linking back to brembs.net






My lab:
lab.png
Welcome Guest
Username:

Password:


Remember me

[ ]
[ ]
[ ]
 Currently Online (18)
 Extra Information
MicroBlog
NeuroTwitter

[14 Oct 11: 05:45]
Just received an email from a computer science student - with an AOL email address?

[03 Jul 11: 16:26]
Google citation alerts suck: I just found out by accident I rolled over h-index of 13 and 500 citations http://bjoern.brembs.net/citations.php

[21 May 11: 12:14]
6.15pm: Does god have Alzheimer? No #rapture in Europe...

[01 May 11: 05:31]
w00t! Just been invited to present at OKCon 2011! #OKCon2011

[13 Mar 11: 05:59]
Review on "Spontaneous decisions and operant conditioning in fruit flies" out http://dx.doi.org/10.1016/j.beproc.2011.02.005

[09 Feb 11: 12:01]
Rule of thumb for revising manuscripts: for every point the reviewer raises, change something, anything in the manuscript.


Networking
Random Video
SciSites
GeoCounter
outils webmaster
There's a new paper out by the group of Nicky Clayton, Cambridge, in the April 4 issue of Current Biology. I know Nicky from the Social Science Research Council sponsored workshop "Behavioral Organization in Animals" in Bodega Bay (see program).
In their study, the researchers used the popular ‘trap-tube’ task to assay the extent of physical cognition (specifically, an understanding of the operation of gravity) in rooks. In the experiment, food is placed inside a horizontal tube which has a vertical, blind-ended tube attached to it. The animals had to push (or pull) the item from the appropriate end of the tube using a stick, so that they do not lose the item in the trap. Seven out of eight rooks learned the task and all seven passed a transfer test, in which the food had to be dropped into the trap to be accessible (see picture).



The authors hypothesize that the rooks have a sense of gravity ("physical cognition") and may use learning to abstract rules to acquire it.
There was a fairly recent (2002) report in the journal Science by Weir, Chappell and Kacelnik, who showed that New Caledonian crows are able to shape unfamiliar materials to create a usable tool for a specific task.
These corvids don't have birdbrains, it seems
Posted on Tuesday 04 April 2006 - 09:59:49 comment: 3
rooks   

Comments
Matt posted on10 May 06: 20:16: Clever rooks
Guest


Reply to this
Hi, I read that article too. I guess I just dont get the big difference. The 'recursive' sequences were of the form AB, AABB, AAABBB, etc. But I fail to see why those sequences require a radically different form of recognition. On the other hand, the fact that the tamarins couldn't learn these sequences is interesting. I would be interested in your comments. Thanks, Matt

bjoern posted on11 May 06: 03:49: Clever rooks
Comments: 283


Reply to this
For us, it doesn't seem to be such a big deal to differentiate AABB from ABAB. I guess this is because it's so engrained in our language. But for animals, it's not that easy. I guess that's because both sequences contain the same number of As and Bs and so are very similar in "content". My prediction would be that quite a few species will be able to discriminate ABAB from AABB, but you would need to train them operantly, i.e. they need to really have to work for the food and be engaged. Moreover, you might need to use very specific stimuli that match the sort of stimuli the animals are faced with in their "normal" life (as the song sounds for the starlings).
It may be that the researchers studying the tamarins also thought "ha, this should be no big deal for them, let's see if they can do it!". More studies clearly need to be performed on more species to find out if the distinction really is difficult, or maybe we're just not asking the animals in the right way.
Does that address your line of thought?

bjoern posted on11 May 06: 03:49: Clever rooks
Comments: 283


Reply to this
BTW, shouldn't these comments have goe to the starling entry? Did you accidentally post yours in this entry, or is there a bug in the comments function?

Submit comment
Subject
Username:
Comment:

//
Render time: 0.8219 sec, 0.3938 of that for queries.
UA-4203938-1