After learning about smart rooks, there's a new story out on clever birds. This time, European Starlings (Sturnus vulgaris, see image) are reported to "accurately recognize acoustic patterns defined by a recursive, self-embedding, context-free grammar". It essentially means that the animals have been trained to accoustically distinguish patterns such as ABAB from AABB. Or, more generally, to distinguish AⁿBⁿ from (AB)ⁿ. See also Scotsman.com newsreport.


For me the important part of the study was the difference in the procedure to another study in which the researchers tried and failed to train tamarins (monkeys) the same distinction: in the tamarins, the animals were asked to distinguish the patterns after a brief exposure. In the starlings, the researchers trained them using an operant conditioning procedure involving extensive positive and negative feedback. The accompanying News and Views article suggests that maybe the operant control of the patterns is what makes starlings succeed and tamarins fail to acquire recursive grammar? I'll have to add this piece of evidence to my essay on operant conditioning.

Tags: neurobiology, neuroscience, operant conditioning