contact Patricia

Please enter your email address and a message.


123 Street Avenue, City Town, 99999

(123) 555-6789


You can set your address, phone number, email and site description in the settings tab.
Link to read me page with more information.

New Paper Friday

Nurse plants and desert communities

Patricia Jones

The nurse shrub,  Mimosa luisana . The Spanish common name is 'madre de los tetechos' or 'mother of the tetechos'. Tetechos,  Neobuxbaumia tetetzo , are a type of columnar cactus. Photo by  naturalia .

The nurse shrub, Mimosa luisana. The Spanish common name is 'madre de los tetechos' or 'mother of the tetechos'. Tetechos, Neobuxbaumia tetetzo, are a type of columnar cactus. Photo by naturalia.

Deserts can be some of the harshest environments on earth, particularly for young plants. The presence of some plant species (called 'nurse plants'), however, can make it easier for young plants to establish. Mimosa luisana is a legume that fixes nitrogen. Other plants can take up this nitrogen that would not otherwise be available to them. This week's paper is lead by Alicia Montesinos-Navarro at the Universidad Nacional Autónoma de México, and is in Ecology. Montesinos-Navarro and her colleagues investigated which plants took up the nitrogen fixed by M. luisana in the Valley of Zapotitlán, Mexico. M. luisana is a nurse plant because it provides shade and nitrogen enabling the establishment of young plants under it, especially tetecho cacti, Neobuxbaumia tetetzo, earning it the Spanish common name 'madre de los tetechos'. 

Montesinos-Navarro and her colleagues investigated which plants take up the nitrogen fixed by M. luisana because they were interested in a classic question in ecology, which is how is diversity maintained in communities? Why are plant communities not monocultures of only one species but rather diverse assemblages? The traditional explanation has been that closely related species (or individuals of the same species) compete too much to coexist in high numbers. They occupy the same 'niche' in terms of their needs for certain soil, light, or water conditions. Different species in contrast can coexist more peacefully because they have slightly different niches and compete less. Montesinos-Navarro and her colleagues looked at the coexistence question from a different perspective and asked whether rather than competition, facilitation could be a driving force in community interactions. In particular, whether plants might facilitate the establishment of more distantly related species more than closely related species, generating community diversity. 

What they did

Montesinos-Navarro and her colleagues selected 14 plant species that co-occur with and are facilitated by M. luisana. They calculated the phylogenetic distance (distance of relatedness) between M. luisana and each of these species. Additionally they examined the total nitrogen content of the leaves of each of these species in comparison to M. luisana. They then soaked the leaves of the M. luisana plants in a stable isotope of nitrogen that is very rare in nature (15N). The M. luisana plants took up this nitrogen and it was transferred into the soil and into neighboring plants through shared fungal interactions. Finally they measured the amount of 15N in the neighboring plants.

What they found

Plants that were more distantly related from M. luisana were more different in the total nitrogen content of their leaves (not the 15N here, just naturally occurring leaf nitrogen). It makes intuitive sense that nitrogen fixing plants such as M. luisana have lots of nitrogen in their leaves in comparison to their non-nitrogen fixing distant relatives. This creates a source-sink gradient, where more distantly related plants may have more demand for nitrogen than plants more closely related to M. luisana. Correspondingly, these more distantly related plants took up more of the 15N added to the M. luisana than closely related plants did. 

The takeaway

The nitrogen fixed by the nurse plant, M. luisana, is taken up more by distantly related plants than closely related plants. This means that nitrogen fixing could facilitate the coexistence of distantly related plant species in the community. So rather than co-occurring because they don't compete as much with M. luisana, they could co-occur because they benefit more from the presence of M. luisana.  

Smashing & Spearing Stomatopods

Patricia Jones

A smasher,  Gonodactylus childi.  Photo from  CalPhotos .

A smasher, Gonodactylus childi. Photo from CalPhotos.

A spearer,  Raoulserenea  sp. Photo by  J. Poupin .

A spearer, Raoulserenea sp. Photo by J. Poupin.

This week's paper is by Maya deVries, a postdoc at the Scripps Institute for Oceanography, and is in Biology Letters. deVries has examined how stomatopod (the mantis shrimps) diet diverges with their hunting appendages. Some stomatopods have spearing arms that allow them to stab prey. Other stomatopods have amazing smashing appendages. These smashers are some of the fastest moving animal parts on earth, moving so fast (20 meters per second!) through the water that they cause the water to vaporize (called cavitating). The vapor bubble then implodes with heat, light and sound. A smashing mantis shrimp therefore breaks a snail shell with both it's own strike and the implosion of vaporized water! Much of this research was done by Sheila Patek, and you can watch her give a TED talk about it.

In this week's paper deVries examined the diet of two small stomatopods, a spearer and a smasher, that live in the same habitat in French Polynesia. deVries hypothesized that they would have very different diets in accordance with their different morphologies. The spearer would be eating more soft-bodied prey and the smasher more snails and other hard prey. 

What she did

deVries determined the diets of these two stomatopods using stable isotope analysis. She collected the stomatopods as well as eight potential prey including shrimp, crabs, hermit crabs, clams, fish, and worms. Different animals (and plants for that matter) have different ratios of carbon to nitrogen in their tissues. After measuring the ratios in the prey species de Vries can then plot the ratios of the two stomatopods on top of the prey isotope ratios, and where the stomatopod ratios fall is indicative of which prey they are eating. 

What she found

deVries found that the spearing stomatopods had been eating lots of fish, and the smashing species were eating a lot of clams, but both species consumed all of the prey (they were eating about 70% fish and clams respectively). This was less specialization than deVries expected to see in these species given that they have to compete with each other when they eat the same diets, and their very different weapons. 

The takeaway

deVries proposes that the evolution of the amazing smashing appendages in stomatopods rather than causing them to specialize on hard-bodied prey, thereby narrowing their diet, just broadens their current diet by allowing them to add lots of clams to a fish diet. Rather than dividing up the available food using their different weapons the two species eat a lot of the same things but are able to expand their diet in slightly different directions with different weapons. 

Stomatopods are amazing animals for lots of reasons. They also have unique color vision (there's a good RadioLab on this) which I guess is not a surprise when they can look like this: 

The truly spectacular peacock mantis shrimp. Photo by George Graff. 

The truly spectacular peacock mantis shrimp. Photo by George Graff. 

Bumblebee learning in the lab and performance in the wild

Patricia Jones

A  Bombus terrestris  foraging in the wild. Photo my own.

A Bombus terrestris foraging in the wild. Photo my own.

This week's paper is in Scientific Reports, co-lead authored by Lisa Evans and Karen Smith in collaboration with Nigel Raine (Guelph and Royal Holloway). They have tackled one of my favorite topics, which is the link between learning, or cognitive abilities, and fitness. To do this they used colonies of a European bumblebee species, Bombus terrestris, studied learning in assays in the lab and then let those same bees forage in the wild and assessed their foraging performance. 

What they did 

The authors used five bumblebee colonies which they split in half using a piece of mesh screen. Half of the colony was connected to a foraging arena where they conducted learning experiments, and the other half had access to the outside through a window. Individual bees that emerged on the arena side were given colored, numbered tags, and tested in a learning assay. The learning assay was an array of yellow and blue flowers, with nectar in the yellow flowers and the blue flowers empty. Bumblebees generally (although often not!) innately prefer blue flowers. The assay examined how long it took bumblebees to learn to only visit yellow flowers. After a bumblebee had completed the learning assay it was then tagged with an RFID tag and moved to the side of the colony connected to the outside. As these bees then foraged outside they had to pass over a scale that weighed them as they exited and re-entered and recorded their individual ID from the RFID tag. By weighing the bees before and after foraging the authors were able to estimate the amount of nectar brought back to the colony by each bee, and they also visually estimated the amount of pollen bees were carrying.

What they found

First off, they found substantial variation in how long it took individual bees to learn to only go to yellow flowers. Bees that learned faster in the assay did not collect more nectar or pollen than bees that learned slower in the assay. But, as bees continued to forage in the wild, the amount of nectar they brought back over successive trips increased. Bees that learned faster in the assay, however, foraged for fewer days over their lifespan in comparison to slower learners. Because there is no difference in the amount of resources fast and slow learning bees brought back to the colony, slow learning bees that foraged for more days collected more resources for the colony than fast learning bees. 

The takeaway

The authors propose that being a faster learner has physiological costs that manifest as shorter lifespan. This would explain why bees that learned faster foraged for less days than bees that learned slower. But we generally think that learning should increase the foraging efficiency of bees. Why did the fast learning bees not bring back more nectar and pollen on their foraging trips than slow learning bees? Well, there are a number of possibilities. It could be that this "learn to avoid blue" assay is just not a relevant measure of learning for how bees use learning in the field. Maybe a test of learning to handle flowers or extract pollen would be a better measure. Another possibility is that the field conditions these bees were foraging in was just not challenging (or variable?) enough for learning abilities to produce variation in nectar and pollen collection. Maybe the pickings are so good that it doesn't matter how fast a learner you are, you can always collect plenty of food. The shortened foraging lifespan of bees that are faster learners is the most intriguing part of this research for me. We expect maintaining brain tissue to be costly, but measures of the costs of cognitive abilities for fitness are few and far between. Additionally, I think any study that measures the learning abilities of an animal using a controlled lab experiment and then ties that to foraging performance in the wild is taking behavioral ecology in the right direction!

Venom in Fangblennies

Patricia Jones

A shorthead fangblenny looking innocent. Photo by  Paddy Ryan . 

A shorthead fangblenny looking innocent. Photo by Paddy Ryan

The fangblenny reveals its fangs! Photo by  Serge Abourjeily .

The fangblenny reveals its fangs! Photo by Serge Abourjeily.

Blenny is a general term that refers to six families of fish (the suborder Blenniodei), most of which are cute little elongated fishes with big eyes and mouths. The fangblenny (or sabre-toothed blenny), however, is another story! The fangblennies encompass many species in 5 genera, only one of which (Meiacanthus) has venom as well as fangs. This week's paper on the evolution of fangs and venom in blennies is in Current Biology, lead authored by Nicholas Casewell from the Liverpool School of Tropical Medicine. 

What they did

Casewell and his colleagues first created a phylogeny of the fangblennies using five molecular markers. They then conducted extensive morphological analyses of the tooth and venom gland structure in each fangblenny species. Next they analyzed the composition of the fangblenny venom. 

What they found

First off, they were able to confirm that in blennies the evolution of the fang structure proceeded the evolution of venom. The ability to bite appears to function in blenny defense, as fish that have eaten fangblennies have been seen to spit them out (unharmed). In accordance with this defense hypothesis, most fangblennies have bright, distinctive, aposomatic coloration which may have evolved because a distinctive looking fish that gets eaten, bites, and is spit back out is more memorable and less likely to get eaten again. 

They also showed that one of the components of fangblenny venom is an opiod hormone that functions by binding to opiod receptors in the the bitten individual causing numbness in that area. Another component of blenny venom reduces blood pressure. This is very different from the effects of the spine venom that is most common in fishes. Spine-venom is quite painful, and therefore appears to function as a defense. So why would fangblenny venom numb and relax the bitten animal? Apparently it allows the fangblenny to get away while the bitten animal is impaired by numbness and lethargy. It is interesting that venom has evolved along two very different pathways (pain versus numbness) that both function in defense. 

The takeaway

One of the coolest things about this group is their mimicry. The venomous blennies are only one genus, and blennies from other genera look and swim like these venomous blennies, presumably allowing them to be sheep in wolf's clothing. One genus of fangblennies (Plagiotremus), however, is what the authors call an "aggressive mimic". These fangblennies feed by swimming up to other larger fishes and taking little bites out of them (or micro-predation). Some of these blennies mimic cleaner wrasse, so the victim fish might think they are about to be cleaned, when instead they get nipped. Others of these blennies, however, look like the venomous blennies. This affords them protection against the fish they are biting, because the bitten fish does not want to mess with a venomous fangblenny. 

New phylogeny of the hymenoptera

Patricia Jones

Some representatives of the Hymenoptera, the sawflies, ants, wasps, and bees. Photo by  Alex Wild .

Some representatives of the Hymenoptera, the sawflies, ants, wasps, and bees. Photo by Alex Wild.

This week's paper is in Current Biology, lead authored by Ralph Peters who is a museum curator at the Zoologisches Forschungsmuseum Alexander Koenig in Bonn, Germany. The Hymenoptera is an amazing group. The diversity in diet alone is staggering. Hymenoptera includes sawflies that eat leaves, predacious wasps, parasitoid wasps that lay their eggs in caterpillars, ants that farm fungus gardens, and bees that forage on nectar and pollen. Additionally, the Hymenoptera includes insects that vary from solitary to eusocial (one queen and lots of workers like a honeybee). Not to mention that while many Hymenoptera do not sting at all, some have venomous stings so toxic that they can debilitate a human. This variation makes the Hymenoptera an amazing group for studying evolutionary transitions from one foraging type to another, or from solitary to social lifestyles. The paper reports the most comprehensive phylogeny ( = evolutionary tree) of the Hymenoptera to date. 

What they did

This phylogeny is so comprehensive because of the amount of DNA sequencing that they did. They sequenced the entire transcriptome of 167 species of Hymenoptera. Early in evolutionary biology scientists made phylogenetic trees from morphological characters (how many teeth and how long they are are, etc). Species were then clustered based on how similar they were in these different traits because if they are more similar they are more likely to be closely related. With the advent of DNA sequencing scientists began doing the same thing but with gene sequence similarity. But as with morphological characters, the phylogeny is better the more different genes you include (like measuring teeth, and ears, and foot bones, etc). Most current DNA phylogenies are therefore constructed using multiple different genes, each of which evolves at different rates and in different groups. By sequencing the entire transcriptome this paper sequenced every gene that is transcribed (will be made into a protein), a technique that is increasingly being used to construct phylogenies. 

What they found

The Hymenoptera began to diversity 281 million years ago (for reference the first mammals were around 224 million years ago, so this is an old, old, group). The Hymenopterans first began when a leaf-eating insect diverged from a predacious insect. This herbivorous diet may have opened up new niches for Hymenoptera, allowing them to greatly diversify. Next Hymenoptera evolved parasitoidism in which females lay their eggs inside the bodies of caterpillars, and their larvae eat the caterpillar from the inside-out. The next big evolutionary step was the evolution of the distinctive "wasp waist" body shape that we associate with most wasps. The narrow waist allows insects more maneuverability of their abdomen independent from their thorax. It may have given parasitoid wasps more maneuverability when laying their eggs in caterpillars. The next transition was the evolution of a stinger. The stinger may have evolved in parasitoids because it allowed them to sting and immobilize the caterpillars that would be prey for their offspring. The stinger may have also enabled the evolution and diversification of many predacious Hymenoptera that could now attack and disable prey that they could not have otherwise subdued. Once the stinger had evolved, eusociality evolved repeatedly in the Hymenoptera. I wonder if the evolution of a stinger as a defense mechanism enabled the group living of large numbers of individuals which might otherwise be susceptible to attack from predators. The evolution of collecting pollen as a source of protein then led to further diversification, especially in the bees. The authors state that "the switch from a predatory to a herbivorous lifestyle was a key to the tremendous diversification of bees". 

The take away

This is certainly not the first phylogeny of Hymenoptera, but it is the most extensive. Phylogenies of this group are so important (and interesting!) because it contains so much amazing biological diversity and innovation. The evolution of such a range of foraging strategies, venom, and complex sociality makes the Hymenoptera particularly unique. Having a good phylogeny allows us to speculate and develop hypotheses about the causes of major evolutionary transitions that can be tested empirically. 

Landscape-scale floral resources and bumblebee survival

Patricia Jones

Red-tailed bumblebee,  Bombus lapidarius . Photo by Lucy Hulmes.

Red-tailed bumblebee, Bombus lapidarius. Photo by Lucy Hulmes.

This week's paper is in Nature, lead authored by Claire Carvell who is a Senior Ecologist at the UK's Natural Environment Research Council's (NERC) Centre for Ecology & Hydrology. Bumblebees are important pollinators for many wild and agriculturally important plants (think apples, strawberries, blueberries, tomatoes, etc.). There has been a global decline in bumblebee populations over the recent decades, likely caused by multiple factors that include disease, pesticides, (and most relevant here) agricultural intensification. As agricultural land use becomes more and more uniform (think those vast fields of wheat or corn), the diverse array of floral resources that bees need becomes less and less available. Other studies have examined how flowering strips or hedgerows impact bumblebee density and pollination services to flowering crops. This study addressed the same question but they investigated the effect on bumblebees across generations. 

What they did

In Buckinghamshire, England there is 1000 hectares of farm land with edges sown with wildflower seeds such that each 50 hectare block has 0% to 8% coverage of wildflowers. In this habitat Carvell and her colleagues studied three bumblebee species Bombus terrestris, B. lapidarius, and B. pascuorum. Across this area in 2011 they captured bumblebees of all three species, collected DNA samples, and then used the genetic information to determine which bees were from the same colony. Using the collection locations of worker bees they were able to determine the most likely colony location. They assessed the area surrounding each colony location for floral resources. The next spring 2012, they captured the emerging queens who were out looking for new nest sites, collected their DNA to determine which colony they had come from, and determined how far they had dispersed from that colony.

What they found

When there were more flowering resources surrounding a colony (within 250 meters, 500 meters, and 1000 meters), more of that colony's queens were seen the next spring. They refer to this as "family lineage survival" because it is across generations. When workers are able to forage on flowers close to their colonies, they are able to collect more resources and produce more new queens. Colonies that have to forage further for food do not produce as many new queens. Interestingly, the distance that queens dispersed was not affected by flowering resources, but was affected by nesting habitat availability. When there was more suitable nesting habitat 250, 500, and 1000 meters from their parent nests queens dispersed further. This is a bit counterintuitive, you would think that if there was more suitable nesting habitat close to home queens would not have to disperse as far. But the authors suggest that the nesting habitat "facilitates" dispersal, perhaps by providing shelter for queens as they disperse?

The takeaway

I really like the way that this study used genetics to look at performance of whole colonies, and the next year's queens. That is a cool step from other studies that have just looked at numbers of individual bumblebees. It is always reassuring that something can be done to mitigate species declines, the next task is providing land-owners with suitable wildflower seeds, especially local genotypes of native species!

Cannibalism and care in poison frogs

Patricia Jones

An  Allobates femoralis  carrying tadpoles on its back. Photo by  Seabird McKeon . 

An Allobates femoralis carrying tadpoles on its back. Photo by Seabird McKeon

It doesn't get much cooler than male parental care in poison dart frogs. This week's paper is in Scientific Reports, lead authored by Eva Ringler at the University of Veterinary Medicine Vienna, Austria. Allobates femoralis is distributed across the Amazon basin of South America. Paternal care is widespread (and thought to be ancestral) in the poison frogs. Males defend territories in the rainforest where females come to mate with them, and lay their eggs. A female will lay her clutch of eggs surrounded by jelly in the leaf litter on the forest floor. The eggs will develop for three weeks at which point the male will move the tadpoles to water sources. Males move tadpoles on their backs to little pools made by the leaves of plants such as bromeliads. Particularly relevant to this study is that males will even move tadpoles from clutches that are not their own, but were placed inside their territory. 

This study examined how male A. femoralis respond to the presence of unrelated clutches in their own territory, in comparison to males that are moved to a new territory (representing a territory "takeover"). The design was really simple and detailed in the picture below. 



In the 'resident' treatment (A) males were removed from their home terrarium, an unrelated clutch was placed in their terrarium, and then they were returned. In the 'takeover' treatment (B) males were removed from their home terrarium and placed in a new terrarium with an unrelated clutch of eggs. So in both cases males experienced being captured and were given an unrelated egg clutch. They then recorded what males did with those eggs. Males in the 'takeoever' treatment were more likely to eat the eggs than males in the 'resident' treatment, and males in the 'resident' treatment were more likely to transport the tadpoles. The authors propose that males follow a simple strategy that allows them to maximize their fitness while minimizing energetic and cognitive costs. Clutches in their own territory are most likely their own, so they should care for them. Clutches in a new territory are certainly not their own, so they should eat them. Given that males defend their territories from invading males and only mate with females on their territories, they may not experience selection to be able to identify their own eggs from those of other males. It would be interesting to see if this type of territoriality trades off with offspring recognition in other systems. 

Social learning strategies in house-hunting ants

Patricia Jones

Ants leading their nest-mates to a new colony using "tandem runs". Photo by Thomas O'Shea-Wheller. 

Ants leading their nest-mates to a new colony using "tandem runs". Photo by Thomas O'Shea-Wheller. 

A new social learning strategy paper is always exciting! Especially when it uses a less common system such as the ant, Temnothorax albipennis. This week's paper comes from Nigel Frank's lab at the University of Bristol, lead-authored by Nathalie Stroeymeyt, and is in Scientific Reports

The "teaching" ant

Temnothorax albipennis is a tiny European ant species that live in cracks in rocks, or even can make homes in hollow acorns (how adorable!). The latin name, albipennis, means "white feathers", I assume because the ants' bodies have a smattering of white hairs.  This ant species has become well known because of it's extraordinary behavior when searching for and moving to a new nest. Their fragile acorn homes are frequently destroyed, forcing a colony including its brood (little larval ants that can't do much on their own) to move to a new nest. Scout ants go out and search their environment for a new nest. When they find a possible new home they return to their old nest (or the ruins of it) and attempt to "recruit" other scouts to come and visit the nest they have found. They do this through "tandem running" which involves one ant leading the way while the other ant follows right on its heels (so to speak). The ant that leads goes slowly, and only proceeds when the following ant taps its back legs with their antennae. Ants that have followed in a tandem run then are able to recruit other ants to the same location. This behavior is so cool because it is one of very few examples of teaching in animals. In order for a behavior to be designated teaching in a strict sense, the "teacher" individual must modify their behavior at some cost to themselves. They cannot be simply doing whatever they like while the student watches. Because leader ants in tandem runs travel distances more slowly and only proceed when tapped by following ants, this behavior qualifies as teaching. 

Once an ant has visited a new home it then uses tandem running to bring other scout to the new home. When the number of ants in the new home (there through individual exploration or recruitment) reaches a threshold number (this is a form of quorum sensing) , then the ants go back to their old nest and start carrying other nest mates and brood over to the new nest. Because ants only recruit other ants if they find the new potential nest satisfactory, this quorum sensing is a means for insect groups to make a group decision based on a conglomeration of the available data each ant has collected. It is also used by honeybees when deciding where to make a new nest (although it is slightly different in honeybees). 

Social learning strategies

Social learning strategies are the strategies that animals employ when deciding whether to use social information (information acquired from others) versus private information (information acquired through personal experience). I often use an airport food court as an analogy. Let's say you walk into the food court section of an airport in a foreign country where you do not speak the language. There are lots of options of where to eat, but you cannot read the signs or the menus, so how do you decide? What if most of the restaurants are empty, but one has 20 people eating there? I would go an check out the place where there are other people eating, with the assumption that these people have information I do not about the superior quality of that restaurant. This risks the danger of an "information cascade" where everybody is relying on other people in making their decisions and therefore all eating crappy food, but when you have no personal information the presence of other people may be your best indicator of food quality. In contrast if you walk into a familiar airport where you have eaten in some of the restaurants before, you may ignore the other people (or even avoid them! because they are competition after all!) and eat at the place you know that you like. This social learning strategy is called "copy when uncertain" because you only use the social information when your personal information is "uncertain". 

This paper

This week's paper examined whether ants use the social learning strategy of "copy when uncertain" when deciding to move to a new home. They set up the ant colony in an arena so while still living in their old home they could explore a potential new home ("new home 1"=NH1). When the researchers destroyed the old home, forcing the ants to move, they added in a second potential new home (NH2) that the ants had not had the opportunity to explore. They then recorded the proportion of ants visiting each of the new homes that were "transporting" other ants as a function of how many ants had been in the NH the last time they were there. Ants that had been to NH1 many times before quickly started transporting nest mates to NH1 regardless of how many other scout ants they saw at NH1. In contrast, ants that had made only a few previous visits to NH1 waited until there were more other ants present at NH1 before beginning to transport ants. These ants were dependent on social information (the presence of other ants) because they had less personal information. Similarly, ants visiting NH2 did not start transporting other ants until there was a quorum of other ants present at NH2. Essentially the number of visits an ant has previously made to a nest decreases the quorum sizes (number of ants present) necessary for that ant to start carrying its nest mates to the new nest. If an ant has been to a nest many times before, there do not have to be a lot of other ants there for it to decide to start moving its nest mates, but if it has never been before, or only a few times, there have to be more ants there for it to decide to start moving. A "copy when uncertain" strategy! 

    Ball-rolling bumblebees

    Patricia Jones

    Bumblebees for two weeks in a row! This week's paper comes from the Chittka Lab at Queen Mary University, lead-authored by Olli Loukola, published in Science. As you can see in the video above (which was provided in their online supplement), they have trained bumblebees to roll little balls into target holes upon which they receive a bit of sugary nectar solution. This paper has two parts which I will detail in different sections here:

    Part I: Learning the task

    The first section of this study simply examined whether bumblebees could learn this ball-rolling task. First, bees were trained to drink sugar water from a yellow ball that was stationary within the goal area. Then the ball was displaced from the goal area, requiring the bees to roll it into the goal to get the reward. Bees that did not move the ball into the hole on their own in the first few attempts had the ball-rolling solution demonstrated for them by a model clay bee on a stick that pushed the ball along. The author's don't say how many bees were able to solve the ball-rolling task without a demonstration, I suspect not many, but I am curious. The nine bees that made it through the 30 training trials were all able to solve this task in all 10 of the test trials. Not only that, but over the course of the training trials the bees got more efficient at solving the task, getting the ball in the goal faster and over shorter travel paths. 

    Part II: Social Learning

    The second part of this study examined how bees learn this task. Naïve bees were exposed to one of three scenarios: 1) social demonstration - in which a demonstrator bee which had already been trained to solve the task rolled the ball into the goal and then both bees got sugar rewards from the ball. 2) "ghost" demonstration - this was the same as (1) except that instead of another bee moving the ball the experimenter moved the ball into the goal using a magnet in the ball and a magnet held under the table. 3) control (no demonstration) - the naïve bee entered the arena to find the ball already in the goal with a sugar reward. For the social and ghost demonstrations there were three balls in the arena, and the demonstrator moved the ball that was furthest from the goal. In the 10 five-minute test trials that followed bees could move any of the three balls. 

    Bees performed the most correct trials when there was social demonstration, then in the ghost demonstration, and the least number of correct trials in the control. Although the demonstrator always moved the ball that was furthest from the goal, the experimental bee would usually move the ball that was closest to the goal. This was even the case in an additional experiment where they made the closest ball black instead of the yellow that they had seen demonstrated. The bees therefore are able to generalize this ball rolling task from the yellow ball they had seen demonstrated to a ball of a different color. 

    So what? 

    So why should we care that bees can learn to move balls into goals? Well, I think for a couple of reasons. In a general sense this is yet another demonstration of the extraordinary cognitive feats that insects are capable of. What is particularly interesting about this example is that is more distant from normal bee foraging tasks. Many studies have demonstrated bee learning of flower colors, shapes, scents, and even electrical fields. Additionally, studies have shown bee learning of fairly complicated flower handling skills where they have to manipulate flowers to access nectar or pollen. What makes this task different is that they have to go away from the award location (the goal area) to locate the ball and roll it into the goal. It is more similar, therefore, to tool use than other studies on bee cognition. The authors, however, never use the term tool-use. I suspect that this study does not qualify because they first trained the bees to associate the ball itself with nectar rewards. Generally the tool itself is not associated with rewards in studies of tool use. I think the bees' choice to move the closest ball, and the black balls when those were not what they had seen demonstrated, is the most interesting result in this study. This result shows the ability of bees to generalize solutions to problems and apply them in different ways. I think this type of innovation on demonstrated behavior has the most important implications for our understanding of bee problem-solving.

    Foraging specialization in bumblebees

    Patricia Jones

    A foraging eastern bumblebee,  Bombus impatiens . Photo by  Michael Motto . 

    A foraging eastern bumblebee, Bombus impatiens. Photo by Michael Motto

    One of the advantages of living in a social group is division of labor, or task specialization. An animal living on its own must collect all of the resources necessary for its survival, as well as find shelter, find a mate, and (in some cases) rear its young. An animal living in a group can specialize on one or a few of these tasks while other group members perform the other necessary tasks. The eusocial bees, such as honeybees and bumblebees, are an extreme example of social living. Accordingly, it has been proposed that there is extensive task specialization in bees, even just within foraging. Bumblebees are believed to specialize on either foraging for nectar or foraging for pollen, and within each foraging type they may specialize on particular plant species or patch location. Specialization is thought to be particularly advantageous because learning how to forage for nectar or pollen, and learning different flower types, takes time. Specializing, therefore, makes bees more efficient foragers. It is unknown, however, over what kinds of timescales this specialization occurs. Additionally, there are aspects of bee physiology that may influence how bees specialize. The sensitivity of the sensory systems of honeybees determines what they specialize on when foraging. Individual bees with more sensitive olfactory systems for pollen, and bees with less sensitive olfactory systems forage for nectar. Bumblebee workers vary broadly in size. Larger bees are more efficient foragers, and also have larger olfactory sensory systems, but it is unknown how these factors influence bumblebee specialization.

    This week's paper addresses the timescale over which bumblebees specialize, and the role of physiology in specialization by bumblebees. It is in Scientific Reports, lead authored by Avery Russell at the University of Arizona. Russell and colleagues examined the nectar and pollen foraging behavior of individual foragers within a colony over their lifespan. To do this they attached a colony to two foraging chambers through a branched tunnel. Individual bees leaving the colony could therefore make a choice in the branched tunnel to go to the foraging chamber that always had pollen or the chamber that always had nectar. To record the choices of individual bees there was an radio frequency identification (RFID) reader at the entrance to each foraging chamber, and a 1.5 × 1.0 × 0.5 mm RFID tag was glued to the back of each foraging bee. The RFID system then conducted automated data collection of which bees entered and exited which foraging arena over their lifespans. Once a bee died, Russell and colleagues measured a wide range of morphological characteristics of that bee including body size and the number of sensory receptors on the antennae. 

    The authors found that most bees were generalists over their lifetime (foraging for both pollen and nectar) but over the course of a day or two, many bees (about 50%) would specialize on one food type. Additionally, the work of foraging was performed predominantly by a minority of bees. 17% of bees performed half of the foraging bouts on a daily basis. Bees that more olfactory sensory receptors foraged more for pollen over their lifespan, but neither olfactory receptors nor body size influenced the extent of specialization demonstrated by bees. 

    Specialization on a day to day basis but not over a lifetime could be ideal if the needs of the colony changes over time. As brood size increases there is more need for pollen to feed developing larvae, whereas as colonies age there is less brood and thereby need for pollen. Additionally, the resources brought back by other foragers may influence colony needs. Specializing for a day or two may enable a bee to forage efficiently but also update foraging behavior with colony needs. This is such a cool experimental design to study the long term foraging choices of individual bees, with a lot of potential to address a wide range of questions!

    Long-term memory for food quality in tortoises

    Patricia Jones

    The red-footed tortoise,  Chelonoidis carbonaria . Photo by  Paul Jones .

    The red-footed tortoise, Chelonoidis carbonaria. Photo by Paul Jones.

    I am fascinated by how animal cognition can shape ecological communities. While my research is generally focused on learning, memory is another (inherently connected) aspect of cognition. Memory allows animals to retain information about food sources, roost or burrow locations, and social partners across time periods when they may not encounter those resources or individuals. If the environment changes in a consistent way, such as seasonally available food, memory may allow an animal to more quickly take advantage of a resource as it becomes available again. This week's paper is lead-authored by Francesca Soldati at the University of Lincoln in the UK. Soldati and colleagues examined long-term (18 months) memory in captive red-footed tortoises, Chelonoidis carbonaria

    Red-footed tortoises are native to the northern South America, from Panamá through northern Argentina, and common as pets worldwide. Wikipedia says in reference to the coloration on their legs "the species name carbonaria means 'coal-like' referring to a dark coal with glowing patches" which is just lovely. They are omnivorous, but in particular consume large amounts of fruit. Red-footed tortoises are becoming a model system for studying reptile learning, as they quickly learn associations between colors and food, socially learn from other tortoises, and can be trained to use touchscreens to solve spatial tasks. On an irrelevant side note, they do not, however, exhibit contagious yawning

    So what food to red-footed tortoises particularly care about the quantity and quality of? Fruit jelly of course. And in case it wasn't obvious, mango is better than apple. Soldati and colleagues trained tortoises to associate different colored cards with mango vs. apple jelly cubes (quality) and with different sizes mango jelly cubes (quantity). 18 months later the tortoises were shown the same colored cards, and exhibited the same preferences for colors paired with mango jelly and larger jelly quantities. [As an aside, I am a bit concerned about the Clever Hans phenomenon in these experiments. It would be nice to know that experimenters were blind to what the tortoises were trained to when they conducted the tests.] 

    18 months is longer than the time between fruiting for most plants in the native tortoise habitat. The authors propose that memory for high quality or large fruits may be retained by tortoises across seasons when those fruit are not available, potentially increasing the frequency or speed with which tortoises visit fruiting trees as they come into season. The amount of frugivory by tortoises may in turn have consequences for seed dispersal of the plants. This postulation raises a lot of further questions: how far does a tortoise range over 18 months? what is the density of fruiting trees of various qualities? do they remember the locations of fruiting trees across seasons in order to revisit them when they begin to fruit? do they associate the fruit quality with other seasonal cues enabling them to revisit high quality fruiting trees at the right time of year? It is known that red-footed tortoises exhibit spatial memory in captivity (memory for the location of food in a maze). Studying the ways in which tortoises use memory in the wild is important (albeit difficult) research to understand the effects memory may have on seed dispersal. 

    Frog saliva is a non-Newtonian fluid

    Patricia Jones

    This week's paper is about frog saliva. It is in the Journal of the Royal Society Interface lead authored by Alexis Noel at the Georgia Institute of Technology. A non-Newtonian fluid is a fluid with a non-linear relationship between its sheer stress and shear rate. Another explanation is that non-Newtonian fluids "change their viscosity or flow behavior under stress". Ketchup is the classic example of a non-Newtonian fluid, but others include blood, honey, cream, and the infamous oobleck. This paper adds frog-saliva to the list. 

    The authors used a combination of high-speed video of frogs and toads catching insects, and biomechanic studies of the physics and chemistry of frog tongues and saliva. They have some great quotes such as "we laboriously collected 0.3 ml of saliva from 15 leopard frogs and placed it into a cone and plate rheometer". They show that when frog saliva hits the prey insect its viscosity decreases, allowing it to spread over the insect and into all the nooks and crannies of the insect exoskeleton. This drop in viscosity is achieved by the speed of the frog's tongue, which creates very fast saliva sheer rates. To swallow the insect, the frog retracts its eyeballs (yes! really!) which again creates a sheer force parallel to the tongue, pushing the insect off the frog's tongue and down its throat. But how does the frog keep the insect stuck to its tongue as it flips its tongue back in its mouth? During this retraction, the force is pulling perpendicular to the tongue at which the viscosity of the saliva increases so it stiffens, holding tight to the insect. A non-Newtonian fluid! 

    Convergent evolution in pandas

    Patricia Jones

    The giant panda Tai Shan in the Smithsonian National Zoo. Photo by Jessie Cohen. 

    The giant panda Tai Shan in the Smithsonian National Zoo. Photo by Jessie Cohen. 

    The red panda. I cannot find a source for this photo. But it is so excellent! I apologize. 

    The red panda. I cannot find a source for this photo. But it is so excellent! I apologize. 

    This week's paper is from PNAS, lead authored by Yibo Hu at the Chinese Academy of Sciences in Beijing. The authors studied the genetics behind the extraordinary similarities of giant pandas (Ailuropoda melanoleuca) and red pandas (Ailurus fulgens) which have overlapping distributions in the Sichuan province of China. Both giant and red pandas are herbivores that consume almost exclusively bamboo. They also both have appendages on their forepaws called "pseudothumbs" that are projections from bones in the wrist (technically called an enlarged radial sesamoid), allowing them to grasp stalks of bamboo. What is particularly surprising is that these two "pandas" are not each other's closest relatives. Rather, the giant panda is most closely related to other bears, and red pandas are most closely related to skunks, raccoons, and weasels. This means that giant and red pandas separately evolved from a meat-eating ancestor (after all they are both in the order Carnivora). The evolution from meat-eating to bamboo-eating requires a large range of physiological changes. When two organisms that are not closely related independently evolve similar traits it is called convergent evolution. To study the convergent evolution of pandas, Hu and colleagues sequenced the entire genomes of both panda species. 

    The authors estimate that giant and red pandas shared a common ancestor some 47. 5 million years ago. To put that in perspective, humans and chimpanzees last shared a common ancestor around 10 million years ago. The authors found 70 genes that convergently evolved in the two pandas. For the majority of these genes the function is still unknown, but two of them are involved in the development of the pseudothumb, and three genes are involved in the digestion of dietary protein. They also found four convergent genes that function in the utilization of vitamins. Not only have the pandas acquired genes in common, they have also lost a total of 10 genes in common. The gene for the umami taste receptor (often described as savory taste receptor, but involved in the detection of meat flavors) is non-functional in both pandas. 

    Few studies have used entire genomes to study convergent evolution. But as high-throughput DNA sequencing advances in speed and affordability it is likely to become more and more common. 

    The echolocating Vietnamese pygmy dormouse

    Patricia Jones

    This week's paper is lead authored by Aleksandra A. Panyutina and is in the journal Integrative Zoology. It is a documentation of the use of echolocation by the Vietnamese pygmy dormouse, Typhlomys chapensis. Echolocation has been best demonstrated in bats and the toothed whales (especially dolphins). Beyond these groups the authors state that "no other mammal has been shown to rely on echolocation as the main means of echolocation". But there are some other mammals that use something like echolocation. For example, shrews "twitter" to assess their habitat, but it does not appear to be an advanced bat-like echolocation. What makes these Vietnamese pygmy dormice so unusual is that they have very reduced eyes (they are nocturnal) but are fast and nimble tree climbers. This study has two components. They 1) dissected the eyes of dormice to assess their vision, and 2) they recorded dormice exploring their cages using high-speed video and ultrasound recordings. They show that the eyes of these pygmy dormice are highly reduced, with the retinas actually have irregular folds in them. They state that therefore the pygmy dormouse "has no other means for rapid long-range orientation among tree branches other than echolocation". They then recorded the behavior of these pygmy dormice when they are exploring branches in their cages. In the video above you can see them exploring and hear their echolocation calls.

    The echolocation calls made by these dormice are so faint that bat detectors cannot even pick them up. Additionally their echolocation calls are such high frequency (93 kHz) that they are above the hearing of most small rodents. This paper is exciting because there are so few species that we know to echolocate. Additionally, evidence of an arboreal mammal that echolocates adds an interesting point to the debate about whether echolocation or flight evolved first in bats. A non-volant echolocating mammal indicates that it is possible echolocation evolved first. It was long ago debunked that bats are rodent relatives, so they would not have evolved from something similar to the Vietnamese pygmy dormouse, but rather a very different arboreal mammal. It is particularly exciting that their echolocation is so quiet and directional that it is very hard to detect. This means that there may be other echolocating mammals out there that we have overlooked. 

    So you're an unattractive fairy wren...

    Patricia Jones

    An "attractive" male red-backed fairy wren,  Malurus melanocephalus . Photo by  Greg Miles .

    An "attractive" male red-backed fairy wren, Malurus melanocephalus. Photo by Greg Miles.

    To start we have to point out that you are still pretty darn cute. This week's paper is by Jenélle Dowling and Mike Webster here at Cornell, published in Biology Letters. Red-backed fairy wrens live along the northern coast of Australia. The older adult males have the striking black and red plumage you see in the photo above, while the females and younger males are tawny brown. One of the cool things about this species is that males turn red-black at different ages, so there are brown males that are the same age as red-black males. Females prefer red-black males to brown males, but brown males still form pairs and raise offspring. Fairy wrens, however, have quite a bit of so-called "extra-pair paternity" or EPP. Male fairly wrens sneak off to mate with other females, but when they do they leave their own female un-guarded to mate with somebody else. Males therefore face a trade-off between gaining paternity elsewhere and losing paternity at home. 

    Dowling and Webster predicted that the more attractive red-black males would spend more of their time investing in attempting to get EPP's. In contrast less attractive brown males would spend more time mate-guarding their females at home. Additionally they predicted that the younger red-black males would take an intermediate tactic. When they watched the behavior of fairy wrens in the field this is exactly what they saw. They showed paternity corresponded to behavior in that brown males had more offspring within their mate pair and red-black males had more offspring outside their mate pair. This is a lovely example of how behavioral flexibility can shape the costs and benefits of sexual signals. 

    Nectar feeding bats and evolving robotic flowers

    Patricia Jones

    Nectar foraging  Glossophaga commissarisi . Photo by  Merlin Tuttle . 

    Nectar foraging Glossophaga commissarisi. Photo by Merlin Tuttle

    Bats for two weeks in a row! Huzzah! This week's paper is in Science, lead authored by Vladislav Nachev from York Winter's group at Humboldt University in Berlin. They addressed how animal perception can influence the evolution of another species. In general, nectar feeding animals prefer flower nectar with a higher sugar concentration. Surprisingly, then, plants generally produce nectar with fairly low sugar concentrations. Why are the preferences of pollinators not driving plant evolution of more concentrated nectar in order to receive pollination services? 

    To answer this question Nachev et al. set up an array of 23 robotic flowers in the jungle at the La Selva Biological Station in Costa Rica. A network of tubing provided each flower with nectar of a predetermined concentration and volume. This array looked pretty much like lots of guinea pig water bottles hanging from the roof of an open pole barn. They captured 16 nectar feeding bats (Glossophaga commissarisi) and tagged them with RFID tags (see chickadee paper I discussed previously). The flowers were equipped with sensors that read the RFID tags and recorded which bat visited them and for how long. Each flower produced the same amount of sugar, but the amount of water added to that sugar (and thus the concentration and volume of the nectar) varied depending on its virtual "genome". Each flower had a diploid set of four virtual genes each with two possible co-dominant alleles (versions of a gene) that controlled the amount of water in nectar. They started with a set of plants of different genotypes, some with high sugar concentrations in nectar (42.2%) and some with low sugar concentrations (17.8%). When a bat visited a two flowers in a row, the computer determined a pollination event and created a "seed" for those two plants. 23 of one night's seeds were randomly selected to be the flower genotypes (with associated nectar concentrations) for the next night. Flowers that received more visits therefore made more seeds, and more of their seeds would be present as flowers the next night. After 10-12 generations (nights) all of the flowers had converged in nectar concentration at around 36%. Why did they not converge at the highest concentration possible? Because there is a trade-off between concentration and volume, and the perceptual biases of bats makes them more selective about volume (see below). 

    Although different flowers produced different nectar volumes dependent on their genotype, flowers do not fully refill after every visit. Therefore when there are lots of bats foraging, an individual bat may encounter a flower that is partly drained already. This is where the bat perception plays in, in the form of what is called Weber's law. Weber's law has to do with the ability to distinguish amounts. If you have two candies, and somebody adds one, you will probably notice. If, however you have 60 candies and somebody adds one you mostly likely will not notice. That is to say, we are much better at distinguishing proportional differences than real differences, so we are best at distinguishing real differences for very small amounts. This means that when there are lots of bats foraging, so bats are encountering smaller nectar volumes, they are better at distinguishing these nectar volumes, and therefore more selective for higher volumes which results in lower sugar concentrations. 

    I am really excited about this paper, because it demonstrates how the brain's perception of the world can have consequences for the evolution of other species. Plant-pollinator relationships are an ideal system in which to study this, but the same processes are also occuring for other biological relationships such as predation, competition, mutualism, parasitism etc. 

    What we can learn from squabbling bats

    Patricia Jones

    Egyptian fruit bat,  Rousettus aegyptiacus . Photo by Eran Levin.

    Egyptian fruit bat, Rousettus aegyptiacus. Photo by Eran Levin.

    This week's paper is in Scientific Reports and comes from Yossi Yovel's group at Tel Aviv University, lead authored by Yosef Prat. If you have spent any time around bats, you are well aware that they do a lot of chattering. The Yovel group tackled the question of what is all that chattering about? To do this they video and audio recorded the behavior of 7 female Egyptian fruit bats, Rousettus aegyptiacus, (in a group of 22) over 75 days, totaling almost 15,000 individual vocalizations. They then used a machine learning approach to assess variation and information in bat calls. They were able to identify calling individuals with 71% accuracy, indicating that bats have enough variation in their calls to tell them apart. Now it starts to get really cool. Not only could the machine learning identify which individuals were calling, it could also detect differences in whom they were addressing. That is bats make slightly different calls when they are addressing different bats. Some of this variation is due to sex, they make different calls when they are talking to males versus to females, but also they are making slightly different calls when talking to different individuals within a sex. This means that an eavesdropping bat could potentially tell not only which bat is talking, but also which bat they are addressing. What the authors don't clarify is how repeatable these individual differences are across vocalizing bats. That is, are all of the bats referring to an individual using particularly vocal variants, like a name? Or do different bats use slightly different "names" for each of their roostmates? The authors notes that there was one bat who did not vary in their vocalizations depending on whom they were addressing. That one jerk who is yelling indiscriminately at everybody. 

    The authors then looked at the contexts in which bats were making vocalizations and showed that bats make different calls in different types of arguments (apparently Egyptian fruit bats mostly argue). The vocalizations they make are different when squabbling about food, or sleep space, or when somebody is making unwelcome sexual advances. The authors are also able to predict from the vocalizations what the outcome of the squabble will be. It would be great to know more about this, are the calls of the bats who win the squabbles louder, deeper, more chaotic?

    Regardless of my desire for more detail, this shows how much can be gained just from making lots and lots of careful observations.

    Tradeoffs between learning and memory in mountain chickadees

    Patricia Jones

    Mountain chickadee,  Poecile gambeli . Photo by  Nick Athanas .

    Mountain chickadee, Poecile gambeli. Photo by Nick Athanas.

    This week's paper is in Animal Behaviour and comes out of the Pravosudov lab at the University of Nevada at Reno, lead authored by Rebecca Croston. Mountain chickadees, Poecile gambeli, live in the Western USA and Canada where they occur from low elevations up into the Sierras. Previous research from the Pravosudov lab has shown that the high elevation chickadees (which experience harsher climates) have better spatial memory and larger hippocampi (the brain region associated with spatial memory), than low elevation chickadees. This is thought to be because high elevation chickadees must cache more food to make it through the winter, and therefore they must have better spatial memory to locate all those food caches. In this particular paper the authors examined the "cognitive flexibility" of high and low elevation chickadees. Chickadees were captured in the field and fitted with individual PIT tags. These PIT tags were registered by feeder boxes that identified and recorded the bird that landed, and determined whether that individual is allowed access to food. (As a side note, this PIT tag and feeder technology is such a cool way that field research on animal behavior is being transformed right now!) The authors put arrays of 8 feeder boxes out in the field, and chickadees only got food when they landed on their assigned "correct" box. They therefore couldn't follow each other to the correct feeder box, because the PIT tag system assigned the correct box to be different for each bird. After a bird learned the correct box, it was then tested for behavioral flexibility using a reversal task, in that the "correct" box was changed to a different box in the array.

    The authors found that high and low elevation chickadees performed equally on the initial learning task (which is contrary to their previous reported results that high elevation chickadees perform better at spatial learning tasks), and high elevation chickadees are much worse at the reversal learning task. The authors suggest that selection for spatial memory abilities and the associated larger hippocampi, may be a trade-off with cognitive flexibility. Perhaps the high elevation birds are really good at remembering initial food cache locations, but this limits their ability to learn novel caches. I don't know that I buy this. The authors say that it was a harsher year than other years, as a potential explanation for the lack of a difference in initial learning between high and low elevation chickadees, which is incongruous with their previous research. It seems possible to me that this reduced reversal learning is also a consequence of some other factor. The concept that selection for memory may limit flexibility is, however, interesting. I hope the Pravosudov lab pursues this further to really pull apart what factors are at play. 

    Associative Learning in Plants

    Patricia Jones

    Pea plant,  Pisum sativum . Ink and watercolor by W. Giglioli. Hulton Fine Art Collection

    Pea plant, Pisum sativum. Ink and watercolor by W. Giglioli. Hulton Fine Art Collection

    Yes. That's right. You heard me. LEARNING IN PLANTS! And not just in some crazy special plant, but in the lowly garden pea! Although everybody else is talking about the feathery baby dinosaur tail discovered preserved in amber, I just desperately need to talk about learning in plants.  The paper is in Nature Scientific Reports led by Monica Gagliano from the University of Western Australia.

    But first let's talk about learning in general. In animal behavior there are two categories of learning: associative learning, and non-associative learning. Associative learning is the formation of an association between a stimulus to which you previously had no response (called the conditioned stimulus, or CS) and a stimulus to which you have an innate response (the unconditioned stimulus, or US). The classic example is the work of a Russian physiologist Ivan Pavlov, who together with his assistant, Ivan Tolochinov, developed the concept of associative conditioning in 1901. The Ivans' experiments showed that when you present a dog with a dish of food (the US) they will salivate (called the unconditioned response, or UR). When the presentation of the US is consistently paired with another stimulus, such as a sound (apocryphally the ringing of a bell in the Ivans' experiments, the CS), eventually the dogs will salivate at the sound of the bell. They have formed an association. Non-associative learning includes other forms of learning such as habituation and sensitization. Habituation is when an animal decreases it's response to a stimulus over repeated exposure (you stop jumping at every firework explosion after listening to them for twenty minutes) and sensitization is an increase in response over repeated exposure. 

    A 2014 paper also from Monica Gagliano demonstrated habituation in Mimosa pudica, which is often called "the sensitive plant". When touched, Mimosa folds its leaves down. There is a pretty silly video of people tickling these plants here. It has long been known that if you repeatedly touch a Mimosa plant it will stop responding, which is highly parallel with habituation learning in animals.

    But associative learning??! In this new paper Gagliano and colleagues used garden peas, Pisum sativum. The unconditioned response was that plants grow towards light. The conditioned stimulus was the breeze from a fan. They put a little pea seedling in a split PVC pipe y-maze (see below). In one treatment the pea plant would have the fan blow on it from one side of the y-maze for 60 minutes, and then a light would come on on that same side. In a second treatment the fan would blow for 60 minutes on one side and then the light would come on on the opposite side. In both treatments the location of the fan is a predictable indictor of where the light is going to be (the same side as the fan, or the opposite side). They switched the sides that the stimuli were presented on for each training session. To test the plants they just gave them the fan from one side. 

    "Figure 1. Training and testing protocol for associative learning in pea seedlings. (A) During training seedlings were exposed to the fan [F] and light [L] on either the same arm (i) or on the opposite arm (ii) of the Y-maze. The fan served as the conditioned stimulus (CS), light as the unconditioned stimulus (US). During testing with exposure to the fan alone two categories of responses were distinguished. Correct response: Seedlings growing into the arm of the maze where the light was “predicted” by the fan to occur [green arrow; iii (corresponding to scenario i) and iv (corresponding to scenario ii)]; Incorrect response: Seedlings growing into the arm of the maze where the light was not “predicted” by the fan to occur (black arrow; iii and iv). (B) Seedlings received training for three consecutive days before testing. Each training day consisted of three 2-h training sessions separated by 1-h intervals. The 90-min CS preceded the 60-min US by 60minutes so that there was a 30-min overlap. (i). During the 1-day testing session, seedlings were exposed to the fan alone for three 90-min sessions (ii). Seedlings of the control group were left undisturbed (no fan, no light; iii)." From Gagliano et al. 2016 .

    "Figure 1. Training and testing protocol for associative learning in pea seedlings. (A) During training seedlings were exposed to the fan [F] and light [L] on either the same arm (i) or on the opposite arm (ii) of the Y-maze. The fan served as the conditioned stimulus (CS), light as the unconditioned stimulus (US). During testing with exposure to the fan alone two categories of responses were distinguished. Correct response: Seedlings growing into the arm of the maze where the light was “predicted” by the fan to occur [green arrow; iii (corresponding to scenario i) and iv (corresponding to scenario ii)]; Incorrect response: Seedlings growing into the arm of the maze where the light was not “predicted” by the fan to occur (black arrow; iii and iv). (B) Seedlings received training for three consecutive days before testing. Each training day consisted of three 2-h training sessions separated by 1-h intervals. The 90-min CS preceded the 60-min US by 60minutes so that there was a 30-min overlap. (i). During the 1-day testing session, seedlings were exposed to the fan alone for three 90-min sessions (ii). Seedlings of the control group were left undisturbed (no fan, no light; iii)." From Gagliano et al. 2016 .

    The majority of plants grew in the direction they had been trained (toward the fan if the fan had been paired with light and away from the fan if the fan had been in the opposite arm from the light).  In a second experiment, Gagliano and colleagues showed that plants are only capable of this associative learning if they are trained during the "day" part of a light/dark cycle. That is, they can only learn to pair light with the fan cue during the time of day when they would normally be exposed to light. 

    My mind is a bit boggled by this paper. In particular I am really curious about what the mechanism is. Also I wonder if we need some different vocabulary? I am generally for maintaining simplicity, and I think that using the same vocabulary across diverse systems can highlight interesting comparisons and similarities, but does it make sense to discuss learning, or even behavior, in an animal with no brain or even neurons? 

    What do birds hear in bird song?

    Patricia Jones

    A female (left) and male (right) zebra finch,  Taeniopygia guttata

    A female (left) and male (right) zebra finch, Taeniopygia guttata

    This week's paper is a review in Animal Behaviour by Robert Dooling and Nora Prior on differences in human and bird perception of birdsong. It is broadly understood that we don't perceive the world and same way that other animals do. It's is hard for us to even wrap our heads around what it would be like to echolocate like a bat, or use electroreception like an electric fish. Even within the senses that are familiar to us, we know that we see flowers differently from bees, and dogs smell all kinds of things that we don't. So what about bird song? Apparently what birds are better at is what Dooling and Prior call "extremely fine temporal processing". In particular, this paper focuses on zebra finches which are the model organism for development and processing of bird song. In one of the studies they discuss Dooling tested "fine temporal processing", by creating artificial stimuli (called Schroeder complexes) composed of repeated harmonics that are either rising or falling. Then they created multiple stimuli in which the time interval between each rising or falling harmonic was shorter and shorter and shorter. Dooling then tested how short the intervals had to be before humans or birds could no longer distinguish the rising from the falling harmonics. 

    I really wish that they had provided some audio files of the test stimuli in the supplement, because it is hard for me to grasp how different these positive and negative Schroeder complexes sound. Regardless, Dooling found that at short intervals zebra finches were much much better at distinguishing these two stimuli than humans. This review paper discusses other experiments that also demonstrate superior distinguishing of sounds over very short time intervals in birds compared to humans. These studies indicate that birds are likely able to detect stimuli, or variation in stimuli, in bird song that we simply cannot hear.