Journal of Experimental Biology - Latest Issue

• fish2eod: finite element modelling of active electric sensing

  ABSTRACT

  Understanding the neural basis of animal behaviour requires a thorough description of the associated sensory inputs. This is especially important when behaviour actively shapes incoming sensory information. Weakly electric fish use perturbations in a self-generated electric field as a basis for an electric sense, and these field perturbations are encoded by electroreceptors distributed over their bodies. Thus, swimming movements and body pose shape not only the field but also the orientation of the receptor array. Previous modelling in this context has focused primarily on the so-called electric image in stationary fish and has not addressed how natural electrosensory inputs are generated in freely swimming fish. Here, we present fish2eod, an open-source finite-element-based modelling framework that describes the dynamics of electrosensory inputs during natural behaviours, including social interactions, in complex environments.

• A technical semi-field methodology to measure the effect of nutrition on honey bee brood rearing

  ABSTRACT

  A honey bee colony's well-being is its ability to nurture larvae into healthy adults. Understanding how nutrition supports brood rearing is crucial for developing diets that could aid against environmental threats. Nutritional research on whole-colony brood development has been historically challenging because of difficulties documenting the diet's impact on brood production over time. We describe a novel semi-field method to study the influence of nutrition on brood rearing using standardised small colonies formed de novo (ca. 1500 nurse-age bees and a queen) housed in adapted mating-nucs, placed inside an enclosure and limited to feeding on chemically defined diets. Complete assessments were conducted every 15 days, assisted by a bespoke device to photograph every frame to measure cell contents. A novel metric describes the number of bees generated per gram of diet consumed, measuring the impact of nutrition on brood rearing and overall colony size.

• Ignoring what we don't know in accelerometer-based behaviour classification: the open-set recognition problem

  ABSTRACT

  Supervised machine learning is commonly used to classify fine-scale behaviours from animal-borne accelerometers, assigning new data to predefined behaviour categories seen during training. These models cannot recognise novel behaviours as ‘unknown’, however, and, when exposed to new behaviours, will continue to overpredict the known classes. This issue – known as open-set recognition – is an inevitable, but underexplored, limitation in accelerometer-based behaviour classification. Here, we describe the problem and assess four solutions: (1) a multiclass model with an ‘other’ category, (2) threshold-based models, (3) one-class models and (4) binary one-versus-all models. We show that traditional multiclass models produce high false-positive rates when exposed to behaviours not present during training. We instead suggest the implementation of binary one-versus-all models as a more conservative method, particularly in cases where a single or limited set of behaviours are of interest. Awareness of this challenge will enhance recognition of often unreported uncertainty in real-world applications.

• Otsuchi Coastal Research Center: a remote but vital marine science hub for the Northwest Pacific Ocean

  ABSTRACT

  The Otsuchi Coastal Research Center (OCRC), a field station belonging to the Atmosphere and Ocean Research Institute at the University of Tokyo, was established in 1973 in Otsuchi, a coastal town on the Sanriku coast of Honshu, Japan. Located near a site where warm and cold ocean currents converge, OCRC facilitates research in biology, chemistry, physics and geoscience within a unique marine environment shaped by a complex rocky coastline and river-fed bays. The centre is staffed by resident researchers and technicians, provides research vessels and a dormitory, and supports around 2000 person-days of visiting scientists annually for field observations, aquarium-based experimentation and instrumental analyses. Since 2004, we have pursued biologging studies at OCRC with graduate students and collaborators from Japan and abroad. This research has focused on loggerhead and green turtles, streaked shearwaters, chum salmon and ocean sunfish, producing insights into physiology, behaviour, ecology and environmental science. In 2011, the original research building and dormitory were severely damaged by an earthquake and tsunami, resulting in the loss of field notes, materials and some data. Fortunately, there were no casualties, and a new research building and dormitory were rebuilt on higher ground in 2018. To enhance data preservation, the biologging intelligent platform (BiP) was established to archive biologging datasets with their associated metadata. To better understand how marine animals respond to ongoing environmental changes, continued long-term field research and historical data comparison are essential. With access to diverse ecosystems and robust technical infrastructure, and its collaborative research culture, OCRC is uniquely positioned to potentially meet that demand.

• Flow impairs multisensory tracking and increases active sensing in weakly electric fish

  ABSTRACT

  Weakly electric fish rely on electrosensory, visual and mechanosensory (lateral-line) cues to guide behavior in flowing water, yet the effects of ambient currents on multisensory tracking and active sensing remain poorly understood. We tested the weakly electric knifefish Apteronotus albifrons (n=4) tracking a moving refuge in a recirculating flow tunnel while systematically varying flow speed (0–16 cm s⁻¹), illumination (light versus dark) and refuge structure (windowed versus non-windowed). Tracking performance was quantified with time- and frequency-domain measures (root-mean-square error; gain–phase analyses), and active sensing as movement power outside stimulus frequencies (mean active sensing power, MASP). Increasing flow degraded tracking: relative to still water, RMSE rose by ∼46% at 16 cm s⁻¹. Deficits were largest in darkness and with the windowed refuge, and were concentrated at low stimulus frequencies. Under higher flows, fish showed a trend toward increased off-frequency movement power (by ∼33%), consistent with compensatory active sensing to sustain sensory acquisition. The effects were non-linear and context dependent. This pattern indicates that increasing hydrodynamic noise may drive dynamic reweighting among visual, electrosensory and mechanosensory inputs. Collectively, our data indicate that ambient flow degrades low-frequency tracking and may elicit compensatory active sensing in A. albifrons, extending recent demonstrations of context-dependent sensing and control switches in this species and bridging rheotaxis with electrosensory refuge tracking.