Western scrub-jays (Aphelocoma californica) solve multiple-string problems by the spatial relation of string and reward

Abstract

String-pulling is a widely used paradigm in animal cognition research to assess what animals understand about the functionality of strings as a means to obtain an out-of-reach reward. This study aimed to systematically investigate what rules Western scrub-jays (Aphelocoma californica) use to solve different patterned string tasks, i.e. tasks in which subjects have to choose between two or more strings of which only one is connected to the reward, or where one is more efficient. Arranging strings in a parallel configuration showed that the jays were generally capable of solving multiple-string tasks and acted in a goal-directed manner. The slanted and crossed configurations revealed a reliance on a “proximity rule”, that is, a tendency to choose the string-end closest to the reward. When confronted with strings of different lengths attached to rewards at different distances the birds chose according to the reward distance, preferring the reward closest to them, and were sensitive to the movement of the reward, but did not consistently prefer the shorter and therefore more efficient string. Generally, the scrub-jays were successful in tasks where the reward was closest to the string-ends they needed to pull or when string length and reward distance correlated, but the birds had problems when the wrong string-end was closest to the reward or when the food items were in close proximity to each other. These results show that scrub-jays had a partial understanding of the physical principles underlying string-pulling but relied on simpler strategies such as the proximity rule to solve the tasks.Electronic supplementary materialThe online version of this article (doi:10.1007/s10071-016-1018-x) contains supplementary material, which is available to authorized users.

Performance in Object-Choice Aesop’s Fable Tasks Are Influenced by Object Biases in New Caledonian Crows but not in Human Children

“…In the standard task, a string dangling from a perch is gradually pulled up, piled under the bird’s feet, until the reward attached to the string can be reached [34]. Further, in multiple string-pulling tasks, Californian scrub-jays were sensitive to reward movement and chose strings according to reward distance [35]. To address this perceptual-motor feedback hypothesis, future experiments could impede visual access to the movement of the reward, such as occluding the view of the water level rising within trials.…”

Exploring the relative contributions of reward-history and functionality information to children’s acquisition of the Aesop’s fable task

“…The authors argue that it is the approach of the reward–rather than obtaining it–that reinforces the behaviour. Indeed, when the action-reward ratio is maintained but the movement of the reward is removed, animals in such studies become unable to learn [ 29 , 30 ]. In the present study, after each object-drop, the reward was either located in (or close to) its original position, or it rested detectably closer than it previously had.…”

Peer Review #2 of "Young children do not require perceptual-motor feedback to solve Aesop’s Fable tasks (v0.1)"

“…Recently, Miller and colleaoues demonstrated that in New Caledonian crows, but not in human children, performance on object-choice tasks can be influenced by pre-existino preferences for certain types of objects, castino some doubt on the suooestion that birds' success on Aesop's Fable tasks reflects causal understandino -at least when considerino their selection of objects (Miller et al, 2016). Perceptual-motor feedback has been suooested to underpin spontaneous strino pullino behaviour performed by birds (Taylor et al, 2010;Taylor, Knaebe & Gray, 2012;see also Jacobs & Osvath, 2015;Hofmann, Cheke & Clayton, 2016), as well as performance on certain problemsolvino tasks by oreat apes (Völter & Call, 2012). However, to date, it is unclear whether the opportunity to receive perceptual-motor feedback accounts for non-human animals' ability to rapidly solve various water displacement tasks.…”

“…Therefore, subjects could learn to solve these tasks by repeatino those actions that brino the reward incrementally closer -i.e., by respondino to perceptual-motor feedback (Taylor & Gray, 2009;Cheke, Bird & Clayton, 2011;Jelbert, Taylor & Gray, 2015). This type of feedback is thouoht to underpin the seeminoly 'insiohtful' behaviour by which birds spontaneously learn to pull up strinos to brino in attached rewards (New Caledonian crows: Taylor et al, 2010;Taylor, Knaebe & Gray, 2012;common ravens, Corvus corax: eeinrich & Bugnyar, 2005; California scrub jays, Aphelocoma californica : Hofmann, Cheke & Clayton, 2016), and is a plausible explanation for the birds' behaviour on water displacement tasks. For example, Cheke et al (2011) found that one out of two Eurasian jays could pass an arbitrary task where a reward was pushed incrementally towards the subject each time they dropped a stone into an L-shaped apparatus.…”