Minimum complexity integrated photonic architecture for delay-based reservoir computing.

Reservoir computing is an analog bio-inspired computation scheme for efficiently processing time-dependent signals, the photonic implementations of which promise a combination of massive parallel information processing, low power consumption, and high-speed operation. However, most of these implementations, especially for the case of time-delay reservoir computing, require extensive multi-dimensional parameter optimization to find the optimal combination of parameters for a given task. We propose a novel, largely passive integrated photonic TDRC scheme based on an asymmetric Mach-Zehnder interferometer in a self-feedback configuration, where the nonlinearity is provided by the photodetector, and with only one tunable parameter in the form of a phase shifting element that, as a result of our configuration, allows also to tune the feedback strength, consequently tuning the memory capacity in a lossless manner. Through numerical simulations, we show that the proposed scheme achieves good performance -when compared to other integrated photonic architectures- on the temporal bitwise XOR task and various time series prediction tasks, while greatly reducing hardware and operational complexity.

The effects of tradition on problem solving by two wild populations of bearded capuchin monkeys in a probing task.

The effects of culture on individual cognition have become a core issue among cultural primatologists. Field studies with wild populations provide evidence on the role of social cues in the ontogeny of tool use in non-human primates, and on the transmission of such behaviours over generations through socially biased learning. Recent experimental studies have shown that cultural knowledge may influence problem solving in wild populations of chimpanzees. Here, we present the results from a field experiment comparing the performance of bearded capuchin monkeys (Sapajus libidinosus) from two wild savannah populations with distinct toolkits in a probing task. Only the population that already exhibited the customary use of probing tools succeeded in solving the new problem, suggesting that their cultural repertoire shaped their approach to the new task. Moreover, only this population, which uses stone tools in a broader range of contexts, tried to use them to solve the problem. Social interactions can affect the formation of learning sets and they affect the performance of the monkeys in problem solving. We suggest that behavioural traditions affect the ways non-human primates solve novel foraging problems using tools.

Diversity of nutcracking tool sites used by Sapajus libidinosus in Brazilian Cerrado.

Cracking nuts with tools is a behavior documented in a small number of populations of tufted capuchins, mainly in semi-arid Caatinga and Caatinga-Cerrado transitional environments of northeastern Brazil. Only one of these populations inhabits the less arid Cerrado in Central Brazil, where environments are composed of a heterogeneous mosaic of fields, savannas and forest formations. We conducted surveys in 10 of 20 localities where nutcracking by capuchins was reported by the local inhabitants in the Cerrrado of Northern Goiás and Tocantins. Our purpose was to evaluate nutcracking sites (anvils and associated hammers and nuts) based on indirect evidence of extensive pounding of nuts and seeds. Nutcracking was confirmed at all 10 surveyed localities. A total of 270 sites were identified. Surveyed localities included areas that were ecologically similar to those where capuchins crack nuts in Caatinga, as well as less arid localities with more typical Cerrado habitat. Anvils and hammers were made of materials including quartz, limestone, sandstone and wood, and displayed a wider range of sizes (i.e., 60-3,750 g for hammers' weight) than reported at previously studied localities. Nuts of seven genera were found in association with anvils and hammers. We conclude that nutcracking by capuchins are not restricted to arid environments and argue that the occurrence and diversity of nutcracking tool sites result from complex interactions of environmental variables (e.g., availability of food and mineral resources, density of canopy cover) and social variables (e.g., spatial cohesiveness and tolerance among group members) that need to be examined through long-term research of habituated groups. Localities in the Cerrado of Northern Goiás and Tocantins vary considerable in the ecological conditions faced by wild groups, and therefore offer the opportunity to examine these interactions.

Stone anvil damage by wild bearded capuchins (Sapajus libidinosus) during pounding tool use: a field experiment.

We recorded the damage that wild bearded capuchin monkeys (Sapajus libidinosus) caused to a sandstone anvil during pounding stone tool use, in an experimental setting. The anvil was undamaged when set up at the Fazenda Boa Vista (FBV) field laboratory in Piauí, Brazil, and subsequently the monkeys indirectly created a series of pits and destroyed the anvil surface by cracking palm nuts on it. We measured the size and rate of pit formation, and recorded when adult and immature monkeys removed loose material from the anvil surface. We found that new pits were formed with approximately every 10 nuts cracked, (corresponding to an average of 38 strikes with a stone tool), and that adult males were the primary initiators of new pit positions on the anvil. Whole nuts were preferentially placed within pits for cracking, and partially-broken nuts outside the established pits. Visible anvil damage was rapid, occurring within a day of the anvil's introduction to the field laboratory. Destruction of the anvil through use has continued for three years since the experiment, resulting in both a pitted surface and a surrounding archaeological debris field that replicate features seen at natural FBV anvils.