tag:blogger.com,1999:blog-1812431336777691886.post4689345041202858694..comments2024-02-29T03:57:00.088-05:00Comments on The Mermaid's Tale: The methane made me do it! (But I'll be nicer, now)Anne Buchananhttp://www.blogger.com/profile/09212151396672651221noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-1812431336777691886.post-58546948969470441382013-08-14T09:41:54.920-04:002013-08-14T09:41:54.920-04:00Thanks for the thoughtful reply John but this is w...Thanks for the thoughtful reply John but this is way out of my league right now. I was just happy to see how the same behaviors could be explained differently by different observers, and experts at that. Holly Dunsworthhttps://www.blogger.com/profile/05260104967932801186noreply@blogger.comtag:blogger.com,1999:blog-1812431336777691886.post-18359015492827361292013-08-14T03:11:00.318-04:002013-08-14T03:11:00.318-04:00Holly,
There are at least two worlds of thought ...Holly,<br /> There are at least two worlds of thought in those domains, structuralists and those (like me) who see the structural regularities of behaviour emerging from the memory of past experience. So in addition to my modest contributions, you might want to look at the work of Landauer and Latent Semantic Analysis (LSA) that also uses singular value decomposition (SVD) of large spaces (see http://www.stat.cmu.edu/~cshalizi/350/2008/readings/Landauer-Dumais.pdf). There is also all the work of Burgess and his lab (http://locutus.ucr.edu) that seems to capture the same themes. The point, though, is simple: with enough varied experience, structure emerges as a necessary consequence, especially (or necessarily) if dimension reduction is involved. I won't bore you with the details, the citations are enough. And reading Brooks (1978), the true source of nonanalytic cognition, should set you on the path.John R. Vokeyhttps://www.blogger.com/profile/03822243132435056442noreply@blogger.comtag:blogger.com,1999:blog-1812431336777691886.post-5315108246797515232013-08-13T09:50:54.193-04:002013-08-13T09:50:54.193-04:00Thanks for sharing this John. I'm writing abou...Thanks for sharing this John. I'm writing about things like monkey and ape cognition (in language and social contexts specifically) digging myself deep into fields that aren't so familiar to me ... and it's always great to read about counter-interpretation like yours. It really helps even if I still don't really understand it completely. Holly Dunsworthhttps://www.blogger.com/profile/05260104967932801186noreply@blogger.comtag:blogger.com,1999:blog-1812431336777691886.post-88208388377090658942013-08-13T07:45:28.221-04:002013-08-13T07:45:28.221-04:00The BS-Meter glows red just about every day, somet...The BS-Meter glows red just about every day, sometimes with Big Story reports of basically much ado about very little, but too often total nonsense blown into a Big Story. Yesterday was a real genetic disease howler by an otherwise respectable writer for the Times (I won't name names in this public space, since the story is so egregious anyone can find it).<br /><br />Still, this is how our culture works, how careers are made, and so on. Every now and then there is a grain of value in the hurricane of chaff.Ken Weisshttps://www.blogger.com/profile/02049713123559138421noreply@blogger.comtag:blogger.com,1999:blog-1812431336777691886.post-11127207941811098012013-08-12T23:48:50.699-04:002013-08-12T23:48:50.699-04:00Ken,
I have recently been involved in a case tha...Ken,<br /> I have recently been involved in a case that I think fits your "BSfactor" category in spades. We originally wrote this piece to respond to the Science magazine's (I can no longer call it a journal) over-the-top hyping of an article that Science claimed showed that baboons can learn to engage in orthographic processing (i.e., lexical decisions). The hype was picked up and repeated in many other sources, New Scientist and National Geographic to name just two. And, it appeared in almost every newspaper and online science magazine (including Nature).<br /><br />Science refused even to send our response (a simple neural net simulation of the Science paper results showing that simple familiarity---not orthographic processing---could account for the results) out for review with no justification given. Fair enough, I guess. But then a new paper in Psychological Science appeared that provided new evidence with the same baboons that what we had claimed (no citation given, even though, as Science requires, we had sent our ms to the original authors), expressed as a general concern (again, fair enough, we were most likely not the only people to notice the weaknesses of the original claims), showed that, indeed, it was orthographic processing the baboons were using. So, we immediately updated our simulation to handle the latest results. It is now under review at Psychological Science.<br /><br />As with your piece, our concern is 1) how either of these articles made it through the review process in the first place, and 2) the exaggerated plethora of hype that followed the original publication. Seriously, none of this is rocket science. And the claims are simply preposterous from the outset. The unfortunate downside of all of this is that the authors have developed an amazing system for testing the animals that I am trying to replicate in my own department (but for wild corvids, not baboons), that may get lost in the resulting kerfuffle over over-charged claims.<br /><br />A pre-print of the version submitted to Psychological Science is here: https://dl.dropboxusercontent.com/u/2793590/VokRan2.pdf<br />John R. Vokeyhttps://www.blogger.com/profile/03822243132435056442noreply@blogger.com