An Experimental Investigation of Insight in Common Ravens (Corvus corax)

Abstract

I presented four groups of Common Ravens (Corvus corax) with a problem that they had never encountered before. Could they demonstrate the solution to this problem without first practicing or learning the correct sequence of intermediary steps? The problem posed was the reaching of food suspended by a string. The solution required perching above the string and the food, reaching down, pulling up a loop of string, setting the looped-up string onto the perch, stepping onto the string, releasing the string with the bill while simultaneously applying pressure with the foot onto it, then reaching down again to repeat the cycle six to eight times in that precise order before finally securing a piece of dried meat. The results varied enormously between individuals. However, typically a bird approached the string nervously, pecked or briefly yanked on the string, repeated the approach when given another opportunity, extinguished the approach behavior, or suddenly did the entire string-pulling sequence correctly. One of the wild birds performed the entire sequence correctly on his first approach to the string, even though no other bird of that group had shown the behavior. After a bird had acquired the behavior it thereafter performed the behavior correctly without fail. Other behaviors were associated with successful string pulling. From their first trial, the four hand-reared individuals dropped the meat attached to string (and perch) if they were shooed from the perch. In contrast, other birds that were handed the food attached to string attempted to fly off with it, and it required five to nine trials before they refused to do so, apparently learning the consequences of this behavior. Other problems related to food presented on the dangling string also were solved without first overtly trying out the alternatives. These problems involved: (1) crossing the string with food with another string that held a rock; (2) using a novel string with food, next to the previously-rewarded "old" string; and (3) having food on string next to a rock on string, but with insertion of the string on the perch now displaced laterally. In contrast, the birds performed very poorly at some tasks where simple trial-and-error learning quickly would have resulted in appropriate responses. For example, three birds never once (in 79 trials) pulled the correct string in the crossed-strings experiments that another mustered with no trials. The results are discussed in terms of possible insight and alternative mechanisms, including innate behavior and learning.