:52:33 @tyrell_turing @AdamHantman You're assuming that the brain is capable of delivering that bandwidth effectively. I'm not sure that it is. e.g. getting information into and out of the mind more rapidly doesn't help if the mind isn't able to keep pace. 2021-03-20 15:50:58 @J_D_Crawford Yea I agree. It limits the acceptance of incidental findings, which are, let's be honest, a large portion of scientific advances. 2021-03-20 15:31:14 This is a great thread on how open science is perceived and enacted by many labs. Thank you for writing this. https://t.co/tF9xDphHMc 2021-03-20 15:20:14 @TPVogels @GunnarBlohm Sensory and motor replacement might be much closer on the horizon 2021-03-20 15:13:58 @GunnarBlohm Yeah I've heard tales of bad outcomes from surgery :(. Changing your muscles is probably a better idea than changing your bones 2021-03-20 15:12:28 @TPVogels @GunnarBlohm I was being well the optimistic there. I wouldn't wager a bottle of rubbing alcohol on this prediction. 2021-03-20 13:53:51 @GunnarBlohm Only 20 more years until brain uploading is ready. Hang in there! But seriously, my sympathies. This sounds awful and I heartily wish that you find a way to get permanent relief. 2021-03-20 13:51:20 I am amazed at her ability to mask what must be a torrent of internal emotions when she is forced to respond to questions so clearly rooted in bad-faith politics. https://t.co/Id6mGFrQlt 2021-03-20 13:46:52 @D_Wiwad @Tomdonoghue Now this is terrifying on an even deeper level. Thank you for spotting this. 2021-03-20 13:22:23 @siminevazire It's true that both take strong positions but I don't think it's fair to claim that this is a "both sides" disagreement, since the blog post is an intensely personal and fairly aggressive response directed at the authors. 2021-03-20 13:18:57 @AlexBrady @JoshuaGrubbsPhD I've just registered that domain name. 2021-03-20 11:51:38 Eva has been doing a wonderful job with helping to manage the backstage and website at the Learning Salon. She's a PhD student working on computational neuroscience at Johns Hopkins. Thank you Eva! https://t.co/080UnGzYWh 2021-03-20 11:46:31 @JoshuaGrubbsPhD Friends, I'm preparing a petition to have Josh removed from the internet. We all know it's time. 2021-03-19 20:15:41 If you're having trouble getting into the learning salon, we had to restart the session so please go to this URL: https://t.co/iQ79lOJZ4Z https://t.co/WkPmx5ux5i 2021-03-19 20:02:25 This is starting up now, should be great! https://t.co/6OxiACQAna 2021-03-19 19:53:47 @lukesjulson I disagree. It makes sense to focus more on why certain people have been murdered than why they have the jobs they have. 2021-03-19 16:22:05 @J_D_Crawford @k_v_n_l Agreed! I am sure there are many wonderful people at York. 2021-03-19 16:18:46 @k_v_n_l I'm sure it's not, and I disagree with the use of the term 'York'd'. That's unfair to the many people who work there and treat students with respect. 2021-03-19 13:32:44 @SimoneSchnall Yes, it is quite notable 2021-03-19 13:28:13 This is a very alarming example of what some students have to deal with in navigating the university system. https://t.co/bcihkAeHoi 2021-03-19 11:52:48 @SimoneSchnall Thank you for saying something. It would have been a lot easier to stay quiet 2021-03-19 04:15:50 @chazfirestone I remember 2000, I was about to give up on horses 2021-03-19 04:01:09 @zerdeve @djnavarro When people show you who they are, believe them. Ideally long before stuff like this happens. 2021-03-19 03:59:04 @chrisdc77 What do you think about this? https://t.co/14QCVC0pCE 2021-03-19 03:49:24 @curatescience Can you comment on this? https://t.co/opgW9KfXHY 2021-03-19 03:46:58 RT @SimoneSchnall: Fun fact: I was subjected to an open science *transparency audit*. Tons of work for my former students, all early career… 2021-03-19 03:45:43 @SimoneSchnall This is straight up bullying. I'm sorry that you had to deal with this awful behavior. 2021-03-19 03:44:26 @PsychScientists @SimoneSchnall It was already creepy. This has crossed into to straight up bullying. 2021-03-18 01:44:36 @timmytimmytim I can guess what this is about but what did I miss? 2021-03-17 02:10:27 @timmytimmytim Did I say lumen? I meant microlumen. 2021-03-16 23:57:01 @timmytimmytim Because you have kids that wake up at the first lumen of the morning? 2021-03-16 20:17:57 @JohnSakaluk https://t.co/ADWKsVzIFW 2021-03-16 14:11:55 RT @awprokop: The GA election call most people think of is the Raffensperger call which broke (with accurate audio) 1/3. Capitol-storming w… 2021-03-16 14:09:45 RT @TIME: "I'm fully who I am." Actor Elliot Page and the fight for trans equality https://t.co/ozlpoqa2sb https://t.co/OAKogfmoRx 2021-03-16 12:56:12 This really is a great paper for anyone interested in the fundamental mechanisms of visual attention. It digs deeply into the literature and discusses how we might improve our understanding of attention by thinking of it as a temporally continuous process. https://t.co/wxYlLwUi9r 2021-03-14 21:38:31 @sarabwtroutman Aww, I'm so sorry to hear that Sara 2021-03-14 12:52:24 @KordingLab @roydanroy I think that in order to build complex understandings, we need to incorporate multiple ideas in one coherent document. I keep running into the problem where a good solution emerges when considering several ideas at the same time. 1 idea/paper is too limited for this. 2021-03-13 15:24:17 @tdverstynen Also, hearing something from two different sources is a "majority" 2021-03-13 15:23:23 @tdverstynen N of 2 is a "clear pattern". You don't rise to "systematic" until an N of 3 2021-03-13 13:39:22 @timmytimmytim https://t.co/In5cOnaOTi 2021-03-13 13:36:48 @tdverstynen Yes but he said "so many" which means that if there is even 1, he's correct. https://t.co/KBtngBkAeJ 2021-03-13 04:15:15 RT @todd_gureckis: What professors can learn from #youtubers (10 tips in 5 minutes): https://t.co/Ml78I1UBh9 from @nyuniversity 2021-03-09 22:10:39 @timmytimmytim If you're going to make up terms, own it. How about 12.4 quizznacks? 2021-03-09 19:35:36 @nickgaspelin Looks awesome, I love the different views approach. 2021-03-09 05:57:24 RT @AJLUnited: Happy #InternationalWomensDay! In celebration, meet the Black women who continue to inspire & 2021-03-09 05:40:21 @cian_neuro @ClopathLab @FieteGroup @GjorJulijana @mon_sci @rajankdr @RangamaniUCSD @SaraASolla @SeriesPeggy @suhitais @TTchumatchenko @YiotaPoirazi @AnnaSchapiro @o_guest And @criticalneuro of course! 2021-03-09 04:09:00 RT @fairfightaction: The GOP-led Georgia Senate just voted 29-20 to PASS #SB241, one of the worst voter suppression bills in the country. I… 2021-03-09 03:16:01 @cian_neuro @ClopathLab @FieteGroup @GjorJulijana @mon_sci @rajankdr @RangamaniUCSD @SaraASolla @SeriesPeggy @suhitais @TTchumatchenko Also @YiotaPoirazi @AnnaSchapiro @o_guest 2021-03-09 00:06:57 RT @cian_neuro: For #InternationalWomensDay here are some top computational neuroscience PIs on twitter whos work I find insightful @Clop… 2021-03-08 23:08:53 RT @criticalneuro: It's international women's day. For the next week please try as hard as you can to NOT troll women, mansplain things to… 2021-03-08 22:48:43 Interested in getting paid to teach deep learning this summer? https://t.co/6WZoEN4Ut9 2021-03-08 18:12:55 I haven't read this yet, but my priors are pretty strong (because Alon) that it's going be very insightful. https://t.co/wxYlLwUi9r 2021-03-08 18:12:04 @AlonZivony Great terminology Alon! I look forward to reading this. 2021-03-07 23:01:57 @neuro_data I would say model building 2021-03-07 22:35:13 RT @IATSE: Don’t let the party that tried to cancel Dungeons & 2021-03-07 21:43:52 @sbloyd Yep, I can see the rationale, but he's still terrible. 2021-03-07 17:16:02 It is not ok for public officials to be broadcasting these conspiracy views, even if they try to cover it up. The people they are speaking to hear it for the support that it is. https://t.co/GB8V3YLN6n 2021-03-07 17:09:06 @DrGBuckingham @KangTarandeep yea, strange. I've not been following them long, so I can't really imagine what you did to earn a block. 2021-03-07 16:48:14 Awesome. With one laptop on each, that's 7 * 50 tabs! (Double that if I use firefox and chrome) https://t.co/jJfM5O3PlJ 2021-03-07 16:44:28 @IrisVanRooij Yes and of course this is a common practice. 2021-03-07 16:25:21 @J_D_Crawford But what I meant is that it seems like you're drawing a false equivalence. Marginalized people having to watch what they say and other people not having to watch what they say are not equivalent positions, nor should it be a political divide 2021-03-07 16:14:43 @TaraViegen @uygun_tunc @lakens @o_guest @IoanaA_Cristea @RickCarlsson @tunc_necip I don't even think of it as a responsibility so much as just taking small steps to make science a nicer place to operate 2021-03-07 16:02:15 Firefly https://t.co/bkBstaZhhK 2021-03-07 14:47:44 @IrisVanRooij Thanks Iris! 2021-03-07 14:04:47 @tyrell_turing But the "crap wording" exemplifies the all-to-common attitude that the social aspects of any practice are unimportant and can be ignored in our quest to penetrate to the truth. I see this attitude all over the place in the sciences and it's harmful. 2021-03-07 14:00:14 @tunc_necip @BimboPolitic @o_guest @uygun_tunc @lakens Even if the figures are not fundamentally similar (and I disagree) why not be charitable with citations? It costs you nothing. 2021-03-07 05:20:27 @J_D_Crawford This isn't a both sides kind of thing. 2021-03-07 05:19:53 @iamknighton @ceptional It works in that context too! 2021-03-07 01:27:47 @ceptional This kind of is mentioned in textbooks when we talk about the central executive and how it handles manipulations of numbers. See also Oberauer's single focus of attention and the global workspace theory 2021-03-06 22:19:51 This seems like a fairly accurate description of why cancel culture feels like a big deal to some people. https://t.co/PXJdeVpfYV 2021-03-06 20:22:31 @azhir_io I'm so sorry Azhir, it's such a heedless and terrible disease 2021-03-06 04:31:02 @PsyPost Please stop with this stuff. 2021-03-05 20:50:11 This is starting in just a few minutes! I've been following Randall's work for ages. I can't wait. https://t.co/1mkjpQohSv 2021-03-05 17:21:04 @Raamana_ @mbeisen @PittCogNeuro @LeonardoTozzi1 @NeuroPolarbear @ApertureOHBM @PLOS The reason that this is an issue is that transparency is not at all the norm in scientific societies. So that's another battle that has to be fought. 2021-03-05 17:20:26 @Raamana_ @mbeisen @PittCogNeuro @LeonardoTozzi1 @NeuroPolarbear @ApertureOHBM @PLOS Agreed, some of those coffers are important for sustainability. But without transparency it's hard to know. 2021-03-05 17:16:08 @Raamana_ @mbeisen @PittCogNeuro @LeonardoTozzi1 @NeuroPolarbear @ApertureOHBM @PLOS Agreed, but, society journal APCs are complex. Some of that money probably gets diverted into other coffers. It's not nearly as bad as for-profit, but without financial transparency it's hard to know what that money goes for. 2021-03-05 12:44:49 @tage_rai Sun revolves around the earth? Cancelled! 2021-03-05 04:30:35 RT @natalie_schwob: Penn State grad students: Vaccination appointments just opened for us. You can make an appointment here https://t.co/pj… 2021-03-05 04:25:13 @MelMitchell1 How does broccoli come up but the muffin doesn't?? 2021-03-05 04:23:41 @azhir_io @elonmusk @SpaceX https://t.co/wAiH77sANY 2021-03-05 04:22:44 @chazfirestone The salient point that I take from these machine studies is that 99.7% confidence score. Yes people are weird but we have a wide variety of fall back mechanisms that saves us from failure modes in the tails. 2021-03-05 01:04:52 @chazfirestone @GaryMarcus this is a good one. https://t.co/dJSLD3gpJW 2021-03-05 00:42:17 @azhir_io @elonmusk @SpaceX That wasn't a leak, it's an intentional feature to encourage rapid disembarking 2021-03-04 22:51:36 @ColleenDerkatch That's usually how I do it 2021-03-04 19:37:33 @neuroMDL @k_v_n_l @BrianOdegaard2 @tsawallis of course, all welcome 2021-03-04 17:59:04 @k_v_n_l @BrianOdegaard2 @tsawallis Yep, Dm me your email 2021-03-04 16:46:19 @BrianOdegaard2 @tsawallis Sure, DM me your preferred email? 2021-03-04 13:26:55 @timmytimmytim @JennRichler Check the bio. Wow!!! 2021-03-04 12:58:03 @meganakpeters @tsawallis omg yes 2021-03-04 12:57:42 @tsawallis Sure, it's very low bandwidth. Basically a place to deposit links and resources. 2021-03-04 12:45:17 @o_guest @azhir_io @tdverstynen @rasmansa https://t.co/fCfbxpIPuR 2021-03-04 12:05:43 RT @tanyaofmars: This view is AMAZING. Straight out of science fiction. Holy hell. #SN10 #SpaceX https://t.co/VZKPMO2x9A 2021-03-04 05:00:17 @tsawallis Do want to join a slack for teaching perception? 2021-03-04 04:18:48 This picture is so yikes at first. Gestalt effects create horrible image. https://t.co/yFEbFmEhgA 2021-03-04 04:17:09 RT @JBeachFerrara: I am Democrat Jasmine Beach-Ferrara. I am running against Madison Cawthorn. I have 56K followers. My opponent has 256K.… 2021-03-04 00:27:32 @azhir_io @tdverstynen @o_guest @rasmansa Yea when you're giving that response it pretty much proves the point. 2021-03-04 00:07:12 @tdverstynen @o_guest @rasmansa That is pretty much how it went down 2021-03-03 23:58:13 @tdverstynen @o_guest @rasmansa Oh that's just what we need. I've been called a racist for pointing out that someone benefits from white privilege. 2021-03-03 20:16:48 @GunnarBlohm @elneurozorro @micahgallen @meganakpeters @schrater_paul @computingnature @katebonnen @AthenaAkrami @seanescola @aarlennek @aina_puce So glad to have fallen in with such good people. 2021-03-02 18:14:07 @rmgrieves @ElDuvelle @GunnarBlohm I don't mean that I am being paid for individual reviews, rather that my annual salary includes a broad set of duties which includes service to the field, such as doing reviews. There are lots of other things in this category, like writing evaluation letters for peers. 2021-03-02 18:04:51 @patrickmineault Head of Llamas, nice. 2021-03-02 18:03:39 @neuralengine @rmgrieves @GunnarBlohm This is good advice. Academic job responsibilities are typically vague by design. It is part of what provides the freedom to explore different avenues of progress. There is a lot of checking in to align expectations for general responsibilities. 2021-03-02 16:59:39 @rmgrieves @ElDuvelle @GunnarBlohm I mean that I am being paid from the university as a part of my larger duties for the university. This is different from being paid per review by the journal company. 2021-03-02 16:58:31 @rmgrieves @GunnarBlohm @neuralengine "Service to the field" is how it is typically worded, and yes this is vague. However the specific definition is typically clarified to include things like reviewing, editing, etc. 2021-03-02 16:57:17 @rmgrieves @neuralengine @GunnarBlohm Yes, and as I mentioned, reviewing has other benefits as well. It's not a line item in my contract, but it's something my employer expects, and it also useful for me as an academic to communicate with the literature in this way. So it's similar to mentoring in some ways. 2021-03-02 16:18:13 @neuralengine @rmgrieves @GunnarBlohm There are many other things not in my contract that I don't expect to be able to bill for. Mentoring students, being on committees, etc. 2021-03-02 16:16:37 @ElDuvelle @rmgrieves @GunnarBlohm I am being paid, this is part of my annual reporting to my department. 2021-03-02 15:23:35 @rmgrieves @GunnarBlohm Because you will have a widened gap between the free and the moneyed journals. And the moneyed journals will continue to gain in prestige as they become more expensive. 2021-03-02 15:14:01 Men, please stop trying to correct women all the time. It's embarrassing at this point. https://t.co/R1CrzIvmJZ 2021-03-02 15:12:15 @rmgrieves @GunnarBlohm I don't understand how you view this as a win. That is a clear path to making academia even more class-stratified than it currently is. 2021-03-02 15:11:35 @rmgrieves @GunnarBlohm It's only insane if you view it as entirely a chore. I get a lot out of reviewing. I learn about the field, I help to teach other authors my perspectives on good science, I improve the quality of the literature that I will be publishing in. It's also part of my job description 2021-03-02 14:51:13 @GunnarBlohm @rmgrieves And it makes these papers less accessible to the less funded portions of academia (either to publish or to read, depending on the access model). Basically, they will find a way to skim more money, and we will pay for it. What we get in reviewer pay will not equal the new fees 2021-03-02 14:36:38 @rmgrieves @GunnarBlohm Do you seriously think they are just going to eat that cost out of their profit margin? 2021-03-02 14:19:29 @rmgrieves @GunnarBlohm I think the goal should be reducing the influence of money for those profit-seeking journals. Paying reviewers just for those journals is going to increase their budgets, their admin overheads, and make them seem even more prestigious. 2021-03-02 05:09:12 @neuro_data @tyrell_turing @KordingLab Yea, that was my first interpretation, but when I read down the list I decided that they might have a different interpretation of a theory's truth value than I had thought. 2021-03-02 03:15:57 @tyrell_turing @neuro_data @KordingLab I think the alternative on that branch is to suggest that theories are fundamentally disconnected from reality 2021-03-02 03:15:03 @tyrell_turing @neuro_data @KordingLab I think of truth in terms "reflecting reality" and theories that we devise have the potential to be true, although, which is why I go the "yes" route on the Instrumentalism branch. This doesn't imply that we have actually found any true theories yet, it just means they exist 2021-03-01 23:18:40 @SBMost No, thank you!! You gave us quite a lot of references to follow up. Let's see where this goes 2021-03-01 14:11:22 @neuralreckoning @GunnarBlohm Yea good point. 2021-03-01 14:07:26 @Jonna_Singh_ @neuroecology @SfNtweets @eLife This would be cool but the problem of oversight is non-trivial. Permanent public spaces online have a tendency to attract problems. 2021-03-01 14:01:56 @neuralreckoning @GunnarBlohm Doesn't that mean more free labor donated to labs that are already popular? 2021-03-01 13:42:53 @GunnarBlohm I don't see how this can be addressed without a drastic decrease in the amount of publications. Or an enormous amount of funding to hire code reviewers. 2021-03-01 13:14:23 @SBMost Yep! Thanks again for your help! 2021-03-01 12:57:58 @SBMost I like it Steve! 2021-03-01 12:57:08 @filippie509 @DrDorotaSK Filip why are you assuming that you understand the situation she's facing when you are not in her situation? 2021-03-01 05:17:00 Excellent advice from Talia, as usual https://t.co/oR0fuonFTk 2021-03-01 04:50:46 @filippie509 @srchvrs Is it possible you're underestimating the gender bias in hiring? Or was that your point? 2021-02-28 13:44:22 Historic dunkage on a transphobe https://t.co/8BGz6CFUzY 2021-02-27 20:18:36 @Hyatt shows that they do not understand the dangers of allowing Nazi beliefs and symbology to be used. @slpng_giants https://t.co/0T7X4BdEa5 2021-02-27 14:39:23 RT @Neurocircuits: Postdoc position is available in my lab, supported by a new 5-year BRAIN Initiative R01. Looking for a neurophysiologist… 2021-02-27 14:20:48 @E_Bonawitz @AlisonGopnik https://t.co/T08GERTFju 2021-02-27 14:15:48 RT @AndyPerfors: Hi everyone. I wrote a blog post with my thoughts on the recent kerfluffle involving Holly Lawford-Smith, the UniMelb prof… 2021-02-27 14:14:54 @tsawallis @AndrewHaun3 For sure, childhood fascination with science is amazing. 2021-02-27 05:23:49 @timmytimmytim Mom, why can you remember 3.7 objects but I can remember only 3.5? 2021-02-26 20:41:01 RT @jakhmack: Champalimaud Research Symposium #CRS21 on #neural and #artificialintelligence-- will certainly happen in October, hopefully n… 2021-02-26 15:42:14 RT @AriDrennen: Not one single trans person is quoted in this @nytimes article about the Equality Act passing today, but they did give thre… 2021-02-26 14:13:09 I think I was 6 years old the first time I started to get really excited about the neurocognitive mechanisms of working memory consolidation. https://t.co/3gQJ9c4bTt 2021-02-26 05:44:52 @JoshuaGrubbsPhD Uh that's a toy potato, it's totally different. 2021-02-26 05:30:13 @KordingLab @azhir_io @Raamana_ @andpru @tyrell_turing @prokraustinator It truly is (in some corners). I'm very glad to have gotten to know all of you. 2021-02-26 01:02:14 @JhendersonIMB My secret is that when one computers browsers get too full of, tabs I buy another computer 2021-02-26 01:01:13 @anne_churchland @NeuroPolarbear If that's the rule then the chemistry kids that used to come with radioactive isotopes back in the 1920s should be extremely illuminating 2021-02-25 19:21:39 @KordingLab @CousinAmygdala @PsychScientists Yell out "the brain is not a computer" and within a few moments they will all have torn one another to pieces. 2021-02-25 13:24:38 RT @drosophilosophy: oh boy, the viewers at home are really gonna write in on this one. i suppose, trigger warning? this is about body imag… 2021-02-25 12:07:14 @aneeshmetanoia @neuromatch I've got teaching this morning but I'll aim for 9 for a bit 2021-02-25 12:04:31 @aneeshmetanoia @neuromatch Sure isn't, I'd be happy to meet. When are you there today? 2021-02-25 12:01:19 RT @AndyPerfors: Sorry to keep banging on about this terrible transphobic website, but one other thing you can do if you're looking for how… 2021-02-25 12:01:14 @AndyPerfors Signed! I can't belief this is launched by a faculty member. Wtf 2021-02-25 04:25:04 Narrator: and that's how we powered humanity's first intergalactic voyage https://t.co/MYk1YwXRqn 2021-02-25 04:17:10 RT @marceelias: The fact that we are about to be hit with a tidal wave of voter suppression legislation by Republican legislatures througho… 2021-02-25 02:48:09 @tdverstynen Yea, seriously. 2021-02-25 02:47:34 @AndyPerfors @tdverstynen Awesome! 2021-02-24 23:43:52 RT @meganakpeters: friday morning i get to join these fantastic people to chat about how we can make science -- conferences, publications,… 2021-02-24 21:34:24 RT @neuro_data: #LearningSalon is thrilled to host @katjahofmann this Friday at 4PM EST on https://t.co/lBK8aSHglb who will be speaking abo… 2021-02-24 13:30:01 @SSarabipour @Merz @AHollandLab @iaincheeseman @neuralreckoning @AnnaHscientist @DORAssessment I'm not saying that we shouldnt work on it, I just don't see how OPR solves the problems. But then again I also wasn't very optimistic about virtual conferences until we were forced to have them 2021-02-24 04:12:55 Are you ok @RonJohnsonWI? Tell us how we can help you. https://t.co/qVickOmLGZ 2021-02-24 04:11:38 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning @AnnaHscientist @DORAssessment Which is not to say that there is no better solution, I just haven't seen evidence of good solutions lurking around the corner (unlike online conferences which have obvious advantages to a wide range of problems) 2021-02-24 04:10:45 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning @AnnaHscientist @DORAssessment I'm more with Alex when it comes to peer review. I don't think that open PR or post-pub-Peer-review are effective solutions to the big problems in our literature. Pre-prints are great, but I don't think we can develop a post-pub review system. The incentives just aren't there 2021-02-24 01:41:38 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning The advantage of the last year is that it is forced us to experiment with alternative models rather than doing so just out of protest against th existing system 2021-02-24 01:40:08 @AndyPerfors @spiantado It's a really good point. I think that only 20% but the effort I put into a paper is there just to appease the reviewers 2021-02-24 01:23:36 @spiantado @AndyPerfors Exactly. And then when you rewrite the introduction you realize you should have run a different experiment. 2021-02-24 01:22:51 @AndyPerfors @spiantado It's so tricky, I think tweets + figures+ code is like 90% of the useful content of our recent working memory model. The problem is that getting to that place required a lot of pressure from peer-review. 2021-02-24 01:19:45 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning It's a hard set of problems for sure. The easiest way to get started is by innovating positive models that will start to shift the needle on how online events are perceived. 2021-02-24 01:17:34 @AndyPerfors @spiantado The one thing that I do like about paper writing is that it forces a lot extra effort, which is often where unanticipated thoughts/discoveries happen. My v1.0 of any paper is way less informative than even the first pre-print I end up submitting. 2021-02-24 01:13:25 @SSarabipour @neuralreckoning @Merz @AHollandLab @iaincheeseman I mean it's kind of like a form of subsidy of local economies. which isn't the worst thing in the world, it helps to pump money into service industries all over the world. But that doesn't mean that that money wouldn't be better spent in other ways, opportunity costs and all 2021-02-24 01:06:43 @SSarabipour @neuralreckoning @Merz @AHollandLab @iaincheeseman 1.3 billion dollars every year for conference attendance just in the physical sciences.. wow. My mouth is agape. 2021-02-24 01:00:11 @SSarabipour @Merz @AHollandLab @iaincheeseman @neuralreckoning Thank you for this paper! It's a valuable resource for us. 2021-02-24 00:56:05 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning Agreed, but is anyone seriously making those claims? 2021-02-24 00:55:19 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning nonsense, these talks are no substitute for papers, either in terms of dissemination, or career building. 2021-02-24 00:53:40 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning COVID has certainly been a forcing factor, and it's clear if R0> 2021-02-24 00:52:22 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning These talks are still being attended regularly, 4 months after they were given. We shouldn't compare these events pound for pound, they have other benefits that are being discovered. We have no idea what else will accrue from the new format as we drift away from the legacy model. 2021-02-24 00:49:29 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning We've also run 3 Neuromatch conferences last year. The third one had 800 talks, and 600 of them are now on Youtube, indexed and available for use by anyone. 2021-02-24 00:48:20 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning I do agree with you that online conferences are usually less immersive, but this is counterweighted (and then some) by the ability to bring in people that would not otherwise have a chance to get access to these resources. E.g. in Neuromatch, we had > 2021-02-24 00:45:26 @Merz @SSarabipour @AHollandLab @iaincheeseman @neuralreckoning We ran Neuromatch last year which taught comp. neuro to 1700 students for a modest fee that was voluntarily waivable. We had > 2021-02-23 19:48:42 RT @talia_konkle: YES. https://t.co/F8bBkzrArQ 2021-02-23 19:36:55 @mcxfrank @Nate__Haines @RickCarlsson Yep, that's how it came across to me. 2021-02-23 19:13:08 @Nate__Haines @RickCarlsson @mcxfrank I agree that there's a drive to frame the pro-modelling argument in this prescriptive way, but it's not the position anyone is advocating for. I.e. doing modelling is not a policy directive. Whereas in many cases (like this post) pre-reg is a described as a policy directive. 2021-02-23 16:34:46 @PsychScientists @azhir_io @mcxfrank The word resonates in different ways with different people, depending on how their self-image. Some people are probably pretty fed-up with being considered as radicals. This is why I didn't understand the choice. 2021-02-23 12:38:47 @NunezKant Congrats!! 2021-02-23 01:30:40 @mcxfrank @IrisVanRooij @PeterKvam @zerdeve My terminology was confusing. By marginalized, I was referred to the difficulty that modelling approaches can face because they're outside the norm of conventional approaches, as Peter mentioned in the OP. 2021-02-23 01:26:40 @mcxfrank @IrisVanRooij @PeterKvam @zerdeve You're right, I shouldn't have framed it that way, it isn't helpful. However I wish I could convince you that these policies have the danger of being worse than the disease they aim to treat. 2021-02-23 00:20:52 @mcxfrank @IrisVanRooij @PeterKvam @zerdeve Consider how your lasts paragraph sounds: in the long run, we'll come around to your (radical) agenda, but for now, let us make our policies. Why do you think that's going to sound like making friends? 2021-02-23 00:16:44 @mcxfrank @IrisVanRooij @PeterKvam @zerdeve That's awesome (really!, I admire the work you've done in making such tools) but it's orthogonal to the point that these new policies can have harmful consequences on some ways of doing science (often ones that are already marginalized). > 2021-02-22 23:44:38 @mcxfrank @IrisVanRooij @PeterKvam @zerdeve This is still missing the point, the one group is telling the other how to behave, the other is trying to provide helpful tools. Relaxing the restrictions is not a compromise between these positions. 2021-02-22 18:39:56 @ipanalysis @o_guest @mcxfrank In science, there is a lot of implicit value in being perceived as in the center when it comes to accruing authorship and grants. (IMO) We talk about valuing novelty but rarely do we actually put that into practice. 2021-02-22 18:37:16 @wrychrd I'm sure that Tesla wouldn't charge more than $25/month for such a transmitter 2021-02-22 18:19:42 @mcxfrank I'm curious why you chose the framing "center vs radicals". It implies a certain proper perspective for the one group and seems to belittle the other. Why write it this way if your goal is to build bridges? 2021-02-22 16:17:12 @rolandVM hrm. How do you get to 1/7 with a Bayes model given the problem as framed? I don't think the probability of the orientation can be discounted given that the face-down component is explicit. 2021-02-22 16:15:43 @ira_hyman @itjohnstone @rj_redden @chazfirestone @XamilyGuy @ShaneFs5cents You got it right, 1/2. I don't think the problem is ambiguous as regards how many sides are blue, unless you assume that there are blue sides on the inside of the initial white cube, which is a stretch. 2021-02-22 14:55:00 @o_guest Btw, note how I didn't go into a deep dive on the etymology of the phrase "straight faced". https://t.co/D2OKaMkzLO 2021-02-22 14:53:15 @rolandVM I don't really agree. The randomness of which side is facing downwards is explicitly built into the decision outcome. And regardless, Bayesian accounts of cognition are supposed to account for explicit as well as implicit aspects. 2021-02-22 14:50:26 @o_guest ha ha yes, very 2021-02-22 14:37:41 @o_guest You allllllmost kept a straight face there... 2021-02-22 13:54:16 TIL that @JohnFetterman lived in the basement of a freezing church while converting it into a youth center. https://t.co/0DMzGjyY9P 2021-02-22 13:51:30 Sad to see people trying to redesign streets to be more convenient for self-driving cars and less convenient for pedestrians and cyclists. With enough money, we could see awful things like this picture become a reality. Note that bikes are not allowed to cross the intersection. https://t.co/e49WSOI7Lr 2021-02-22 13:45:08 @zpkilpat @chazfirestone @XamilyGuy @ShaneFs5cents Nice explanation 2021-02-22 13:43:26 @SebastianDeri @chazfirestone @BMtHart @XamilyGuy @ShaneFs5cents I think it's just that the odds of a one-blue cube showing 5 whites is 1/6, while the all white cube will always show 5 whites so this has to be factored into the probability. 2021-02-22 13:40:51 And this is why I'm not an advocate of Bayesian as a general approach to understanding cognition. Bayesian just kicks the can down the road in terms of figuring out how we make decisions 2021-02-22 13:29:20 It's interesting that the reason we get this one wrong is because we compartmentalize probabilities to make the problem more tractable. But The correct answer falls right out of a simple Bayesian approach. https://t.co/yFLheemM5K 2021-02-22 13:14:57 @azhir_io @aina_puce @datingdecisions https://t.co/vCQB7NXEEi 2021-02-22 13:01:57 @rj_redden @chazfirestone @XamilyGuy @ShaneFs5cents What I was forgetting is that it is very unlikely for a one-blue sided cube to land blue side down while the all white cube will always have 5 whites showing. So this probability has to be factored in. 2021-02-22 12:56:39 @sharoz @chazfirestone @XamilyGuy @ShaneFs5cents Ps. Yes I get it now 2021-02-22 12:54:46 @sharoz @chazfirestone @XamilyGuy @ShaneFs5cents https://t.co/moTo1mOa0H 2021-02-22 12:32:16 @chazfirestone @XamilyGuy @ShaneFs5cents If the answer isn't 1/7 I'll be interested to know why.. guessing it's not 2021-02-22 04:12:17 Spending $250 on a hardware upgrade so that your computer can display lowercase letters https://t.co/yv8sbQcRg4 2021-02-22 03:58:10 @nicholdav Yea Pashler is definitely one of the big names in attention. Very good intuitions on complex processes. 2021-02-22 03:22:10 @shengokai Having seen you at the Learning Salon many times now, I agree that this is accurate. 2021-02-22 03:21:14 @arankomatsuzaki @GaryMarcus 2021-02-22 03:19:45 @nicholdav This 1987 paper was one of the first papers that got me to start seriously second guessing the dominant paradigms in cog science. 2021-02-22 03:12:24 @nicholdav Yup. 20 some years prior. 2021-02-22 03:02:46 @timmytimmytim @nicholdav We need to assemble the attention Justice League and put this claim to the test. 2021-02-22 02:36:48 @nicholdav Check out this paper by Pashler from 1987 which has some similar ideas.. kinda. https://t.co/tTV8ssjdcF 2021-02-22 02:35:42 @nicholdav @neuromatch *ps* I know you're not crediting me for THAT idea, just one of the experiments.. but I couldn't resist the bit. 2021-02-22 02:33:52 Do we need attention to explain set size effects in search if DNNs can do simulate the same data with no attention/binding mechanisms? Great paper by @nicholdav Wondering what other data in search relies on an attention explanation? https://t.co/QO2vF1yGPJ 2021-02-22 02:31:12 @nicholdav @neuromatch Oh shit, you're crediting me for the idea of blowing up the attention-in-search framework? Honey we gotta go, now!! Get the dog! *packs bags* 2021-02-22 01:24:55 @datingdecisions Raise your hand if you've been told that diversity it's own little subcategory when it comes to improving science and teaching. 2021-02-22 00:25:39 RT @MGreenePhD: New preprint alert: Dynamic Electrode-to-Image (DETI) Mapping Reveals the Human Brain’s Spatiotemporal Code of Visual Infor… 2021-02-22 00:21:45 @tyrell_turing @GunnarBlohm @jmdelgadogarcia https://t.co/WXGbIZf8kz 2021-02-21 21:59:22 RT @OPAMconference: The deadline to apply to be an OPAM organizer is in one week! If you're interested in being part of the team, a researc… 2021-02-21 20:04:52 @hedderik @derxen I dunno, I think this is pretty sensible if drivers didn't ignore it. It seems safer to cluster both directions of bikers in the same lane than to split them across the road (which doubles the interactions with cars). 2021-02-21 19:43:06 RT @timnitGebru: I relate to this. But when you internalize the trauma & 2021-02-21 05:30:32 @NunezKant @GunnarBlohm @tyrell_turing https://t.co/3W7PbZn3xL 2021-02-20 23:44:30 RT @aneeshmetanoia: @neuromatch Any #neuromatch academy participants interested in a paid RA position at @Columbia ? DM me for more info… 2021-02-20 20:38:36 @derxen @hedderik Yes and this is how places over here in the Americas often approach the bike lane problem. Hence the tweet :) 2021-02-20 20:32:42 @andpru @PaulMinda1 @ScottMacShack Yea, perhaps we are. It's definitely harder on students who have to work from home in smaller spaces and with less good equipment (among other differences). I'm not suggesting work from home for everything, just that I see a advantages in zoom exams. 2021-02-20 20:31:01 @andpru @EllaBatty @ScottMacShack @PaulMinda1 Are you talking about at the after-party? 2021-02-20 20:29:39 @derxen @hedderik Yes, the bike lanes need to be accompanied by either physical or social infrastructure. Just allocating budget for a painted lanes in a city that isn't used to having them will not change behavior. 2021-02-20 20:15:04 @PaulMinda1 @andpru @ScottMacShack Oh yes, I agree there. That part is sorely missed. 2021-02-20 20:11:24 Painted bike lines are fucking useless. Pass it on. https://t.co/7E9VP7A1I5 2021-02-20 20:09:32 @EllaBatty @ScottMacShack @andpru @PaulMinda1 It might be better now that we've been in lockdown for a year? 2021-02-20 20:09:01 @PaulMinda1 @andpru @ScottMacShack That's true but on the other hand, no way was I giving an in-person talk in Austria in the middle of a teaching term. I think in-person lectures are badly distributed towards prestige, because those profs have more budget (personal and professional) and less teaching. 2021-02-20 20:05:12 @ScottMacShack @EllaBatty @andpru @PaulMinda1 Those are worse IMO. I would have thought it's better to be 100% zoom or room. 2021-02-20 20:03:17 @EllaBatty @andpru @PaulMinda1 @ScottMacShack Sure I can understand the preference for in-person. Ideally we'd let students choose. 2021-02-20 20:02:25 @andpru @PaulMinda1 @ScottMacShack You mean zoom based vs room based defenses? Sure I've done both. Maybe I'm obtuse but I haven't seen a clear disadvantage for the students in zoom. What do you mean by "dump stuff" in this context? 2021-02-20 20:00:46 Good thread on why we need to think about how to rebalance admission decisions so that we can fairly consider a wider range of applicants, especially 1st gens who haven't been preparing for grad school for most of their lives. https://t.co/7pvNd7O7XP 2021-02-20 19:50:17 @andpru @PaulMinda1 @ScottMacShack But in what ways is it worse? That's what I don't understand. 2021-02-20 19:27:11 @andpru @PaulMinda1 @ScottMacShack I'm curious why you think it's worse for them. I would have thought that for people who have some kind of anxiety about being in the same room with their examiners this might actually help alleviate that. 2021-02-20 13:23:48 RT @terrinwatson: Dear Black Scholars, Cornel West does not need Harvard University. Harvard University needs Cornel West. The lesson: K… 2021-02-20 13:18:28 RT @dmetaxak: Google recently added a feature allowing users to search for "Black-owned businesses near you". They're advertising it all ov… 2021-02-20 13:07:46 RT @katerodenhurst: If they had put the people who send out university alumni magazines in charge of Track and Trace we’d all be in the pub… 2021-02-20 12:52:02 @dlowd You freely admit to reviewing and criticizing a BERT-class language model, a very expensive piece of equipment 2021-02-20 02:17:12 @o_guest Agreed. One of these things is not like the other. 2021-02-20 01:51:29 Goddamnit. Please stop going down this path Google. https://t.co/TaWJAX85Gi 2021-02-19 23:47:50 RT @ilyseh: I am appalled by Joe Manchin voting against Neera Tanden for OMB, a committed progressive after he voted for Bill Barr and othe… 2021-02-19 20:38:57 RT @neuro_data: Starting in an hour https://t.co/HEzmHbsJDu 2021-02-19 19:38:02 RT @Chican3ry: Solidarity and lots of to random LGBT people and services who opt to jump on a grenade for the trans community on the regu… 2021-02-19 19:36:42 @sanjanacurtis Twitter: It's where bad ideas need to be explained. You can have that one for free jack 2021-02-19 19:33:41 @sanjanacurtis Isn't that how Twitter works? 2021-02-19 13:07:27 @AndyPerfors Twitter forced Ted Cruz to abandon a vacation in Cancun. 2021-02-19 05:17:47 @sarabwtroutman Sounds fair to me! 2021-02-19 01:22:20 @nicholdav @ODonnell_R_E Yea there's no better time to be there. 2021-02-19 00:54:22 @nicholdav @ODonnell_R_E There's nothing unusual about it. Ryan's been in lockdown in an Antarctic research base since last March. 2021-02-19 00:08:28 RT @math_rachel: I admire & 2021-02-18 23:48:12 RT @tage_rai: This thread is so lolsob https://t.co/0cdDFUM35D 2021-02-18 22:00:34 A beautiful illustration of the acute predictive capabilities of the human mind in a very high dimensional space. Not tweeting this for scoring political points, it's just science. ok maybe it's a little bit about politics... https://t.co/pUWxmSElcS 2021-02-18 21:58:04 @niais The fact that he thinks this makes him more relatable is just beyond... anything I can process right now. https://t.co/rJRXcPVH97 2021-02-18 21:54:29 @o_guest To me your background revealed was the breaking rays of hope that computational approaches bring! Seriously, it was great. I had hoped you'd talk for 30 minutes but you hung around for an hour a half and the time just few by. It was a hugely deep but accessible discussion. 2021-02-18 21:50:34 RT @hardsci: Meanwhile the first images are coming in from Perseverance Total Landscaping https://t.co/ClHDAWTwDE 2021-02-18 21:50:14 RT @MHendr1cks: Against Reproducibility: A thread. https://t.co/nAEY4OS8Ab 2021-02-18 21:16:29 We (@Shekoofeh93,@ODonnell_R_E, @j__tam) were so fortunate to have a chance to talk with @o_guest about her recent paper on building theories through computational modelling. It was loads of fun to talk with her and her perspectives were incredibly helpful https://t.co/zlVqcRsBXr https://t.co/Lu3Jpl9Dtv 2021-02-18 20:07:37 My mistake, it's apparently the fault of *checks notes* his children. https://t.co/fuxBwD3u7K 2021-02-18 19:34:21 @manes Or just the general difficulty of using ML to solve hard social problems. 2021-02-18 18:42:30 @azhir_io @tedcruz @TuckerCarlson @rcbregman A classic! 2021-02-18 17:56:54 I have it on good authority that AOC, with the backing of the Antifa, the Big Green New Deal and cancel culture were responsible for booking @tedcruz's vacation to Cancun. Why aren't you covering this @TuckerCarlson?? 2021-02-18 15:27:07 *stares in "yes, obviously"* https://t.co/TUBtWw8Nmq 2021-02-18 14:30:19 RT @ianbremmer: Greetings from Cancun Cruz https://t.co/wW4Xb3ndPJ 2021-02-18 05:31:15 @crozSciTech @patrickmineault Oh you said badge.. https://t.co/26LdxpZFDi 2021-02-18 05:15:31 RT @Helena_LB: The @neuromatch Neural Data Science course will run for 3 wks (approx July 5th-23rd), and the cost will be about $100 (fee w… 2021-02-17 22:24:14 RT @GunnarBlohm: This looks awesome! Congrats! https://t.co/XzPcY8RXeh 2021-02-17 04:08:37 RT @jeffbigham: @palvaro two hardest problems in computer science are (i) people, (ii) convincing computer scientists that people are the h… 2021-02-17 03:23:43 @tdverstynen Why are you trying to hurt America Tim? 2021-02-17 00:04:38 @tdverstynen Oo, do tell? 2021-02-16 23:43:52 RT @GeeDee215: Nextdoor be like: “I saw a brown lady in public. Be advised that this is happening.” https://t.co/NuuXrAeOok 2021-02-16 23:40:19 RT @BennieGThompson: The lawsuit filed today under the Ku Klux Klan Act of 1871, is to hold Trump, Giuliani, Proud Boys, & 2021-02-16 23:29:14 RT @socmathpsych: The Society for Mathematical Psychology is invested in providing an environment in which diverse participants may learn,… 2021-02-16 22:16:42 @sbloyd @BillRoper Smart! I'm glad to hear that you're doing ok. Sounds like some places are out just because capacity is reduced even if lines are intact so I hope that doesn't roll your way too. 2021-02-16 21:59:47 @sbloyd @BillRoper How are you doing on power? 2021-02-16 17:47:00 https://t.co/SplzXtKSyg https://t.co/6gyxpgiu5Y 2021-02-16 17:42:57 RT @KordingLab: The @neuromatch conference paper is finally out: https://t.co/F4LNB1XzPm We learned so much! 2021-02-16 14:34:36 Danielle reminding us (again) that we can and should do better at elevating awareness of the plight of the trans community in academia. Especially because some academics *actively promote* transphobia. https://t.co/njdmfAPoPN 2021-02-16 14:31:41 @timmytimmytim Thanks dark corners of the internet! 2021-02-16 14:02:53 This subreddit is a wonderful resource for people teaching sensation & 2021-02-16 13:55:33 RT @JenMsft: My brain refuses to believe there are 4 people in this photo https://t.co/GOwglY3vyw 2021-02-16 12:44:57 RT @meganakpeters: this beautiful, massive description of the content from @neuromatch Academy in summer 2020 is now up on OSF! special tha… 2021-02-16 03:13:27 @prokraustinator @MHendr1cks @michael_syglass Oh how I miss the 70's 2021-02-16 01:57:47 @MHendr1cks #3 all the way 2021-02-16 00:58:24 @djnavarro This is one of my favorites of this sub-sub genre 2021-02-15 23:33:40 @JohnCornyn Really? Is that what you think people are asking? 2021-02-15 23:02:49 @IrisVanRooij These are too good 2021-02-15 20:57:19 @matherion @krstoffr @dsquintana I don't think so, it must be missing a lot of parts of a full OS 2021-02-15 19:43:50 @krstoffr @matherion @dsquintana Isn't another option to release a complete VM with the software installed? Then it should run on any x86 architecture for a good 10 years 2021-02-15 16:08:31 RT @BFriedmanDC: Roads in Dallas are snowy and icy and the government is telling me to stay home or I might die. But this is infringing on… 2021-02-15 15:01:41 @MHendr1cks We are well poised to be world leaders in coal fired steam locomotion. 2021-02-15 13:53:53 @dsquintana You forgot the big one: personnel management. 2021-02-15 13:39:45 RT @andpru: Reflections on a recent process of finding a faculty position. Valuable readings for anyone on the job market. #AcademicTwitter 2021-02-15 12:37:43 RT @mattyglesias: 184,000 people in Oregon have gotten two doses — this is an absurd, socially harmful way to report the impact of a highly… 2021-02-15 04:06:55 @TheAtlantic haven't we spend enough time trying to understand these people? Can we just move on? You're only giving their ideas unnecessary, harmful airtime with pieces like this. https://t.co/UU6f7WwNxz 2021-02-15 04:02:19 @KordingLab @kashdogcicero @TaraViegen @pebenjamters @KriegeskorteLab @TimKietzmann @bethgelab @dyamins @summerfieldlab @cpilab Yep, that's the easy way. You just need a delay so that you can generate a prediction X milliseconds ago and compare against the present input. 2021-02-15 03:18:49 @kashdogcicero @TaraViegen @pebenjamters @KriegeskorteLab @TimKietzmann @bethgelab @dyamins @summerfieldlab @KordingLab @cpilab Predicting exactly when a given visual stimulus will move to a given location 2021-02-15 03:18:19 @kashdogcicero @TaraViegen @pebenjamters @KriegeskorteLab @TimKietzmann @bethgelab @dyamins @summerfieldlab @KordingLab @cpilab Ah, no I don't think the dorsal stream cares very much about binding. I would assume that It's more about predicting sequential information, particularly when self-motion is involved. 2021-02-15 02:56:31 @kashdogcicero @TaraViegen @pebenjamters @KriegeskorteLab @TimKietzmann @bethgelab @dyamins @summerfieldlab @KordingLab @cpilab How do you define temporal stability? I was thinking temporal precision more than stability. 2021-02-15 02:54:01 @TaraViegen @kashdogcicero @pebenjamters @KriegeskorteLab @TimKietzmann @bethgelab @dyamins @summerfieldlab @KordingLab @cpilab I'm sure it would work (somewhat) with different objective functions. E.g. in Dorsal you penalize for small errors in temporal reconstruction, in ventral you treat time more loosely. 2021-02-14 23:38:35 RT @ABC: JUST IN: Pres. Joe Biden calls on Congress to take action on gun reform on 3rd anniversary of Parkland shooting: "This Administrat… 2021-02-14 18:06:07 @PsychScientists She has strong opinions and often interjects. Many of us have had experiences with her on Pubpeer, facebook etc. I respect her right to object to the conduct of science but she is fairly upset about everything and it's often unproductive to talk with her. 2021-02-14 17:26:05 RT @skoularidou: I feel like stressing out again that I am determined to support & 2021-02-14 14:14:33 RT @Abebab: Facial recognition can do no such thing 2021-02-14 13:34:55 RT @ryanjreilly: So strange how early in the process there was a two-thirds vote to find Trump guilty and then they just kept “finding” all… 2021-02-14 13:05:13 RT @AkiyoshiKitaoka: Wi-Fi https://t.co/lZ9Bj47EFJ 2021-02-14 12:58:00 @NickFoxstats @chrisaberson @wgervais That doesn't mean it's not useful. 2021-02-14 12:57:17 @wgervais But that has already happened. Calling the witnesses wouldn't fundamentally change that. I think we're overestimating the impact of witnesses. Remember the Ukraine trial? Those witnesses should have been devastating but weren't. 2021-02-14 12:54:48 @wgervais @chrisaberson @NickFoxstats But he actually delivered. We saw the speech, it's recorded. 2021-02-14 06:32:18 @wgervais It's not that simple. Mitch gave them a speech they can use going forward in prosecutions and campaigns. The alternative could have been weeks of acrimony in which the R's just muddied the waters about who knew what. Don't assume witnesses would have produced a clean result. 2021-02-14 06:29:23 @NickFoxstats @chrisaberson @wgervais I think what they bought with move was mitch's speech and it's not nothing for to admit publicly that trump was responsible for all of it. That's arguably more useful to them than a war on the Senate floor in prosecutions 2021-02-14 03:44:59 RT @shantalrenee: .@AlisonLedg I remember realizing how the PhD was built for people very unlike me. Thought it would be easier to co… 2021-02-14 01:31:15 @sbloyd Those long jumps were a lot of fun. 2021-02-14 01:08:06 @GunnarBlohm Kinda, there was no real chance of a different outcome in all this, even if they had called witnesses. I think the dems probably did the right thing. They can start immediately on reforms after the recess and Trump can now be prosecuted with all of this new evidence available. 2021-02-13 23:43:21 RT @IrisVanRooij: Hey, I just noticed that @TrishaVanZandt is on twitter with a new account! Go follow her if you aren't yet. She does awes… 2021-02-13 23:41:36 @GunnarBlohm Impeachment wasn't the justice system, it's a completely different mechanism and set of rules. Trial is not about evidence but primarily about politics. The real courts probably aren't done with him yet. Stay tuned. 2021-02-13 22:36:53 @ravi_jayakumar I think it probably could. The Ragnaroc model is really driven by stimulus input but this doesn't preclude top down control. 2021-02-13 22:35:48 @neuralengine @neuralreckoning @azhir_io @JAlGallego @KordingLab @GunnarBlohm @venpopov yes but also includes real-time editing by multiple collaborators distributed around the world. That wasn't happening in 2005. 2021-02-13 22:35:13 @rudzinskimaciej @TaraViegen I don't think those RT=capacity findings tell you very much about how long the information bite is. Rather it's a correlation driven by some other cognitive variable. 2021-02-13 22:33:44 @neuralreckoning @neuralengine @azhir_io @JAlGallego @KordingLab @GunnarBlohm @venpopov No, but you can use a latek equation generator, convert those into PDFs and then insert the PDFs. Obviously clunky, but similar to inline latek equations in appearance. 2021-02-13 20:40:59 @ElDuvelle @KordingLab @neuralengine @neuralreckoning @JAlGallego @GunnarBlohm @venpopov If you use a tool that other people are reluctant to use, you will certainly be doing most of the job 2021-02-13 19:50:17 @tyrell_turing How can you reconcile that belief with a 5 year old's insanely acute ability to figure out exactly how to push their parents' buttons to get what they want? 2021-02-13 19:26:02 @azhir_io @neuralreckoning @neuralengine @JAlGallego @KordingLab @GunnarBlohm @venpopov Overleaf does offer this, although in my experience it's been a bit clunky as an interface. 2021-02-13 18:42:34 @neurobongo That's not fair, the interface is ideally suited for people who like preferred shows to be hidden among a sea of other shows they haven't watched yet. 2021-02-13 18:27:17 RT @futuredoctorate: https://t.co/h2Q6I4wQRr 2021-02-13 18:13:34 @PaulMinda1 @kimberlyquinn Let's just agree to use what school they came from. No bias there. 2021-02-13 17:52:03 @tyrell_turing So you don't think 5 year olds have a causal understanding behind their speech? 2021-02-13 16:11:53 @rudzinskimaciej @TaraViegen I think it's a different mechanism. There are so many mechanisms at play here that it's nearly impossible to link one specific phenomenon to one parameter. 2021-02-13 15:33:45 @texesmom @cmclymer I'm guessing because they know she'd agree pretty easily and there's no chance of her lying about it. Then it becomes easier to ask McCarthy. 2021-02-13 15:26:06 Lordy I hope there are witnesses https://t.co/HtVCxyYk2O 2021-02-13 14:58:55 RT @jchristie: https://t.co/p2BWY8ZNkR 2021-02-13 14:29:22 @lisamurkowski are you ok with this? https://t.co/HRHqTxZOLn 2021-02-13 14:03:28 @kashdogcicero I'm so sorry to hear this. I don't understand people. 2021-02-13 13:45:14 RT @databoydg: QA redacted: but there was a really interesting question about how we can prevent our communities from being "fake/hostile".… 2021-02-13 13:30:20 @TaraViegen Great question, there is an inhibitory mechanism built into the LV layers that does this. 2021-02-13 13:29:17 @ChombaBupe @rudzinskimaciej I meant that the connections could probably be trained to more efficiently build memories "instantly". Not trained during memory formation 2021-02-13 12:52:26 @ChombaBupe @rudzinskimaciej Oh they definitely do. Optimization could be done. 2021-02-13 12:40:04 @azhir_io Thanks Azhir! 2021-02-13 12:37:16 RT @Commish_Schmidt: .@LeaderMcConnell The former POTUS incited supporters to threaten to kill my children and put their “heads on spikes”… 2021-02-13 05:07:00 @marcorubio Did you spend even 4 seconds thinking about the implications of saying that someone who has left office is immune from all repercussions? 2021-02-13 02:58:32 Here's the talk recording in case you want to see it. This talk explores 3 different models that address different aspects of making memories of visual input. https://t.co/9JVa6dWvut https://t.co/HlFBUpoaJb 2021-02-12 20:35:32 RT @neuro_data: #LearningSalon (hosted by @criticalneuro, @blamlab, & 2021-02-12 19:59:30 They're using words.. sometimes sentences.. it's the real deal! https://t.co/9s6Uydgyk7 2021-02-12 19:56:41 RT @JRubinBlogger: THIS! and if state bars do not police their own, states should take back the power of licensing and discipline https://… 2021-02-12 19:28:18 @TimKietzmann @jake_fisher https://t.co/Ki2WiEhZfp 2021-02-12 19:26:45 @neuralreckoning @Timothy0Leary I would say borderline inappropriate (unless there was a promise of payment) rather than unethical. If they want to farm their program requirement checking out to other people, they should pay for it. 2021-02-12 16:06:46 @VenkRamaswamy @TPVogels ha ha :) OBS studio is a freeware video streaming toolkit. Loads of fun (and also useful). 2021-02-12 16:01:18 @VenkRamaswamy @TPVogels I just shared the OBS virtual camera 2021-02-11 22:34:54 @andpru @TimKietzmann It's a lot less costly to the University to break the contract before arriving. 2021-02-11 22:11:01 @andpru @TimKietzmann Technically is there any difference to taking that other interview after having been in a contract for 3 years? It's still a contract. I don't see the distinction. I'd rather people left after 0 years than after 1 year. 2+ years you've had a chance to actually contribute. 2021-02-11 21:16:18 @FenCameron @pfau Working 90 hours In a week is not being productive. It's just working a lot. 2021-02-11 20:58:56 RT @will_ngiam: This is now online at Psychophysiology! "Estimating the statistical power to detect set-size effects in contralateral delay… 2021-02-11 20:58:04 @tyrell_turing @KordingLab @MattiaRigotti @neurograce We eventually wrapped our heads around it. It's a neat idea although it sounds a bit fiddly in the original incarnation 2021-02-11 16:36:35 @zoom_us has so many weird interface issues, not the least of which is that when you are leaving a breakout room, your mouse cursor is like 5 pixels away from a different button that would completely ruin your meeting by ending it. Can they please hire some UI people? 2021-02-11 14:54:32 @jfrickuga @andrewshtulman Please tell me about these "grants" 2021-02-11 14:12:37 I'm really looking forward to giving this talk! Thanks @TPVogels for setting it up. https://t.co/IzQnRNyHeX 2021-02-11 13:34:53 @dannysullivan @databoydg @timnitGebru Danny I think everyone here understands that the tab is the missing part. You are doing a pretty bad job of engaging with and addressing the concerns. A person in your position shouldn't start with a defensive posture, especially for a contentious issue. 2021-02-11 13:17:53 @jfrickuga @andrewshtulman https://t.co/bay0SZK2wL 2021-02-11 13:16:16 RT @dlowd: No, you don’t have to work ridiculous hours to do meaningful work in AI, and it’s harmful to normalize the idea that long hours… 2021-02-11 13:13:32 @tyrell_turing @nicholdav But sleep is just another form if behavior isn't it? 2021-02-11 05:21:35 @djnavarro No, it doesn't look good in the short term. You're right that this issue needs more attention. It's too far down on the progressive agenda, especially given the strong pushback it's receiving from feminists. 2021-02-11 05:08:46 @djnavarro Oh god, I'm sorry Danielle. This is awful. 2021-02-11 04:24:44 @hannah_merseal @sarabwtroutman https://t.co/LosBjFoc1N 2021-02-11 03:53:36 @RCownie @rasmansa I used to really enjoy his scholarship on language. Now I don't think I can really trust his perspective on anything. 2021-02-11 03:47:03 @rasmansa https://t.co/SDrBD0ybqx 2021-02-11 03:43:08 @rasmansa Wasn't he the one who suggested a debate? 2021-02-11 02:44:12 @davidskemp @_danilo @BerkeleyLaw Any idea where they found this? It seems to be an old Dell filter. 2021-02-10 21:04:44 @J_D_Crawford @andpru @chazfirestone @action_brain The same was true in perception science was very much about auditory processing until computers became common and now we've moved towards vision. It's not that vision is better/more interesting, it's just more flashy and technologically accessible. 2021-02-10 21:03:51 @J_D_Crawford @andpru @chazfirestone @action_brain Yea I completely agree that perception has been prioritized unnecessarily in cognition and to our detriment as a field. I suspect this was due to technological factors (i.e. it was easier to precisely control stimuli than to measure movement until recently) > 2021-02-10 20:33:32 @andpru @chazfirestone @action_brain @J_D_Crawford That said, I do see a number of tweets in this thread that lean pretty heavily towards motor/action chauvinism, so maybe I wasn't overinterpreting. 2021-02-10 20:22:15 @timmytimmytim Most, best, kind regards Sincerely, Thanks! 2021-02-10 20:20:31 @timmytimmytim Sincerely, Tim 2021-02-10 20:19:16 @andpru @action_brain @J_D_Crawford IMO, this is where the case is overstated. One can build abstract representations for some domains that are not tightly linked to specific actions, but instead provide a broad based set of capacities that can be flexibly employed. 2021-02-10 20:17:12 @DrGBuckingham @J_D_Crawford @andpru @action_brain Yea, I don't think anyone doubts that this is true from an evolutionary perspective. The question is whether we should shape our research agendas around that fact. 2021-02-10 20:16:05 @andpru @chazfirestone @action_brain @J_D_Crawford Yes I think maybe I was just reading too much into the initial post here about what is meant by primacy. In the typical version of this debate there is implicit comparison about which perspective is more "important". So I may have overinterpreted what was meant by "primacy" 2021-02-10 18:00:07 @andpru @action_brain @J_D_Crawford e.g. in an autoencoder, the compressed latent space is more economical for reasoning/deciding than the L1 layer. Similarly, I don't build models at the subatomic level because it's not the best representation. I think the brain also chooses efficient representations 2021-02-10 17:56:39 @andpru @action_brain @J_D_Crawford Yea I've heard these arguments and they're not convincing that, in all cases, action is the primary way of framing representations. The fact that X is causal for Y through some succession of transforms doesn't mean that X is the most efficient representation for processing > 2021-02-10 17:20:30 @cmacsbrain @andpru I think one should view it as a continuous process of update. Action gives rise to perception, perception gives rise to action.. But I think in some cases the action or perception component is more or less important. 2021-02-10 17:19:03 @criticalneuro I think they used the latest version of Adam. 2021-02-10 16:08:54 @andpru I would be similarly opposed to a framing that makes perception primary as well. I don't think it's helpful to mandate a given perspective 2021-02-10 16:02:48 @andpru You lost me at "has to". Many cognitive processes are not closely tied to action. For example reading. Sure there's eye movements, but the construction of a narrative from reading is not necessarily related to action, even imagined action. 2021-02-10 15:40:47 Here's a great tutorial video on machine learning. https://t.co/V2mq5NAzgQ 2021-02-10 15:38:35 @mccormick_ted That is so disappointing. I used to like McWhorter for his takes on language evolution and history but he has apparently fallen down some pretty deep holes. 2021-02-10 15:22:52 @andpru He's right of course that muscles came first, but does that mean this primacy question is useful for framing perception and action in us? I don't see the use of this "which comes first?" question. It's a loop. 2021-02-10 02:36:42 RT @blamlab: Exciting line-up @criticalneuro @neuro_data https://t.co/Lh39jtB2Lh 2021-02-10 01:26:19 RT @abidlabs: I'm shocked how hard it is to generate text about Muslims from GPT-3 that has nothing to do with violence... or being killed.… 2021-02-10 01:25:11 RT @o_guest: So so excited this is out! @Abebab is wonderful to work with. Please check out our paper, if you have not! 2021-02-09 23:09:20 RT @Timodc: The GOP run senate went on vacation for 2 weeks after the capitol was stormed and a cop was killed while the perp was still con… 2021-02-09 23:07:31 RT @NBCNews: WATCH: To begin impeachment trial against former President Trump, House impeachment managers present a 13-minute video montage… 2021-02-09 18:45:02 @will_ngiam @ira_hyman Right, I think we can think of there being fast/automatic labels (it's a 9!) and slower labels that have to be created with controlled processes. But once the labels are created/stored, access is fast. 2021-02-09 18:33:11 @will_ngiam @ira_hyman Right, so maybe they develop a new label on the fly, a rapid form of compression. This is something we hadn't considered, but it could be a faster form of learning than full-on perceptual learning (thousands of trials e.g. Pelli et al. experiments) 2021-02-09 18:31:39 @will_ngiam Wonder if this is what the CDA entails? (i.e. repeated operations to pack and unpack) while for stimuli that can just reside in memory passively without this cycle, there's no CDA. 2021-02-09 18:25:50 @will_ngiam @ira_hyman In the model, you can skip the back and forth process and just set the feature encoding parameters to 0 if you know that the labels are all that you need. We think subjects do this in our 30 trial experiment we mentioned. 2021-02-09 18:24:32 @ira_hyman @will_ngiam Sort of! It doesn't involve learning though, so it's limited. 2021-02-09 18:23:59 @will_ngiam I think @Shekoofeh93 will! I'll be talking about the attentional blink, revisiting more familiar ground.. 2021-02-09 18:23:06 @will_ngiam Yes, in the binding pool, unpacking is basically the reverse of the packing, and each time you do it, subtle information would be lost, though labels could be retained. Perceptual grouping and associations: currently out of scope. The model needs a bigger retinal canvas. 2021-02-09 18:15:50 @will_ngiam Wow, that is a really good idea. Thanks! 2021-02-09 18:14:11 @will_ngiam Hi Will! The token is what clusters features together into an object.. I think it would be awesome to show that with repeated encoding/retrieval operations on the same stimulus the labels remain strong while the features degrade. I suspect that happens... 2021-02-09 17:57:13 @PaulMinda1 @graemedmoffat @MHendr1cks @tyrell_turing That looks like one of those cookie casseroles, with either blueberries or chocolate chips 2021-02-09 16:58:43 @tyrell_turing @MHendr1cks I think this is because we don't really understand high-dimensional problems so they seem low-dimensional to us. And therefore it is easy to imagine that they are similar. 2021-02-09 16:57:48 @tyrell_turing @MHendr1cks The more I think about it, the more absurd it is. It's as if we view all problems in a high-dimensional space as being more similar than Low-d problems we understand, when in reality, locations in high dimensional spaces are LESS similar 2021-02-09 16:47:47 @AlonZivony Panicking and redoing all your analyses is what marks you as a good scientist. Others might wave their hands and assume that it's probably fine. 2021-02-09 15:58:05 RT @DemocracyDocket: Trump’s second impeachment trial begins TODAY at 1:00 PM ET. Watch our feed for LIVE updates throughout the day. In t… 2021-02-09 15:29:46 @psmaldino @glupyan @Slate WTF indeed! 2021-02-08 04:26:40 RT @Amy_Siskind: The Chiefs should just say they won anyway. Maybe Ted Cruz and Josh Hawley can help. 2021-02-07 23:19:48 @KordingLab @GunnarBlohm I think it's about these: https://t.co/ta1p4rztHL 2021-02-07 23:17:37 RT @ferrisjabr: A thread of very good, wonderful, truly Super Bowls. Translucent agate bowl with ornamental grooves and coffee-and-cream m… 2021-02-07 21:27:09 @o_guest @drussellmrichie I've got it down to 120 bytes, I've got 8 whole bytes left! I can add so many more features. 2021-02-07 20:18:14 @o_guest @drussellmrichie Yea, your entire game state was saved in < 2021-02-07 20:17:01 @o_guest @drussellmrichie And that is not a typo. 2021-02-07 20:16:44 @o_guest @drussellmrichie 128 bytes of RAM is all they had to work with. 2021-02-07 20:14:02 @djnavarro @richarddmorey Zaxxxxxxon, so good. And incredibly difficult to track your position relative to those missiles. 2021-02-07 20:12:57 @o_guest @drussellmrichie I'm this many years old https://t.co/mczfZFuL3m 2021-02-07 17:11:15 @LaurenHallion @Raamana_ System1 and system2 makes some sense to me. System 1 is like the first, primary system (both evolutionarily and in terms of proximity to the control pathways), and system 2 is the elaboration. Type I and Type II is the worst ever. 2021-02-07 04:12:51 @todd_gureckis That was amazing! What a great demo of what all of us could be doing 2021-02-07 04:08:33 RT @todd_gureckis: i spent 48 hours watching youtubers explain how to make content that is engaging so as to get lots of subscribers and th… 2021-02-07 03:30:43 @glupyan @susanwagnercook @AndreiCimpian Though as has been pointed out many times, there were plenty of people in the 1960's who didn't have these views. 2021-02-07 03:13:44 @talyarkoni @andpru @KriegeskorteLab @aina_puce @JSchleiderPhD @MHendr1cks Again not necessarily, could be lots of ancillary effects 2021-02-07 02:24:07 @talyarkoni @andpru @KriegeskorteLab @aina_puce @JSchleiderPhD @MHendr1cks it's impossible to measure reliability in the context of the decision process I just described. We classify grants into a category and then adjust them across the threshold afterward. And yes, it impacts a minority of proposals. Why is that an issue? Isn't this a good thing? 2021-02-07 02:13:18 @talyarkoni @andpru @KriegeskorteLab @aina_puce @JSchleiderPhD @MHendr1cks I'm telling you, at the NSF, it's not about scores moving, it's about deciding how to classify grants in the grey zone. I have literally seen the decisions arising through discussion with my own eyeballs. I'm not sure what else I can provide. Video recording is not allowed 2021-02-07 02:07:46 @neuralreckoning @ampanmdagaba @tyrell_turing Yea I got to 11 just by viewing it larger on a full monitor. 2021-02-07 02:04:26 @talyarkoni @andpru @KriegeskorteLab @aina_puce @JSchleiderPhD @MHendr1cks It's possible that I read that opinion into your position given that you spent a day on this, but I've seen you spend more time on ice cream so I shouldn't have done that... 2021-02-07 02:02:09 @talyarkoni @andpru @KriegeskorteLab @aina_puce @JSchleiderPhD @MHendr1cks I don't want to say that though. I think we have no idea of what the real benefits (or costs) of discussions at panels are, although the Canada experiment suggests that their abrupt removal causes the sudden expiration of the subject. 2021-02-07 01:59:35 @KordingLab @neuroecology @kashdogcicero Seems so. 2021-02-07 01:58:39 @talyarkoni @andpru @KriegeskorteLab @aina_puce @JSchleiderPhD @MHendr1cks So why aren't you railing against them with the same conviction as you are against panels? 2021-02-07 01:57:20 @neuralreckoning @andpru @talyarkoni @KriegeskorteLab @aina_puce @JSchleiderPhD @MHendr1cks Yea I know, my point is that panel discussions are .0001 of the footprint of conferences and have a higher SNR in terms of discussing science. I'm just wondering why the panels are the thing we should be worrying about if we doubt social interactions as a useful aspect of science 2021-02-07 01:24:32 @andpru @talyarkoni @KriegeskorteLab @aina_puce @JSchleiderPhD @MHendr1cks I also don't understand why discussion panels are so objectionable. Surely getting together and talking has a variety of benefits and it's not that expensive compared to e.g. conference travel. Why are they worth it? Where's the peer reviewed evidence that we should give talks? 2021-02-06 23:05:19 @JohnKubie @NinaStrohminger hrm... yea me too actually. 2021-02-06 23:00:24 @JohnKubie @NinaStrohminger Who are you gonna believe. John Kubie, or the New York Times? 2021-02-06 22:01:42 RT @Fair_Decoy: The reactions to this are fascinating. I love traveling & 2021-02-06 21:51:54 @RhiannonDauster Ouch.. to even think this. Let alone say it out loud Let alone say it out loud to a conference of educators 2021-02-06 21:15:28 @talyarkoni @andpru @aina_puce @KriegeskorteLab @JSchleiderPhD @MHendr1cks I think that is one of several different benefits of panel. Science is, and should be, a community activity. Panels are one of the ways that people assemble to corroborate our intuitions and sharpen our review process. It is hard to know all of the things this entails. 2021-02-06 18:47:26 @R3RT0 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD This illustrates why it's so hard to measure "good". #papers is at best an incredibly noisy measure of better science. It may even be anticorrelated. It is much easier to determine empirically that e.g. trepanning is useless because outcome variables are better defined. 2021-02-06 18:01:39 @R3RT0 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD Can you point me to some examples you consider good and useful? 2021-02-06 16:54:32 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD I'm not opposed to measuring it. I'm worried about saying: panel discussion is worthless if we can't show improvement in the data so let's just average the numbers together and apply a threshold. 2021-02-06 16:51:41 @aina_puce @talyarkoni @KriegeskorteLab @JSchleiderPhD That's the issue really, we have no idea of 10 year let alone 50 year outcomes in terms of scientific progress. And absent time machines and parallel universes, we never will. 2021-02-06 16:50:25 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD We have much better data on Covid mitigation than we have on grant review policy outcomes. These aren't really comparable. 2021-02-06 16:48:55 @R3RT0 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD Figuring out the brain is a centuries/millenia scale problem. And that's fine, we all know what we're in for. 2021-02-06 16:10:13 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD Maybe I'm just making excuses for myself. 2021-02-06 16:08:13 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD I guess over the past few years I've started to partition some things into a category that is so difficult to measure we're better off using our personal ethical and moral frameworks. 2021-02-06 16:01:09 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD Correct, I don't know how to *measure* which of those is better with e.g. a 50 year time horizon on scientific progress. I just have my opinion. 2021-02-06 15:56:35 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD But the point is is that it wasn't the reviewers who were driving the ECR push here It was the program officers. 2021-02-06 15:55:31 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD I guess that I'm not asking anyone else to believe me. I'm trying to solve a select group of problems in my life and field of study, not improving the grant review process. Therefore I probably shouldn't have bothered speaking up since I don't have more evidence 2021-02-06 15:47:11 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD I don't think "improve the process" is something we can objectively determine. 2021-02-06 15:46:08 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD I was in the room and saw the decisions happen. 2021-02-06 15:39:24 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD I'm happy that there is a diversity of approaches to this complex problem. 2021-02-06 15:38:27 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD I didn't say it helps in some concrete sense, I said it affects some career outcomes by shifting money to younger people. Whether that is a " good" thing is hard to measure objectively but I'm happy with it. 2021-02-06 15:33:42 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD Since I doubt that we have any useful metrics of the best way to evaluate grants in a long-running sense I'm happy to go with opinions and what we should do. Boosting early career people seems like a good idea to me. 2021-02-06 15:30:59 @neuroecology @kashdogcicero I was actually thinking the opposite. It seems like number of followers has a nonlinear excitatory effect on the likelihood of a tweet taking off. 2021-02-06 15:29:17 @talyarkoni @aina_puce @KriegeskorteLab @JSchleiderPhD How do you define helps versus hinders? 2021-02-06 15:08:11 @aina_puce @talyarkoni @KriegeskorteLab @JSchleiderPhD That discussion contextualizes certain grants and also works to upweight junior people or those who face other challenges for reasons that might not be obvious otherwise. 2021-02-06 15:05:40 @aina_puce @talyarkoni @KriegeskorteLab @JSchleiderPhD At NSF, the discussions have little effect on the majority at the ends of the distribution but have a huge influence on the ones close to threshold. Would look like noise in aggregate but is huge in terms of its effect on careers and direction. 2021-02-06 14:44:18 @samuelmehr It's back up 2021-02-06 14:15:56 @samuelmehr I guess we did. Stay tuned I guess. 2021-02-06 14:03:24 Here's the T-index, which might be this weekend's science-twitter obsession (compute yours from a link in the first reply) https://t.co/GPSd9ANG3u 2021-02-06 13:18:09 RT @shengokai: Here's an angry job market tweet for you: despite numerous reports that the global pandemic has affected academic production… 2021-02-06 13:15:21 RT @scalzi: Christopher Plummer did it. Now it's your turn. https://t.co/EXkZMtL41V 2021-02-06 13:08:32 This is a great natural illusion. See if you can figure it out before the end. https://t.co/vyKJsnO4XN 2021-02-06 03:04:55 RT @KT_So_It_Goes: RIP lou dobbs https://t.co/uxbHgLNnpZ 2021-02-06 03:03:34 @itjohnstone @talyarkoni This is me thinking about all the times I've had this debate. https://t.co/nZlFGQbpBS 2021-02-06 02:50:15 @kashdogcicero You might want to add a short warning to let people know that there will be a lag between pressing Add and the update. I suspect some people are adding themselves several times, assuming that the website has failed to update. 2021-02-06 02:26:39 @BuckmasterFresh @evopsychgoogle Yes it is 100% ridiculous. How did the interviewer not crack up when he brought up Legolas as a data point? 2021-02-06 02:23:46 @neuro_kim Thank you so much for keeping us all entertained and educated at the same time. 2021-02-06 02:21:00 @kashdogcicero Although the h-index isn't normalized at all. so... https://t.co/KOuo6fVO8I 2021-02-06 02:16:17 @kashdogcicero Now divide this by (#tweets * #followers) for an average quality metric? 2021-02-06 02:14:01 @talyarkoni All I'm trying to say is that some problems are easier than others. Evaluating subjective paper quality in a distributed system, across many domains, without paying the reviewers, and in a hurry. Multiply all those factors and you come up with "hard" 2021-02-06 02:07:23 RT @sm_haigh: Gosh these are pretty. Early Illustrations of the Nervous System by Camillo Golgi and Santiago Ramón y Cajal https://t.co/nD… 2021-02-06 02:06:53 @talyarkoni I refuse to believe that we can calculate the trajectory of a hurricane but can't calculate the pathway of a water droplet on a dirty surface. 2021-02-06 00:50:32 RT @mmitchell_ai: I am concerned about @timnitGebru 's firing from Google and its relationship to sexism and discrimination. I wanted to sh… 2021-02-05 20:39:47 @patrickmineault You were at that meeting on decaf? my hat's off to you for staying awake and being coherent. 2021-02-05 20:13:38 RT @neuro_data: Starting in 52 minutes....! https://t.co/zsoVca5gCn 2021-02-05 14:41:38 @lmauricecpr Yes 2021-02-05 13:55:51 @AthenaAkrami Great point Athena. It's interesting that convolution has become so baked into our concept of deep learning that I forget sometimes how deeply vision-centric it is. 2021-02-05 13:49:47 @tyrell_turing Don't tell me you're also not forgetting cite someone for the relationship between mass and measured weight??? 2021-02-05 12:39:21 RT @teemu_roos: Could we please fix an attribution problem? The nearest neighbor classifier was first proposed and studied by Evelyn Fix an… 2021-02-05 04:16:02 @TaliaRinger It takes a long time to found/develop a good reading group with colleagues. If you try to set one up, don't be discouraged if you spend a year before it starts to pay off with deeper discussions beyond methods critiques. Many people will drop out, don't take it personally. 2021-02-05 04:08:44 Thank you for writing this Claire! https://t.co/0aHBSNiFqB 2021-02-05 03:47:59 RT @justicedanielh: Fun fact: this is manifest bullshit. I review multiple manuscripts a year for academic presses and have never once seen… 2021-02-05 01:10:16 I'm really looking forward to this https://t.co/sXvuMOV6L6 2021-02-05 00:48:44 This is a powerful video. He digs deep into how harmful MTG's messages are. https://t.co/K7xwILwnV6 2021-02-05 00:37:53 RT @LeaderHoyer: No member ought to be permitted to engage in the kind of behavior that Rep. Greene has and face zero consequences. Tonight… 2021-02-04 18:02:16 @MHendr1cks @prokraustinator Eh, I dunno. This is just a dumb programmer error that could be made easily in any language. Matlab has spent a lot of time polishing their engine and there are some good arguments for using it in certain contexts. 2021-02-04 15:56:50 @Dr_Koraly I can't believe we don't have to pay for these features. It's so great. 2021-02-04 15:54:05 @Dr_Koraly Just spicing it up for you! 2021-02-04 15:53:38 @VenkRamaswamy No this is Mojave actually, but still. 2021-02-04 15:39:49 Me: Hi Mac! Mac: Hi! Me: Are you going to recognize my USB keyboard today or do I have to unplug it again? Mac: That's a great question! Me: ... Mac: ... Mac: No. 2021-02-04 14:39:33 RT @neurobongo: A nice naturally-occuring bistable percept https://t.co/uNistKjpzV 2021-02-04 05:42:03 RT @KordingLab: If there was an intensive, almost free, global, NMA style, TA led, driven by community, deep learning course this summer, w… 2021-02-04 03:56:34 RT @AkiyoshiKitaoka: The first paper has been accepted in Journal of Illusion! https://t.co/BaKbEY71Fy Kitaoka, A. & 2021-02-04 03:52:56 I have just hit the one millionth tweet that I cancelled halfway through writing while muttering "ah, I don't really care..." https://t.co/dahi5n4oaY 2021-02-04 00:53:57 This statement absolves someone of everything they did before taking office. What a bizarre policy. It sets a standard of zero accountability for behavior. https://t.co/KR4WNTUvBC 2021-02-03 17:45:11 @bjbalas @kimberlyquinn @stephenawurst I don't really a difference, both equally noseed 2021-02-03 14:46:27 @Abel_TorresM @KordingLab @NunezKant @elonmusk @xaqlab Sure but even then it's still easier to talk into a device. The noise of a shower is easier to filter out than the noise of a brain :) 2021-02-03 13:00:15 @Abel_TorresM @KordingLab @NunezKant @elonmusk @xaqlab There is no way that neuralink is going to beat voice output and recognition for a long time. The low hanging fruit there is to build better algorithms for decoding voice output based on context and learning. 2021-02-03 12:23:26 @Abel_TorresM @KordingLab @NunezKant @elonmusk @xaqlab The rate limit on reading a book is building memories, not reading the words. Speed reading works in the sense that you can process the words faster but comprehension isn't great. 2021-02-03 03:48:55 @andpru @KordingLab @azhir_io @rudzinskimaciej @xaqlab @NunezKant @elonmusk Agreed, assuming it's not a toxic work environment. 2021-02-03 03:37:56 @andpru @KordingLab @azhir_io @rudzinskimaciej @xaqlab @NunezKant @elonmusk Yes, that is an argument that I find somewhat convincing, if he delivers. 2021-02-03 03:36:54 @KordingLab @azhir_io @rudzinskimaciej @xaqlab @NunezKant @elonmusk I think I'd feel less annoyed by the whole thing if he didn't pretend that wireless recordings from a subdermal eletrode array are a novel thing. To be honest, even I didn't know at the time that competitors are mature, but figured it out with a bit of googling. 2021-02-03 03:05:26 @azhir_io @rudzinskimaciej @KordingLab @xaqlab @NunezKant @elonmusk Nothing in the interface demo was close to novel. Wireless, small footprint implants for neural recording have been in active development for a long time and it's not clear that Elon's product is even better. The only novelty is the surgical bot,(which is pretty cool if it works) 2021-02-03 03:02:48 @azhir_io @rudzinskimaciej @KordingLab @xaqlab @NunezKant @elonmusk I'm also (still) not convinced that public awareness is great if it's deeply oversold. People (and VCs and grant agencies) only have so much appetite for claims that don't deliver. Musk may just end up briefly eclipsing the long-running efforts in this area. > 2021-02-03 02:59:29 @azhir_io @rudzinskimaciej @KordingLab @xaqlab @NunezKant @elonmusk It's been discussed after the neuralink demo and it's not clear cut that this is great for the field. Musk makes wild claims about what is novel, while labs that have been working for years on this technology (and have mature/awesome products) are ignored. > 2021-02-03 01:05:31 @xaqlab @KordingLab @NunezKant @elonmusk That's a great point, but the problem with EEG is that it doesn't get much better than this even with training while fingers are a-may-zing 2021-02-03 01:04:00 @xaqlab @azhir_io @KordingLab @NunezKant @elonmusk Yea, I definitely agree that it inspires. I'm just not convinced that it's worth the price in hype. When Elon says he can do X, that raises expectations for all of us. 2021-02-03 00:05:35 And we're go! https://t.co/jyeo2USmgj https://t.co/xwrdhKfq7z 2021-02-02 21:52:40 @AlonZivony This also applies to "Reviewer 1: My suggestion is to rewrite the introduction to blah blah blah" 2021-02-02 21:42:42 @KordingLab @xaqlab @NunezKant @elonmusk That's a good way of putting it. It's a lot easier to improve the environment than the brain. 2021-02-02 21:26:12 RT @AlonZivony: https://t.co/zzgAZT5pAS 2021-02-02 21:25:35 @azhir_io @KordingLab @xaqlab @NunezKant @elonmusk This tweet has changed the outcome variable. I'm not talking about running a successful company and could care less if Tesla thrives. I'm concerned about what's good for the field and science in general. 2021-02-02 21:24:37 @xaqlab @KordingLab @NunezKant @elonmusk Now whether you should add a neural interface for this is unclear. Like Konrad I'm skeptical that this is feasible without some major advances (and I don't just mean smaller amplifiers & 2021-02-02 21:23:21 @xaqlab @KordingLab @NunezKant @elonmusk I'm pretty confident we can. Sensory remapping with the tongue, and other forms of skin-vision sensation work pretty well. VR has shown that people can adapt to having a lobster body with relative ease. 2021-02-02 19:05:20 @tyrell_turing Now we're talking. Can't disagree with that framing. 2021-02-02 19:02:48 @azhir_io @tyrell_turing yea it's a classic debate about the misleading nature of experience. Very weird stuff. 2021-02-02 18:54:06 @tyrell_turing This is especially true for time, which is sloppy when it comes to subjective experience. It's possible to experience ambiguity that gets resolved later and then the resolution is retroactively inserted into our experience such that we have no memory of the other state. 2021-02-02 18:52:08 @tyrell_turing Cognitive psychology as a discipline is built on the premise that we have great difficulty understanding our own mechanisms through subjective experience alone. I know you know all this, so I'm failing to understand why you think this is true because you feel like it's true. 2021-02-02 18:47:27 @tyrell_turing Of course it feels different to have that immediate sensation of understanding, and it does indicate some differences in the underlying processing as well. The question, as regards your statement that relates to prediction, is whether it really IS immediate. > 2021-02-02 17:45:53 RT @DrGBuckingham: This should be getting more love - the best ambiguous/bistable image you'll see all day https://t.co/GuK2lf7Iuf 2021-02-02 17:16:29 @tyrell_turing I'm sorry to be pedantic about a fun and real phenomenon, but it's important to be clear about when something is or isn't strong evidence for something. Maybe I'm not thinking about this the right way, but I can't see why this is evidence for prediction (over other likely ideas) 2021-02-02 17:09:07 @tyrell_turing I'm sure you've heard of the adage that consciousness is a story that we make up after the fact. The words settle into an interpretation in your head over time but when it's fast (because you have expectation), you can't feel that settling process happen so it seems instant. 2021-02-02 17:07:28 @tyrell_turing You're assuming that what you perceive as instantaneous is actually so, but you cannot trust your sense of perceived time, especially when it comes to simultaneity. So it seems instant to you but your brain has already had hundreds of milliseconds to chew on that information. > 2021-02-02 17:03:07 @tyrell_turing For example if you imagine that speech perception is like an SVM (ludicrous simplification but bear with it), the expectation induced by seeing an owl can allow you to turn a 500-way classification into a 5-way classification. Thus accuracy increases dramatically 2021-02-02 17:01:01 @tyrell_turing Yes, I realize now that this wasn't really your point, I should have deleted it. I guess my key issue here is that the effect of expectation on speech perception doesn't strongly support prediction, as there are other likely mechanisms that also explain this observation. 2021-02-02 14:46:43 @KordingLab @NunezKant @elonmusk @xaqlab Exactly! Now if you somehow attached a Threadripper in there.. 2021-02-02 14:23:09 @tyrell_turing Yes, schemas alter perception. This is a well known phenomenon in cognitive psych. Again, I'm not convinced this is a strong argument for predictive processing because an expectation reduces degrees of freedom for after-the-fact decoding, making it easier. 2021-02-02 14:18:28 @tyrell_turing Sure you have to be in the right frame of mind to perceive properly, that's always true of cognition. But the ability to decode is more related to familiarity than expectation. e.g. I couldn't understand chinese just by expecting that I could, no matter how many drugs I'd take. 2021-02-02 14:12:25 @azhir_io @KordingLab @NunezKant @elonmusk @xaqlab Sure, I think all of us would love for this to be a reality. The downside of Musk's wildly aspirational claims is that it sets unrealistic expectations and promotes hype cycles. 5 years later Musk moves on to another issue and has no care for the devastation in his wake. 2021-02-02 14:08:59 @tyrell_turing So this is true but what does have to do with predictive processing per se? The ability to decode a familiar sequence more easily than an unfamiliar sequence is predicted by a lot of theories. 2021-02-02 13:58:15 @KordingLab @NunezKant @elonmusk @xaqlab Got it. Yes I agree with that. I was just confused by the word "but". Elon sees this as a simple case of increasing bandwidth but he's not thinking about the true limit which is cognitive processing. That doesn't magically scale up with Bits. 2021-02-02 13:53:26 @KordingLab @NunezKant @elonmusk @xaqlab Did you mean to say that he has "a convincing argument" rather than "no convincing argument?" Otherwise I can't understand the point. 2021-02-02 13:41:33 @NunezKant @KordingLab @elonmusk @xaqlab The problem with his idea is that Elon wants to achieve this with neural linking rather than fingers, eyes and ears. 2021-02-02 13:39:53 @aina_puce It is beyond belief. I can't wrap my head around it. Meanwhile people are in prison for selling a joint. 2021-02-02 13:39:01 @tyrell_turing @NunezKant @analog_ashley @nicholdav If you do, go in with shields up and a backup plan. 2021-02-02 05:45:14 @filippie509 There's a huge practical argument in favor of this. People with this pile of debt are unable to get on their feet which makes it more difficult for them to be productive members the economy. and this is a debt the government can cancel without massive repercussions. 2021-02-02 05:27:12 RT @EtheHerring: Some personal news that some of you may find surprising (spoiler alert: I quit my job). Why I Am Leaving Academia - by Em… 2021-02-02 05:25:51 @neuro_data @tyrell_turing @blake_camp_1 @KordingLab @TonyZador Right, that seems familiar. Reading that thread again I don't see any very convincing arguments against the concerns you raised (although I didn't read the links) 2021-02-02 04:50:23 @neuro_data @tyrell_turing @blake_camp_1 @KordingLab @TonyZador Thanks Jovo, it was likely you I'd heard it from. Is there a discussion of this somewhere? 2021-02-02 04:30:25 @NunezKant @xaqlab Yup, that's exactly how it seems to me. He doesn't seem to understand how much harder biology is. 2021-02-02 04:09:11 @NunezKant @xaqlab I think he's just used to talking big about things he doesn't really understand and not worrying about whether he's right (and it's worked well for him) 2021-02-02 03:35:53 This is heart wrenching to hear. All the stories need to come out and accountability has to be a first step in unity. https://t.co/QVdtq83eIK 2021-02-02 03:21:07 @hannah_merseal https://t.co/tCKr9JbWLB 2021-02-02 01:04:07 @tyrell_turing @NunezKant Yea I wouldn't hold much hope that this will deliver what he hopes in the short term or the long term, but you will be surrounded by smart people, well funded and working on interesting problems. I'd see what you can find out about the workplace environment though. 2021-02-01 21:01:59 @tyrell_turing @blake_camp_1 @KordingLab @TonyZador @neuro_data Gauntlet thrown: can we train GPT-3 to emulate Blake at a party? (including the dancing) 2021-02-01 20:56:02 @tyrell_turing @blake_camp_1 @KordingLab @TonyZador @neuro_data https://t.co/3Ii4d0KOQG 2021-02-01 20:53:30 @tyrell_turing @blake_camp_1 @KordingLab @TonyZador @neuro_data I am so not willing to take that bet :) Come over to cognitive land for a few decades and see what you learn about how hard it is to be human. 2021-02-01 20:52:11 @tyrell_turing @blake_camp_1 @KordingLab @TonyZador @neuro_data Are you assuming 1 sample per second? 2021-02-01 20:50:50 @tyrell_turing @blake_camp_1 @KordingLab @TonyZador @neuro_data This is only half of it. What is the calculation of the dimensionality of the problem space and the training set size? 2021-02-01 20:46:08 @tyrell_turing @blake_camp_1 @KordingLab @TonyZador @neuro_data I have a recollection that someone failed to replicate it in some important conditions where it had been expected. Unfortunately I can't recall where that was. Thanks for the reference. 2021-02-01 20:42:30 @tyrell_turing @blake_camp_1 @KordingLab @TonyZador @neuro_data Thinking more, I'd guess that it's not possible to defend the brain being over or under hyper parameterized since we know so little about the # of parameters and the size of the search space. 2021-02-01 20:25:10 @tyrell_turing @blake_camp_1 @KordingLab @TonyZador @neuro_data I thought the double descent is not as robust as we thought. Is there any reason to think this is a thing that occurs in a system as complex as the brain? (where we can't even figure out how many parameters there are) 2021-02-01 19:59:59 @micahgallen That's huge-mongous. and if the building heat goes out everyone can stay warm. Wow those heatsinks 2021-02-01 19:35:36 @micahgallen Nice! what specs? 2021-02-01 19:00:17 @azhir_io @KordingLab @blake_camp_1 @tyrell_turing @TonyZador @neuro_data @neuralreckoning Yes, one might have to weigh each parameter's utility against its metabolic cost. I.e. it presumably costs less energy to have a genetically fixed parameter compared to one that is flexible but has to be stable on a learned value. 2021-02-01 18:58:02 @tyrell_turing @KordingLab @blake_camp_1 @TonyZador @neuro_data That's consistent with what I said though. And even given that true, it's not clear that the brain is in such a regime. 2021-02-01 18:07:19 @azhir_io @todd_gureckis Well I haven't tried live yet, that's where it may fail. 2021-02-01 18:04:46 @azhir_io @todd_gureckis I'm getting neither. My chief concern was simply blurring. It seems that OBS spits out a 1080p virtual camera which Zoom mysteriously downgrades. However when you share it using the "second camera" zoom feature it does not downgrade. Let's hear it for undocumented features.. 2001-01-01 01:01:01