Navigation: many senses make efficient foraging paths

Multimodal Information Processing and Associative Learning in the Insect Brain

Simple Summary

Insect behaviors are a great indicator of evolution and provide useful information about the complexity of organisms. The realistic sensory scene of an environment is complex and replete with multisensory inputs, making the study of sensory integration that leads to behavior highly relevant. We summarize the recent findings on multimodal sensory integration and the behaviors that originate from them in our review.

Abstract

The study of sensory systems in insects has a long-spanning history of almost an entire century. Olfaction, vision, and gustation are thoroughly researched in several robust insect models and new discoveries are made every day on the more elusive thermo- and mechano-sensory systems. Few specialized senses such as hygro- and magneto-reception are also identified in some insects. In light of recent advancements in the scientific investigation of insect behavior, it is not only important to study sensory modalities individually, but also as a combination of multimodal inputs. This is of particular significance, as a combinatorial approach to study sensory behaviors mimics the real-time environment of an insect with a wide spectrum of information available to it. As a fascinating field that is recently gaining new insight, multimodal integration in insects serves as a fundamental basis to understand complex insect behaviors including, but not limited to navigation, foraging, learning, and memory. In this review, we have summarized various studies that investigated sensory integration across modalities, with emphasis on three insect models (honeybees, ants and flies), their behaviors, and the corresponding neuronal underpinnings.

Using your nose to find your way: Ethological comparisons between human and non-human species

Olfaction is arguably the least valued among our sensory systems, and its significance for human behavior is often neglected. Spatial navigation represents no exception to the rule: humans are often characterized as purely visual navigators, a view that undermines the contribution of olfactory cues. Accordingly, research investigating whether and how humans use olfaction to navigate space is rare. In comparison, research on olfactory navigation in non-human species is abundant, and identifies behavioral strategies along with neural mechanisms characterizing the use of olfactory cues during spatial tasks. Using an ethological approach, our review draws from studies on olfactory navigation across species to describe the adaptation of strategies under the influence of selective pressure. Mammals interact with spatial environments by abstracting multisensory information into cognitive maps. We thus argue that olfactory cues, alongside inputs from other sensory modalities, play a crucial role in spatial navigation for mammalian species, including humans; that is, odors constitute one of the many building blocks in the formation of cognitive maps.

Orientation to polarized light in tethered flying honeybees

Many insects exploit the partial plane polarization of skylight for visual compass orientation and/or navigation. In the present study, using a tethering system, we investigated how flying bees respond to polarized light stimuli. The behavioral responses of honeybees (Apis mellifera) to a zenithal polarized light stimulus were observed using a tethered animal in a flight simulator. Flight direction of the bee was recorded by monitoring the horizontal movement of its abdomen, which was strongly anti-correlated with its torque. When the e-vector orientation of the polarized light was rotated clockwise or counterclockwise, the bee responded with periodic right-and-left abdominal movements; however, the bee did not show any clear periodic movement under the static e-vector or depolarized stimulus. The steering frequency of the bee was well coordinated with the e-vector rotation frequency of the stimulus, indicating that the flying bee oriented itself to a certain e-vector orientation, i.e. exhibited polarotaxis. The percentage of bees exhibiting clear polarotaxis was much smaller under the fast stimulus (3.6 deg s-1) compared with that under a slow stimulus (0.9 or 1.8 deg s-1). Bees did not demonstrate any polarotactic behavior after the dorsal rim area of the eyes, which mediates insect polarization vision in general, was bilaterally covered with black paint. Preferred e-vector orientations under the clockwise stimulus varied among individuals and distributed throughout -90 to 90 deg. Some bees showed similar preferred e-vector orientations for clockwise and counterclockwise stimuli whereas others did not. Our results strongly suggest that flying honeybees utilize the e-vector information from the skylight to deduce their heading orientation for navigation.

A Neurocomputational Model of Goal-Directed Navigation in Insect-Inspired Artificial Agents

Despite their small size, insect brains are able to produce robust and efficient navigation in complex environments. Specifically in social insects, such as ants and bees, these navigational capabilities are guided by orientation directing vectors generated by a process called path integration. During this process, they integrate compass and odometric cues to estimate their current location as a vector, called the home vector for guiding them back home on a straight path. They further acquire and retrieve path integration-based vector memories globally to the nest or based on visual landmarks. Although existing computational models reproduced similar behaviors, a neurocomputational model of vector navigation including the acquisition of vector representations has not been described before. Here we present a model of neural mechanisms in a modular closed-loop control—enabling vector navigation in artificial agents. The model consists of a path integration mechanism, reward-modulated global learning, random search, and action selection. The path integration mechanism integrates compass and odometric cues to compute a vectorial representation of the agent's current location as neural activity patterns in circular arrays. A reward-modulated learning rule enables the acquisition of vector memories by associating the local food reward with the path integration state. A motor output is computed based on the combination of vector memories and random exploration. In simulation, we show that the neural mechanisms enable robust homing and localization, even in the presence of external sensory noise. The proposed learning rules lead to goal-directed navigation and route formation performed under realistic conditions. Consequently, we provide a novel approach for vector learning and navigation in a simulated, situated agent linking behavioral observations to their possible underlying neural substrates.