An Insula-Central Amygdala Circuit for Guiding Tastant-Reinforced Choice Behavior.
Journal: 2018/November - Journal of Neuroscience
ISSN: 1529-2401
Abstract:
For animals to survive, they must reliably predict during foraging which substances are suitable for consumption. Despite extensive study, the neural circuit mechanisms underlying such adaptive behavior remain poorly understood. Here, using a tastant (sucrose/quinine)-reinforced "go/no-go" task in male and female mice, we examined the anatomical and functional connectivity of the circuit linking the insular cortex (IC) to the central amygdala (CeA) and the role of this circuit in the establishment of appropriate behavioral responses. Using anatomic tracing approaches combined with optogenetics-assisted circuit mapping, we found that the gustatory region of the IC sends direct excitatory projections to the lateral division of the CeA (CeL), making monosynaptic excitatory connections with distinct populations of CeL neurons. Specific inhibition of neurotransmitter release from the CeL-projecting IC neurons prevented mice from acquiring the "no-go" response, and impaired the "go" responses in the go/no-go task. Furthermore, selective activation of the IC-CeL pathway with optogenetics drove unconditioned lick suppression in thirsty animals, induced aversive responses, and was sufficient to instruct conditioned action suppression in response to a cue predicting the optogenetic activation. These results indicate that activities in the IC-CeL circuit are critical for establishing taste-reinforced behavioral responses, including avoidance responses to an aversive tastant, and are sufficient to drive learning of anticipatory avoidance. Our findings suggest that the IC-CeL circuit plays an important role in guiding appropriate choices during foraging.SIGNIFICANCE STATEMENT An animal's ability to predict which substances are suitable for consumption and then produce an appropriate action to those substances is critical for survival. Here we found that activity in the circuit that links the insular cortex (IC) to the central amygdala (CeA) is necessary for establishing appropriate behavioral responses to taste-predicting cues. This neural circuit seems to be particularly tuned to avoid an unpleasant tastant, and is also sufficient to drive learning of such avoidance responses. These results suggest that the IC-CeA circuit is critical for generating appropriate behavioral responses during foraging when facing different choices.
Relations:
Content
Citations
(5)
Drugs
(1)
Genes
(3)
Similar articles
Articles by the same authors
Discussion board
Curr Biol 28(1): 100-105e4.

Selection for Collective Aggressiveness Favors Social Susceptibility in Social Spiders

RESULTS AND DISCUSSION

In virtually any social group, some individuals are more influential than others [1014]. Knowledgeable matriarchs in elephant [15] or orca [16] societies, ringleaders in chimpanzees hunting groups [17], and movement initiators in schooling fish [18] are all examples in which one or a few individuals strongly influence the behavior and success of their group. Such phenomena have inspired many scientists to ask what causes individuals to become so influential [15, 1821]. Yet an equally important, but potentially inconspicuous and overlooked, contributor to these dynamics is the social susceptibility of other group members [1820, 2224]. We define “social susceptibility” as an individual’s tendency to alter its behavior in response to influential social partners (see glossary in Table 1). Previous studies have demonstrated links between individuals’ tendency to assume leader versus follower roles during movement decisions and individual temperament [21, 25] and that the strength of these roles can differ as a consequence of individuals’ previous social experience [5]. Here we experimentally determine the relative contributions of “social influence” versus “social susceptibility” in generating habitat-specific variation in apparent leader-follower dynamics and collective behavior. “Leader” is broadly defined here as an individual that influences, leads, or commands a group.

Table 1

Glossary Describing the Terminology Used Here to Describe Individual Behavioral Tendencies, the Collective Behavioral Tendencies of Groups, and Social Roles

DefinitionMetric and Interpretation
Boldness: an individual’s willingness to place itself at risk. Individuals that tolerate risk are termed “bold,” whereas more cautious individuals are termed “shy.”Boldness is measured in Stegodyphus by administering an aversive stimulus to spiders that causes a death feigning response. Boldness is then measured as the latency for individuals to leave their death feigning posture and resume normal activity. The majority of S. dumicola are shy, and only a small number exhibit a bolder phenotype (3%–6% of the mature spiders assayed here). This metric is continuously distributed.
Aggressiveness: hostile or forceful behavior, here measured in the context of foraging responsiveness.Aggressiveness is measured in Stegodyphus using spiders’ responsiveness toward prey. Greater responsiveness is interpreted as more aggressive foraging behavior. For individuals, this is measured as an individual’s tendency to join hunting groups. Colony-level aggressiveness is measured as the number of colony members that participate in a hunting group.
Leader: an individual that influences, leads, or commands a group.Leadership in S. dumicola is measured as an individual’s ability to alter the collective behavior of their group. In arid regions, bold individuals are able to dramatically enhance the foraging aggressiveness of their colony. Bold individuals from wet sites have no such effect. Although initiators of group movement, typically located at the front of the group, are often hypothesized to be group leaders, leaders need not always reside at the front of groups to exact their social power.
Social susceptibility: an individual’s tendency to alter its behavior in response to influential social partners.Social susceptibility is measured in S. dumicola as an individual’s tendency to respond to the movement or excitatory actions of other individuals. Shy individuals from arid sites are socially susceptible, as evidenced by their willingness to follow the vibratory cues of foraging conspecifics and by their tendency to model the foraging behavior of bold group members, even long after bold group members have been removed from the colony.

The spider Stegodyphus dumicola lives in colonies of up to a few thousand individuals [26]. Social spiders exhibit a highly female-biased primary sex ratio [27], and females perform the vast majority of colony maintenance tasks in these systems. Female social spiders cooperate in collective prey capture and food sharing, web maintenance, and alloparental care [27]. S. dumicola also exhibits temporally consistent individual differences in behavior that are associated with the roles that individuals play in groups [28, 29]. Prior lab studies on S. dumicola have shown that a colony’s boldest group member wields a large influence over the behavior and success of its colony. Boldness is defined as an individual’s willingness to place itself at risk, and bold individuals constitute 3%–10% of a mature S. dumicola colony. Bold Stegodyphus are more likely to be initiators and participators in prey capture events, in both S. dumicola [28, 30] and other social congeners [31]. In S. dumicola from arid sites in the Kalahari desert, previous research has shown that boldness is not associated with individual body condition in these spiders, although bold individuals do tend to be slightly smaller than their shy counterparts (r only 0.03) [32].

Importantly, adding just one bold individual to a laboratory colony of all shy individuals increases the number of colony members that respond to prey by >400% and increases the mass gained by other colony members by 200% [30, 33]. Such effects on colony behavior and performance persist even weeks after bold individuals are removed from groups, thus conveying that their “social influence” over groups is enduring [34]. Yet, other studies have also shown that, in lab environments with enemies, a bold individual can spread misinformation about predator identity and decrease colony mass gain and survival [35]. Given that the same kind of key individual can enhance colony success under some situations but diminish it in others, it is likely that such social dynamics will not be universal across environments. We therefore predict variation among populations in this unique social relationship.

Here we focus on differences between arid and wet environments because prior work has shown local behavioral adaptation in spiders associated with aridity [36, 37]. Specifically, arid habitats are generally prey limited and thus favor more aggressive behavior, whereas wetter habitats have more prey but more enemies too [36, 37]. Thus, we predicted that bold individuals would have a positive effect on colony aggressiveness, and therefore on colony performance, in harsher desert sites where aggressive hunting may be essential.

We established 242 experimental colonies of 19 shy spiders and one bold spider of varying degrees of boldness in the field (see STAR Methods for details). For some colonies, the single bold spider was only slightly bolder than the shy spiders, but in other colonies the bold individual was far bolder than other group members. We first determined whether the boldness of the single boldest group member predicts colonies’ foraging aggressiveness. We measured the foraging aggressiveness of each colony three times by recording the number of spiders that responded to simulated prey. We then deployed 28–32 colonies at each of eight sites along two orthogonal precipitation gradients: one extending from the Namib Desert north (810 km) and one extending from the Kalahari Desert east (981 km) (Figure S2). All of the spiders used to create experimental colonies were mature females, and only native spiders were used to construct colonies at each site. Colonies were established in the field in November–December 2016, and their status was checked in February of the following year. In particular, we noted whether each colony survived this time period and whether produced offspring. We used these data to determine (1) whether the boldness of the single boldest group member predicts colonies’ foraging aggressiveness and (2) whether colony foraging aggressiveness is linked with colony survival and reproductive success.

At the four arid sites, we found that colony aggressiveness toward prey increased up to 300% in groups containing just one very bold spider, as compared to colonies in which a less bold spider was added (F1, 111.4 = 50.17, β = 0.008 ± 0.001, p < 0.0001; Figure 1A). At the four wet sites, we did not find an effect of the boldness of colonies’ boldest spiders on colony aggression (F1, 120.5 = 0.01, β = −3.539e–6 ± 5.95e–5, p = 0.98; Figure 1B). Thus, an apparent leader-follower relationship, in which group members alter their foraging behavior in the presence of a bold individual, exists only in arid sites and not in wet ones. To further test whether the establishment of this social order is beneficial, we examined the colonies’ survival and reproduction in the field. We detected positive selection on colony aggressiveness at all arid sites: aggressive colonies were more likely to survive (likelihood ratio [L-R] χ1 = 75.44, p < 0.0001; Figure 1A) and produce offspring (L-R χ1 = 21.13, p < 0.0001), whereas no selection was detected at the wet sites (colony survival: L-R χ1 = 0.03, p = 0.85; Figure 1B; offspring production: L-R χ1 = 0.002, p = 0.95) (Data S1). Thus, bold individuals have a pronounced positive effect on colony aggressiveness, and this collective aggressiveness has a positive effect on colony survival and reproductive success only in arid sites, as predicted. Perhaps in response to this selection, colonies at arid sites are, on average, more aggressive than colonies at wet sites (F1, 10.23 = 7.22, β = 0.83 ± 0.31, p = 0.02).

An external file that holds a picture, illustration, etc.
Object name is nihms949964f1.jpg
Regional Variation in Colony Performance and Survival

Relationship between the number of spiders that attacked a simulated prey and the boldness of the boldest individual in the group (i.e., the leader) in arid (A) and wet (B) sites. Each site is depicted in a different color. Open circles indicate colonies that did not survive after 2 months in the field, and closed circles indicate colonies that persisted. See Figure S2 for site locations.

Colonies of S. dumicola from arid and wet sites also differ in a number of other respects. For instance, naturally occurring arid-dwelling colonies contain 40% more individuals than their wet-site counterparts (F1, 51 = 12.34, p = 0.0009). However, the more plentiful colony constituents in arid-site colonies are also smaller in size (prosoma width: F = 8.52, β = −0.45 ± 0.15, p = 0.03). Bold individuals are also slightly rarer at arid sites than at wet sites. Bold individuals constitute 5.4% of mature individuals in wet sites but only 3.6% of individuals at dry sites (F1, 51 = 6.13, p = 0.02). Together, the ecology and functioning of natural S. dumicola colonies in arid versus wet sites appear to differ in a number of respects.

The effect of bold S. dumicola on colony behavior and success in arid sites could be due to the “social influence” of bold individuals or the “social susceptibility” of shy individuals, or both. To disentangle these effects, we created 96 chimeric colonies in the laboratory with different compositions of individuals from arid and wet sites. The arid-wet chimeric colonies included two treatments in which the followers (shy) were from either an arid or a wet site and the leader (bold) was from the opposing habitat type. We found that bold individuals from arid sites are not uniquely influential (Figure 2B). When bold individuals from arid sites were mixed with shy individuals from wet sites, we did not recover a positive association between the boldness of the group’s boldest member and colony aggressiveness (F1, 44.24 = 1.07, β = −0.0007 ± 0.0007, p = 0.31; Figure 2B). In contrast, when we mixed shy individuals from arid sites with bold individuals from wet sites, we recovered a positive association between the boldness of singularly bold individuals and group foraging aggressiveness (F1, 44 = 1.07, β = 0.006 ± 0.0008, p < 0.0001; Figure 2C). Thus, habitat-specific differences in leader-follower dynamics appear to be generated by site-specific variability in the “social susceptibility” of shy individuals and not the “social influence” of bold individuals.

An external file that holds a picture, illustration, etc.
Object name is nihms949964f2.jpg
Chimeric Colonies Reveal “Social Susceptibility” of Shy Individuals from Arid Sites

Relationship between the number of spiders that attacked a simulated prey and the boldness of the boldest individual in the group (i.e., the leader) in chimeric colonies. Color denotes the origin of the shy individuals (followers) in the chimeric colonies: orange, arid; blue, wet. Shapes denote the origin of the bold individuals (leaders): circles, arid; triangles, wet; squares, heterospecifics (S. mimosarum). (A) and (D) are controls in which both the leaders and followers come from arid (A) or wet (D) sites. Shy followers from wet sites are not susceptible to bold leaders from arid sites (B) or from another species (F). However, shy followers from arid sites are susceptible to the influence of bold leaders from wet sites (C) and to bold leaders from a different species (E). See Figure S1 for more information on following behavior.

To determine whether mixing individuals across populations that share a similar climate affects colony foraging behavior, we generated two control treatments in which the leader and followers were from two different sites that were both either arid (ncolonies = 24) or wet (ncolonies = 15) (see STAR Methods). These control chimeric colonies revealed that mixing individuals from populations that share the same climate had no effect on colony foraging behavior (Figures 2A and and2D).2D). The leader-follower dynamics observed at arid sites do not break down in response to mixing, despite reduced familiarity or relatedness (F1, 20 = 26.74, β = 0.006 ± 0.001, p < 0.0001; Figure 2A), and the wet-wet chimeras never produced the behavioral patterns observed in arid populations (F1, 11 = 2.96, β = −0.0062 ± 0.0016, p = 0.11; Figure 2D).

To determine whether the shy spiders from arid sites are susceptible to any indication of boldness in their group, we produced interspecific-chimeric colonies. Here we constructed colonies of 19 shy S. dumicola from two arid sites (Kalkrand, Namibia and Upington South Africa; total n = 48) and two wet sites (Rundu, Namibia and Ladysmith, South Africa; total n = 37) with a single heterospecific congener Stegodyphus mimosarum. Like S. dumicola, S. mimosarum exhibits temporally stable individual differences in boldness that are linked with individual social roles [38]. This species has independently evolved sociality, and it last shared a common ancestor with S. dumicola several million years ago [39]. Despite these differences, adding a bold S. mimosarum to a colony of shy S. dumicola from arid sites enhanced the collective foraging aggressiveness of chimeric colonies to the same degree as adding a bold conspecific (F1, 44 = 35.78, β = 0.006 ± 0.001, p < 0.0001; Figure 2E). In contrast, adding a bold S. mimosarum to a colony of shy S. dumicola from wet sites did not enhance colony aggressiveness (F1, 44 = 2.07, β = 0.002 ± 0.0013, p = 0.16; Figure 2F). Thus, shy individuals from arid sites are obligatorily socially susceptible, as evidenced by their willingness to respond to bold conspecifics from any source, as well as heterospecifics, to the same degree that they respond to native bold individuals.

One possible mechanism behind the leader-follower dynamics observed at arid sites is within-colony competition for resources. If colonies at arid sites are more food starved, it may benefit shy colony members to quickly join bold individuals during prey attack sequences, as to secure limited resources, particularly if bold individuals are disproportionately able to monopolize resources for themselves. One might also expect within-colony competition over resources to be less pronounced at high resource sites. However, the majority of the available data are mixed or at odds with this hypothesis. Colonies at arid sites were of no lower body condition than their counterparts at wet sites (F1, 5.89 = 3.21, β = −0.002 ± 0.001, p = 0.12). Furthermore, body condition is unrelated to boldness at both wet and dry sites (boldness: F1, 236.8 = 0.28, β = −3.772e−6 ± 7.172e−6, p = 0.60, boldness × site type: F1, 236.8 = 0.28, β = −2.237e−6 ± 7.172e−6, p = 0.76) (Data S1). Thus, bold individuals are not consistently able to secure more resources than their shy counterparts, nor are they more food deprived, and colonies at arid sites are not in poorer body condition. Instead, shy colony members at arid sites actually benefit by the presence of bold individuals because they enhance colony survival (Figure 1), reproductive output (Data S1), and the collective mass gain of shy colony members [30, 33]. Finally, the social dynamics observed at arid sites are seemingly unresponsive to manipulations to resource availability (see additional results in Figure S1) [30]. In sum, this suggests that ongoing differences in hunger among individuals within colonies or across sites are not sufficient to explain our results. Instead, we propose that an early developmental or evolutionary history of food limitation may be key.

Socially influential individuals can provide their groups with benefits [15, 30] but also situationally associated costs [35, 40]. In the social spider S. dumicola, a leader-follower dynamic is established in a habitat-specific manner and is restricted to arid environments. This social order also enhances colony survival and reproduction at arid sites and is therefore subject to ongoing site-specific selection. Most interestingly, we discover that this leader-follower relationship depends on habitat-specific “social susceptibility” of shy spiders from arid sites, because these shy individuals are uniquely influenced by the social cues of conspecifics (Figure S1). Notably, post hoc behavioral analyses showed that even bold individuals from arid sites exhibit some degree of heightened social responsiveness, at least as compared to their counterparts from wet sites (Figure S1), though not nearly so much as shy individuals. Therefore, the emergence of leadership is not driven by the unique traits of leaders, but by the population majority’s willingness to follow. In the hands of the socially susceptible, anything approximating the phenotype of a leader, even a heterospecific sham, is sufficient to elicit a response. Though naive followership appears to be advantageous in particular settings (e.g., in arid habitats for S. dumicola), such susceptibility may one day provide an easy point of manipulation whereby enemies or selfish phenotypes could evolve to exploit entire societies.

STAR★ METHODS

Detailed methods are provided in the online version of this paper and include the following:

KEY RESOURCES TABLE

REAGENT or RESOURCESOURCEIDENTIFIER
Biological Samples
Healthy mature Stegodyphus dumicola &amp; Stegodyphus mimosarumSouthern AfricaN/A
Deposited Data
Raw &amp; Analyzed Behavioral &amp; Colony Survival DataDryadhttps://doi.org/10.5061/dryad.7p090

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources should be directed to and will be fulfilled by the Lead Contact, Jonathan N. Pruitt (ude.bscu@ttiurp).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Colony Selection Study

Study Sites and Collection

Whole S. dumicola colonies (n = 212) were collected along roadside Acacia mellifera and barbwire fencing at eight study sites. These sites were arranged along two orthogonal precipitation gradients: one extending from the Namib Desert north toward Angola (810km) and a second extending from the Kalahari Desert east toward Lesotho (981km). Along each gradient, care was taken to select two arid desert sites (Namib Gradient: Kalkrand [−24.065027, 17.580452] and Rehoboth [−23.209881, 17.092]; Kalahari Gradient: Upington [−28.403361, 21.071249] and Boegoeberg [−29.037819, 22.027999]) and two wetter savannah sites (Namib Gradient: Outjo [−20.233099, 16.354468] and Rundu [−18.299209, 19.407636]; Kalahari Gradient: Ladysmith [−38.65655, 29.625249] and Weenen [−28.856239, 30.142306]) per gradient (Figure S2). A total of 211 source colonies were collected ranging in size from 75–512 mature female S. dumicola. The sequence of colony creation and deployment was arranged to prevent linearly conflating aridity with time of deployment (Nov 1–Dec 15, 2016): Wet (Rundu) → Wet (Outjo) → Arid (Rehoboth) → Arid (Kalkrand) → Arid (Upington) → Arid (Boegoeberg) → Wet (Ladysmith) → Wet (Weenen)

Colonies were collected by covering the nest containing all the spiders in a cloth pillowcase and snipping off the supporting branches using pruning snips. Whole colonies were then transported to nearby hotels and lodges. Each source colony was then dissected by hand by gently pulling open the nest and counting all of the spiders therein. Spiders were isolated in 59 mL deli containers and their prosoma widths, masses, and individual boldness scores were obtained before assigning them to an experimental colony. When assigning spiders to an experimental colony, care was taken never to mix spiders from multiple source colonies in order to preserve natural levels of within-colony relatedness and familiarity, both of which can impact collective behavior and foraging performance in this or other species of Stegodyphus [4144].

Experimental colonies used in our selection studies were subsequently deployed at the same source site where the original colonies were collected, i.e., experimental colonies were never moved across sites. We avoided moving spiders across sites to preserve natural population genetic structure, which is the subject of ongoing genomic investigations by other investigators.

METHOD DETAILS

Boldness Assay

Boldness is defined as an individual’s willingness to place itself at risk [45]. In S. dumicola, boldness is assessed by measuring the duration of an individual’s antipredator response following an aversive stimulus [31, 46]. Boldness trials are initiated by placing each individual spider in the center of a clean open field container (radius = 12cm) and giving the spider 30 s to acclimate. Two puffs of air are then administered toward the dorsal anterior end of the spider using an infant nose-cleaning bulb. These puffs universally result in S. dumicola pulling in its legs in a death feign, which varies in duration. Spiders that remain in the death feign for longer are deemed shyer than spiders that emerge from a death feign sooner. We recorded individuals’ latency to emerge from a death feign and move one full body length in the open field. Individuals are given a maximum duration of 10 min to emerge and move one body length. Individuals that fail to move a body length in this time receive a maximum score of 600 s. We then take individuals’ latency to resume movement and subtract it from the maximum time provided in the trial (600 s) to obtain a boldness index that scales positively with boldness. All behavioral assays were conducted during daylight hours under artificial lighting in the absence of air drafts.

Individual differences in this test are repeatable over the duration of several months [47, 48]. Moreover, individual differences in boldness are tightly linked with these spiders’ tendency to participate in a variety of cooperative tasks, including colony defense [29], prey capture [28], and web construction and maintenance [49]. The majority (90%–95%) of adult S. dumicola exhibit a shy behavioral type (latency to emerge and resume movement > 600 s) [47]. Bold individuals, while rare, are known to have a large impact on colony behavior and performance in the laboratory [30, 35].

The truncated dataset associated with a 10-min termination to our trials raises the concern that shy individuals from arid sites might be shyer than their counterparts from wet sites (or vice versa), and that this variation could be concealed by the design of our assay. However, a more open-ended 3 hr trial where we tracked the emergence time of shy individuals from wet (Ladysmith, Weenan) versus arid (Upington, Groblershoop [near Boegoeberg]) sites reveals (1) that there are no concealed differences in the shyness of individuals that receive as boldness score of 0 (latency to emerge > 600 s) in our truncated trials, and that many of these individuals would take hours to emerge from their huddles (see supplementary statistics). Thus, conducting these trials over a duration which would not truncate the dataset for the ~ 10,000 spiders assayed here was not feasible.

Experimental Colony Construction

We created experimental colonies of 20 adult female spiders. 19 of these females exhibited a latency to resume movement of 600 s or greater (boldness index = 0), and the 20 spider was randomly pulled from a pool of spiders that varied in their boldness scores from 1–600. This created a set of experimental colonies of identical size and developmental stage, but which varied in the boldness score of the single boldest member of the group. Prior studies have referred to these individuals as ‘keystone individuals’ because they exhibit a disproportionately large influence over the collective behavior and success of their colony.

Experimental colonies were housed in 390 mL plastic cups containing three Acacia mellifera twigs to facilitate web construction. Colonies were provided two days to construct a rudimentary nest and capture web. Colonies were then run through three collective foraging aggressiveness assays (1/day for 3 days) before being deployed into the field. The temperature in the lodge/hotel rooms where individual boldness assays, colony assembly, and colony foraging aggressiveness assays occurred varied from 22.2–25.5C. All behavioral assays were conducted during daylight hours under artificial lighting in the absence of air drafts.

Colony Foraging Aggressiveness Assay

Colony foraging aggressiveness is quantified as the number of colony members that leave the nest during a staged prey capture event, where more aggressive colonies deploy more attackers [30]. Foraging aggressiveness assays are initiated by placing a 1 × 1cm piece of paper in the colony capture web ~5cm away from the nest entrance, and then vibrating the piece of paper using a 1.5mm diameter wire attached to the end of a handheld vibratory device (Magic Purple Bullet). The device was set on a pulse setting, as to mimic the irregular thrashing movements of insect prey captured in the web. We then recorded the number of spiders that left the nest to capture the simulated prey item, until the first individual contacted the piece of paper. Colony prey capture behavior was assessed once per day for three consecutive days. At the end of the third day, colonies were deployed at one of our eight field sites during the evening hours. Nearly all individuals that emerge in response to stimulated prey eventually attempted to attack and subdue the stimulus, which conveys that emergence is an indication of individuals’ desire to attack prey, and not merely to acquire information about it.

This assessment of colony aggressiveness does raise the potential concern that if one individual is very fast and contacts the paper rapidly, then its fellow colony members may not have time to emerge. However, at odds with this inference we actually find that fast-responding Stegodyphus colonies also attack prey with larger numbers of individuals in a smaller amount of time [30, 46, 50]. Additionally, the number of attackers that emerge in response to prey until contact is ecologically realistic because, once the first few individuals contact and subdue the prey, the attack ceases and the prey is dragged back to the next entrance where it is shared. Thus, in nature, the time for the attack sequence to occur is actually smaller for aggressiveness colonies. All colony aggressiveness assays were conducted during daylight hours under artificial lighting in the absence of air drafts in lodges and hotels.

Colony Deployment and Monitoring

Experimental colonies (28–33 colonies per sites) were deployed at our eight sites by using clothespins to attach the colony in its container to resident Acacia trees. We used the preexisting presence of S. dumicola colonies on nearby trees as an indicator of habitat quality and substrate suitability. Care was taken not to place colonies on Acacia trees actively being guarded or patrolled by ants (genus Anoplolepis and Crematogaster), which can destroy S. dumicola colonies before they establish [51, 52]. Prior studies have shown that setting colonies out in the evening hours provides them with an opportunity to produce a functional nest and a sizable capture web before the following sunrise [46].

We then left these experimental colonies in the field for the next several months and checked their performance in February 2017. During these checks, we determined if there were any spiders living within the colonies, noted the presence and number of egg cases, noted the presence of spiderlings, and, in cases where the colony had perished, we noted situational factors that provide clues as to why the colony may have perished. For instance, a large number of predatory ant carcasses remaining in the colony retreat and in the plastic deli cup was taken as evidence of a raid by predatory ants, which kill a large number of natural colonies each year. The presence of sporulated fungus emerging from the carcasses of colony members was interpreted as evidence that fouling by fungus had contributed to colony demise. Entomopathogenic fungus is a common driver of colony demise in this system [52]. In this study we focus on whether or not colonies survived the duration of study, meaning that at least one colony member was left alive, and whether or not colonies successfully reproduced, i.e., produced viable egg cases or spiderlings during this time period.

The phenology of our experimental colonies (e.g., timing of egg case production, presence of spiderlings, age of spiderlings) closely matched those of native naturally-occurring colonies. Experimental colonies were composed of mature females only that came from colonies containing mature males, and thus, were likely mated prior to colony establishment. Although, immigration of some foreign males to the colonies was evident during our February check.

Chimeric Colony Study

Because the influence of bold spiders on colony foraging behavior was only detected at arid sites, we used chimeric colonies (n = 96) to determine whether this dynamic was driven by the “social influence” of bold individuals from arid sites or the “social susceptibility” of shy individuals from arid sites. We created colonies of bold individuals from arid sites with shy individuals from wet sites (N = 48), and vice versa (N = 48). We also executed a set of control chimeras (n = 39) where we mixed individuals from different sites with similar climatic conditions (arid with arid [n = 24], wet with wet [n = 15]). This allowed us to determine if merely mixing colonies alters associations between group composition and group behavior, and whether the mechanisms governing the influence of bold individuals at arid sites are mutually cross-compatible. Mixtures involve 19 shy individuals from one population with 1 bold individual from another, and vice versa (Wet: Ladysmith → Rundu, Rundu → Ladysmith, Arid: Kalkrand → Upington, Upington → Kalkrand).

Chimeric colony studies were carried out under standardized laboratory conditions at the University of California Santa Barbara. Twenty intact S. dumicola colonies were collected form each site and transported to laboratory in January 2017. Colonies were dissected out and the prosoma width, mass, and boldness of each colony constituent were determined over a 72 hr period. We then constructed chimeric colonies (n = 96) of the same behavioral composition (19 spiders with a boldness index of 0, and 1 spider with a boldness index between 1 and 600). Colonies were then subjected to foraging aggressiveness assays identical to those detailed above.

To determine if the shy spiders from arid sites are socially susceptible to any indication of boldness in their group, we produced a third set of interspecific-chimeric colonies. Here we constructed colonies of 19 shy S. dumicola from two arid sites (Kalkrand, Namibia and Upington South Africa, total n = 48) and two wet sites (Rundu, Namibia and Ladysmith, South Africa, total n = 37) with a single heterospecific congener Stegodyphus mimosarum. Like S. dumicola, S. mimosarum exhibits temporally stable individual differences in boldness that are linked with individual social roles [38]. This species has independently evolved sociality and shared a common ancestor with S. dumicola several million years ago [39].

Web Recruitment Study

We designed a final study to determine 1) whether shy individuals from arid sites are uniquely drawn to the cues of conspecific on the capture web, and 2) whether food restriction could cause shy individuals from wet sites to become more socially susceptible (see Figure S1). To do this, we created colonies of 19 shy spiders or 19 bold spiders from two arid (Kalkrand, Namibia and Upington South Africa) and two wet (Rundu, Namibia and Ladysmith, South Africa) sites (total N = 33). We then subjected these colonies to two different sensory stimuli at three time points (in a fully factorial design), where each colony was tested in each of the six different treatment combinations. To test responsiveness to conspecifics we introduced a bold conspecific taken from the same source colony to the experimental colony’s capture web. To test change over time in responsiveness to prey, we introduced a domestic cricket on the capture web. Colonies were presented with each of these stimuli in a randomized order one, two, and five days after an ad libitum feeding event and we recorded the number of spiders that responded to each of these stimuli. The ordering of prey-response treatments (1, 2 or 5 days post-feeding) was likewise alternated among focal colonies.

Timeline

  • Colony Establishment in the Field: November-December 2016

  • Colony Performance Assessment: February 2017, July 2017

  • Chimeric Colony Experiment (Lab): February-April 2017

  • Web Recruitment Studies: April-May 2017

QUANTIFICATION AND STATISTICAL ANALYSIS

Our statistical analyses were performed in JMP 12.0. In brief, we constructed a variety of GLMs and LMMs to evaluate associations between group composition and various behavioral and colony performance response variables (see Data S1 for full model descriptions). Statistical significance was evaluated using an α = 0.05. Sample sizes were determined by the availability of spiders of specific behavioral tendencies and general feasibility based on personnel limitations. No replicates were excluded from our statistical analyses. The distribution of residuals and the nature of our response variables (e.g., binary versus continuous) were used to determine our model structures.

DATA AND SOFTWARE AVAILABILITY

Behavioral and colony performance data summarized in this paper are available in the Dryad Digital Repository: https://doi.org/10. 5061/dryad.7p090.

KEY RESOURCES TABLE

REAGENT or RESOURCESOURCEIDENTIFIER
Biological Samples
Healthy mature Stegodyphus dumicola &amp; Stegodyphus mimosarumSouthern AfricaN/A
Deposited Data
Raw &amp; Analyzed Behavioral &amp; Colony Survival DataDryadhttps://doi.org/10.5061/dryad.7p090

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources should be directed to and will be fulfilled by the Lead Contact, Jonathan N. Pruitt (ude.bscu@ttiurp).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Colony Selection Study

Study Sites and Collection

Whole S. dumicola colonies (n = 212) were collected along roadside Acacia mellifera and barbwire fencing at eight study sites. These sites were arranged along two orthogonal precipitation gradients: one extending from the Namib Desert north toward Angola (810km) and a second extending from the Kalahari Desert east toward Lesotho (981km). Along each gradient, care was taken to select two arid desert sites (Namib Gradient: Kalkrand [−24.065027, 17.580452] and Rehoboth [−23.209881, 17.092]; Kalahari Gradient: Upington [−28.403361, 21.071249] and Boegoeberg [−29.037819, 22.027999]) and two wetter savannah sites (Namib Gradient: Outjo [−20.233099, 16.354468] and Rundu [−18.299209, 19.407636]; Kalahari Gradient: Ladysmith [−38.65655, 29.625249] and Weenen [−28.856239, 30.142306]) per gradient (Figure S2). A total of 211 source colonies were collected ranging in size from 75–512 mature female S. dumicola. The sequence of colony creation and deployment was arranged to prevent linearly conflating aridity with time of deployment (Nov 1–Dec 15, 2016): Wet (Rundu) → Wet (Outjo) → Arid (Rehoboth) → Arid (Kalkrand) → Arid (Upington) → Arid (Boegoeberg) → Wet (Ladysmith) → Wet (Weenen)

Colonies were collected by covering the nest containing all the spiders in a cloth pillowcase and snipping off the supporting branches using pruning snips. Whole colonies were then transported to nearby hotels and lodges. Each source colony was then dissected by hand by gently pulling open the nest and counting all of the spiders therein. Spiders were isolated in 59 mL deli containers and their prosoma widths, masses, and individual boldness scores were obtained before assigning them to an experimental colony. When assigning spiders to an experimental colony, care was taken never to mix spiders from multiple source colonies in order to preserve natural levels of within-colony relatedness and familiarity, both of which can impact collective behavior and foraging performance in this or other species of Stegodyphus [4144].

Experimental colonies used in our selection studies were subsequently deployed at the same source site where the original colonies were collected, i.e., experimental colonies were never moved across sites. We avoided moving spiders across sites to preserve natural population genetic structure, which is the subject of ongoing genomic investigations by other investigators.

Colony Selection Study

Study Sites and Collection

Whole S. dumicola colonies (n = 212) were collected along roadside Acacia mellifera and barbwire fencing at eight study sites. These sites were arranged along two orthogonal precipitation gradients: one extending from the Namib Desert north toward Angola (810km) and a second extending from the Kalahari Desert east toward Lesotho (981km). Along each gradient, care was taken to select two arid desert sites (Namib Gradient: Kalkrand [−24.065027, 17.580452] and Rehoboth [−23.209881, 17.092]; Kalahari Gradient: Upington [−28.403361, 21.071249] and Boegoeberg [−29.037819, 22.027999]) and two wetter savannah sites (Namib Gradient: Outjo [−20.233099, 16.354468] and Rundu [−18.299209, 19.407636]; Kalahari Gradient: Ladysmith [−38.65655, 29.625249] and Weenen [−28.856239, 30.142306]) per gradient (Figure S2). A total of 211 source colonies were collected ranging in size from 75–512 mature female S. dumicola. The sequence of colony creation and deployment was arranged to prevent linearly conflating aridity with time of deployment (Nov 1–Dec 15, 2016): Wet (Rundu) → Wet (Outjo) → Arid (Rehoboth) → Arid (Kalkrand) → Arid (Upington) → Arid (Boegoeberg) → Wet (Ladysmith) → Wet (Weenen)

Colonies were collected by covering the nest containing all the spiders in a cloth pillowcase and snipping off the supporting branches using pruning snips. Whole colonies were then transported to nearby hotels and lodges. Each source colony was then dissected by hand by gently pulling open the nest and counting all of the spiders therein. Spiders were isolated in 59 mL deli containers and their prosoma widths, masses, and individual boldness scores were obtained before assigning them to an experimental colony. When assigning spiders to an experimental colony, care was taken never to mix spiders from multiple source colonies in order to preserve natural levels of within-colony relatedness and familiarity, both of which can impact collective behavior and foraging performance in this or other species of Stegodyphus [4144].

Experimental colonies used in our selection studies were subsequently deployed at the same source site where the original colonies were collected, i.e., experimental colonies were never moved across sites. We avoided moving spiders across sites to preserve natural population genetic structure, which is the subject of ongoing genomic investigations by other investigators.

Study Sites and Collection

Whole S. dumicola colonies (n = 212) were collected along roadside Acacia mellifera and barbwire fencing at eight study sites. These sites were arranged along two orthogonal precipitation gradients: one extending from the Namib Desert north toward Angola (810km) and a second extending from the Kalahari Desert east toward Lesotho (981km). Along each gradient, care was taken to select two arid desert sites (Namib Gradient: Kalkrand [−24.065027, 17.580452] and Rehoboth [−23.209881, 17.092]; Kalahari Gradient: Upington [−28.403361, 21.071249] and Boegoeberg [−29.037819, 22.027999]) and two wetter savannah sites (Namib Gradient: Outjo [−20.233099, 16.354468] and Rundu [−18.299209, 19.407636]; Kalahari Gradient: Ladysmith [−38.65655, 29.625249] and Weenen [−28.856239, 30.142306]) per gradient (Figure S2). A total of 211 source colonies were collected ranging in size from 75–512 mature female S. dumicola. The sequence of colony creation and deployment was arranged to prevent linearly conflating aridity with time of deployment (Nov 1–Dec 15, 2016): Wet (Rundu) → Wet (Outjo) → Arid (Rehoboth) → Arid (Kalkrand) → Arid (Upington) → Arid (Boegoeberg) → Wet (Ladysmith) → Wet (Weenen)

Colonies were collected by covering the nest containing all the spiders in a cloth pillowcase and snipping off the supporting branches using pruning snips. Whole colonies were then transported to nearby hotels and lodges. Each source colony was then dissected by hand by gently pulling open the nest and counting all of the spiders therein. Spiders were isolated in 59 mL deli containers and their prosoma widths, masses, and individual boldness scores were obtained before assigning them to an experimental colony. When assigning spiders to an experimental colony, care was taken never to mix spiders from multiple source colonies in order to preserve natural levels of within-colony relatedness and familiarity, both of which can impact collective behavior and foraging performance in this or other species of Stegodyphus [4144].

Experimental colonies used in our selection studies were subsequently deployed at the same source site where the original colonies were collected, i.e., experimental colonies were never moved across sites. We avoided moving spiders across sites to preserve natural population genetic structure, which is the subject of ongoing genomic investigations by other investigators.

METHOD DETAILS

Boldness Assay

Boldness is defined as an individual’s willingness to place itself at risk [45]. In S. dumicola, boldness is assessed by measuring the duration of an individual’s antipredator response following an aversive stimulus [31, 46]. Boldness trials are initiated by placing each individual spider in the center of a clean open field container (radius = 12cm) and giving the spider 30 s to acclimate. Two puffs of air are then administered toward the dorsal anterior end of the spider using an infant nose-cleaning bulb. These puffs universally result in S. dumicola pulling in its legs in a death feign, which varies in duration. Spiders that remain in the death feign for longer are deemed shyer than spiders that emerge from a death feign sooner. We recorded individuals’ latency to emerge from a death feign and move one full body length in the open field. Individuals are given a maximum duration of 10 min to emerge and move one body length. Individuals that fail to move a body length in this time receive a maximum score of 600 s. We then take individuals’ latency to resume movement and subtract it from the maximum time provided in the trial (600 s) to obtain a boldness index that scales positively with boldness. All behavioral assays were conducted during daylight hours under artificial lighting in the absence of air drafts.

Individual differences in this test are repeatable over the duration of several months [47, 48]. Moreover, individual differences in boldness are tightly linked with these spiders’ tendency to participate in a variety of cooperative tasks, including colony defense [29], prey capture [28], and web construction and maintenance [49]. The majority (90%–95%) of adult S. dumicola exhibit a shy behavioral type (latency to emerge and resume movement > 600 s) [47]. Bold individuals, while rare, are known to have a large impact on colony behavior and performance in the laboratory [30, 35].

The truncated dataset associated with a 10-min termination to our trials raises the concern that shy individuals from arid sites might be shyer than their counterparts from wet sites (or vice versa), and that this variation could be concealed by the design of our assay. However, a more open-ended 3 hr trial where we tracked the emergence time of shy individuals from wet (Ladysmith, Weenan) versus arid (Upington, Groblershoop [near Boegoeberg]) sites reveals (1) that there are no concealed differences in the shyness of individuals that receive as boldness score of 0 (latency to emerge > 600 s) in our truncated trials, and that many of these individuals would take hours to emerge from their huddles (see supplementary statistics). Thus, conducting these trials over a duration which would not truncate the dataset for the ~ 10,000 spiders assayed here was not feasible.

Experimental Colony Construction

We created experimental colonies of 20 adult female spiders. 19 of these females exhibited a latency to resume movement of 600 s or greater (boldness index = 0), and the 20 spider was randomly pulled from a pool of spiders that varied in their boldness scores from 1–600. This created a set of experimental colonies of identical size and developmental stage, but which varied in the boldness score of the single boldest member of the group. Prior studies have referred to these individuals as ‘keystone individuals’ because they exhibit a disproportionately large influence over the collective behavior and success of their colony.

Experimental colonies were housed in 390 mL plastic cups containing three Acacia mellifera twigs to facilitate web construction. Colonies were provided two days to construct a rudimentary nest and capture web. Colonies were then run through three collective foraging aggressiveness assays (1/day for 3 days) before being deployed into the field. The temperature in the lodge/hotel rooms where individual boldness assays, colony assembly, and colony foraging aggressiveness assays occurred varied from 22.2–25.5C. All behavioral assays were conducted during daylight hours under artificial lighting in the absence of air drafts.

Colony Foraging Aggressiveness Assay

Colony foraging aggressiveness is quantified as the number of colony members that leave the nest during a staged prey capture event, where more aggressive colonies deploy more attackers [30]. Foraging aggressiveness assays are initiated by placing a 1 × 1cm piece of paper in the colony capture web ~5cm away from the nest entrance, and then vibrating the piece of paper using a 1.5mm diameter wire attached to the end of a handheld vibratory device (Magic Purple Bullet). The device was set on a pulse setting, as to mimic the irregular thrashing movements of insect prey captured in the web. We then recorded the number of spiders that left the nest to capture the simulated prey item, until the first individual contacted the piece of paper. Colony prey capture behavior was assessed once per day for three consecutive days. At the end of the third day, colonies were deployed at one of our eight field sites during the evening hours. Nearly all individuals that emerge in response to stimulated prey eventually attempted to attack and subdue the stimulus, which conveys that emergence is an indication of individuals’ desire to attack prey, and not merely to acquire information about it.

This assessment of colony aggressiveness does raise the potential concern that if one individual is very fast and contacts the paper rapidly, then its fellow colony members may not have time to emerge. However, at odds with this inference we actually find that fast-responding Stegodyphus colonies also attack prey with larger numbers of individuals in a smaller amount of time [30, 46, 50]. Additionally, the number of attackers that emerge in response to prey until contact is ecologically realistic because, once the first few individuals contact and subdue the prey, the attack ceases and the prey is dragged back to the next entrance where it is shared. Thus, in nature, the time for the attack sequence to occur is actually smaller for aggressiveness colonies. All colony aggressiveness assays were conducted during daylight hours under artificial lighting in the absence of air drafts in lodges and hotels.

Colony Deployment and Monitoring

Experimental colonies (28–33 colonies per sites) were deployed at our eight sites by using clothespins to attach the colony in its container to resident Acacia trees. We used the preexisting presence of S. dumicola colonies on nearby trees as an indicator of habitat quality and substrate suitability. Care was taken not to place colonies on Acacia trees actively being guarded or patrolled by ants (genus Anoplolepis and Crematogaster), which can destroy S. dumicola colonies before they establish [51, 52]. Prior studies have shown that setting colonies out in the evening hours provides them with an opportunity to produce a functional nest and a sizable capture web before the following sunrise [46].

We then left these experimental colonies in the field for the next several months and checked their performance in February 2017. During these checks, we determined if there were any spiders living within the colonies, noted the presence and number of egg cases, noted the presence of spiderlings, and, in cases where the colony had perished, we noted situational factors that provide clues as to why the colony may have perished. For instance, a large number of predatory ant carcasses remaining in the colony retreat and in the plastic deli cup was taken as evidence of a raid by predatory ants, which kill a large number of natural colonies each year. The presence of sporulated fungus emerging from the carcasses of colony members was interpreted as evidence that fouling by fungus had contributed to colony demise. Entomopathogenic fungus is a common driver of colony demise in this system [52]. In this study we focus on whether or not colonies survived the duration of study, meaning that at least one colony member was left alive, and whether or not colonies successfully reproduced, i.e., produced viable egg cases or spiderlings during this time period.

The phenology of our experimental colonies (e.g., timing of egg case production, presence of spiderlings, age of spiderlings) closely matched those of native naturally-occurring colonies. Experimental colonies were composed of mature females only that came from colonies containing mature males, and thus, were likely mated prior to colony establishment. Although, immigration of some foreign males to the colonies was evident during our February check.

Chimeric Colony Study

Because the influence of bold spiders on colony foraging behavior was only detected at arid sites, we used chimeric colonies (n = 96) to determine whether this dynamic was driven by the “social influence” of bold individuals from arid sites or the “social susceptibility” of shy individuals from arid sites. We created colonies of bold individuals from arid sites with shy individuals from wet sites (N = 48), and vice versa (N = 48). We also executed a set of control chimeras (n = 39) where we mixed individuals from different sites with similar climatic conditions (arid with arid [n = 24], wet with wet [n = 15]). This allowed us to determine if merely mixing colonies alters associations between group composition and group behavior, and whether the mechanisms governing the influence of bold individuals at arid sites are mutually cross-compatible. Mixtures involve 19 shy individuals from one population with 1 bold individual from another, and vice versa (Wet: Ladysmith → Rundu, Rundu → Ladysmith, Arid: Kalkrand → Upington, Upington → Kalkrand).

Chimeric colony studies were carried out under standardized laboratory conditions at the University of California Santa Barbara. Twenty intact S. dumicola colonies were collected form each site and transported to laboratory in January 2017. Colonies were dissected out and the prosoma width, mass, and boldness of each colony constituent were determined over a 72 hr period. We then constructed chimeric colonies (n = 96) of the same behavioral composition (19 spiders with a boldness index of 0, and 1 spider with a boldness index between 1 and 600). Colonies were then subjected to foraging aggressiveness assays identical to those detailed above.

To determine if the shy spiders from arid sites are socially susceptible to any indication of boldness in their group, we produced a third set of interspecific-chimeric colonies. Here we constructed colonies of 19 shy S. dumicola from two arid sites (Kalkrand, Namibia and Upington South Africa, total n = 48) and two wet sites (Rundu, Namibia and Ladysmith, South Africa, total n = 37) with a single heterospecific congener Stegodyphus mimosarum. Like S. dumicola, S. mimosarum exhibits temporally stable individual differences in boldness that are linked with individual social roles [38]. This species has independently evolved sociality and shared a common ancestor with S. dumicola several million years ago [39].

Web Recruitment Study

We designed a final study to determine 1) whether shy individuals from arid sites are uniquely drawn to the cues of conspecific on the capture web, and 2) whether food restriction could cause shy individuals from wet sites to become more socially susceptible (see Figure S1). To do this, we created colonies of 19 shy spiders or 19 bold spiders from two arid (Kalkrand, Namibia and Upington South Africa) and two wet (Rundu, Namibia and Ladysmith, South Africa) sites (total N = 33). We then subjected these colonies to two different sensory stimuli at three time points (in a fully factorial design), where each colony was tested in each of the six different treatment combinations. To test responsiveness to conspecifics we introduced a bold conspecific taken from the same source colony to the experimental colony’s capture web. To test change over time in responsiveness to prey, we introduced a domestic cricket on the capture web. Colonies were presented with each of these stimuli in a randomized order one, two, and five days after an ad libitum feeding event and we recorded the number of spiders that responded to each of these stimuli. The ordering of prey-response treatments (1, 2 or 5 days post-feeding) was likewise alternated among focal colonies.

Timeline

  • Colony Establishment in the Field: November-December 2016

  • Colony Performance Assessment: February 2017, July 2017

  • Chimeric Colony Experiment (Lab): February-April 2017

  • Web Recruitment Studies: April-May 2017

Boldness Assay

Boldness is defined as an individual’s willingness to place itself at risk [45]. In S. dumicola, boldness is assessed by measuring the duration of an individual’s antipredator response following an aversive stimulus [31, 46]. Boldness trials are initiated by placing each individual spider in the center of a clean open field container (radius = 12cm) and giving the spider 30 s to acclimate. Two puffs of air are then administered toward the dorsal anterior end of the spider using an infant nose-cleaning bulb. These puffs universally result in S. dumicola pulling in its legs in a death feign, which varies in duration. Spiders that remain in the death feign for longer are deemed shyer than spiders that emerge from a death feign sooner. We recorded individuals’ latency to emerge from a death feign and move one full body length in the open field. Individuals are given a maximum duration of 10 min to emerge and move one body length. Individuals that fail to move a body length in this time receive a maximum score of 600 s. We then take individuals’ latency to resume movement and subtract it from the maximum time provided in the trial (600 s) to obtain a boldness index that scales positively with boldness. All behavioral assays were conducted during daylight hours under artificial lighting in the absence of air drafts.

Individual differences in this test are repeatable over the duration of several months [47, 48]. Moreover, individual differences in boldness are tightly linked with these spiders’ tendency to participate in a variety of cooperative tasks, including colony defense [29], prey capture [28], and web construction and maintenance [49]. The majority (90%–95%) of adult S. dumicola exhibit a shy behavioral type (latency to emerge and resume movement > 600 s) [47]. Bold individuals, while rare, are known to have a large impact on colony behavior and performance in the laboratory [30, 35].

The truncated dataset associated with a 10-min termination to our trials raises the concern that shy individuals from arid sites might be shyer than their counterparts from wet sites (or vice versa), and that this variation could be concealed by the design of our assay. However, a more open-ended 3 hr trial where we tracked the emergence time of shy individuals from wet (Ladysmith, Weenan) versus arid (Upington, Groblershoop [near Boegoeberg]) sites reveals (1) that there are no concealed differences in the shyness of individuals that receive as boldness score of 0 (latency to emerge > 600 s) in our truncated trials, and that many of these individuals would take hours to emerge from their huddles (see supplementary statistics). Thus, conducting these trials over a duration which would not truncate the dataset for the ~ 10,000 spiders assayed here was not feasible.

Experimental Colony Construction

We created experimental colonies of 20 adult female spiders. 19 of these females exhibited a latency to resume movement of 600 s or greater (boldness index = 0), and the 20 spider was randomly pulled from a pool of spiders that varied in their boldness scores from 1–600. This created a set of experimental colonies of identical size and developmental stage, but which varied in the boldness score of the single boldest member of the group. Prior studies have referred to these individuals as ‘keystone individuals’ because they exhibit a disproportionately large influence over the collective behavior and success of their colony.

Experimental colonies were housed in 390 mL plastic cups containing three Acacia mellifera twigs to facilitate web construction. Colonies were provided two days to construct a rudimentary nest and capture web. Colonies were then run through three collective foraging aggressiveness assays (1/day for 3 days) before being deployed into the field. The temperature in the lodge/hotel rooms where individual boldness assays, colony assembly, and colony foraging aggressiveness assays occurred varied from 22.2–25.5C. All behavioral assays were conducted during daylight hours under artificial lighting in the absence of air drafts.

Colony Foraging Aggressiveness Assay

Colony foraging aggressiveness is quantified as the number of colony members that leave the nest during a staged prey capture event, where more aggressive colonies deploy more attackers [30]. Foraging aggressiveness assays are initiated by placing a 1 × 1cm piece of paper in the colony capture web ~5cm away from the nest entrance, and then vibrating the piece of paper using a 1.5mm diameter wire attached to the end of a handheld vibratory device (Magic Purple Bullet). The device was set on a pulse setting, as to mimic the irregular thrashing movements of insect prey captured in the web. We then recorded the number of spiders that left the nest to capture the simulated prey item, until the first individual contacted the piece of paper. Colony prey capture behavior was assessed once per day for three consecutive days. At the end of the third day, colonies were deployed at one of our eight field sites during the evening hours. Nearly all individuals that emerge in response to stimulated prey eventually attempted to attack and subdue the stimulus, which conveys that emergence is an indication of individuals’ desire to attack prey, and not merely to acquire information about it.

This assessment of colony aggressiveness does raise the potential concern that if one individual is very fast and contacts the paper rapidly, then its fellow colony members may not have time to emerge. However, at odds with this inference we actually find that fast-responding Stegodyphus colonies also attack prey with larger numbers of individuals in a smaller amount of time [30, 46, 50]. Additionally, the number of attackers that emerge in response to prey until contact is ecologically realistic because, once the first few individuals contact and subdue the prey, the attack ceases and the prey is dragged back to the next entrance where it is shared. Thus, in nature, the time for the attack sequence to occur is actually smaller for aggressiveness colonies. All colony aggressiveness assays were conducted during daylight hours under artificial lighting in the absence of air drafts in lodges and hotels.

Colony Deployment and Monitoring

Experimental colonies (28–33 colonies per sites) were deployed at our eight sites by using clothespins to attach the colony in its container to resident Acacia trees. We used the preexisting presence of S. dumicola colonies on nearby trees as an indicator of habitat quality and substrate suitability. Care was taken not to place colonies on Acacia trees actively being guarded or patrolled by ants (genus Anoplolepis and Crematogaster), which can destroy S. dumicola colonies before they establish [51, 52]. Prior studies have shown that setting colonies out in the evening hours provides them with an opportunity to produce a functional nest and a sizable capture web before the following sunrise [46].

We then left these experimental colonies in the field for the next several months and checked their performance in February 2017. During these checks, we determined if there were any spiders living within the colonies, noted the presence and number of egg cases, noted the presence of spiderlings, and, in cases where the colony had perished, we noted situational factors that provide clues as to why the colony may have perished. For instance, a large number of predatory ant carcasses remaining in the colony retreat and in the plastic deli cup was taken as evidence of a raid by predatory ants, which kill a large number of natural colonies each year. The presence of sporulated fungus emerging from the carcasses of colony members was interpreted as evidence that fouling by fungus had contributed to colony demise. Entomopathogenic fungus is a common driver of colony demise in this system [52]. In this study we focus on whether or not colonies survived the duration of study, meaning that at least one colony member was left alive, and whether or not colonies successfully reproduced, i.e., produced viable egg cases or spiderlings during this time period.

The phenology of our experimental colonies (e.g., timing of egg case production, presence of spiderlings, age of spiderlings) closely matched those of native naturally-occurring colonies. Experimental colonies were composed of mature females only that came from colonies containing mature males, and thus, were likely mated prior to colony establishment. Although, immigration of some foreign males to the colonies was evident during our February check.

Chimeric Colony Study

Because the influence of bold spiders on colony foraging behavior was only detected at arid sites, we used chimeric colonies (n = 96) to determine whether this dynamic was driven by the “social influence” of bold individuals from arid sites or the “social susceptibility” of shy individuals from arid sites. We created colonies of bold individuals from arid sites with shy individuals from wet sites (N = 48), and vice versa (N = 48). We also executed a set of control chimeras (n = 39) where we mixed individuals from different sites with similar climatic conditions (arid with arid [n = 24], wet with wet [n = 15]). This allowed us to determine if merely mixing colonies alters associations between group composition and group behavior, and whether the mechanisms governing the influence of bold individuals at arid sites are mutually cross-compatible. Mixtures involve 19 shy individuals from one population with 1 bold individual from another, and vice versa (Wet: Ladysmith → Rundu, Rundu → Ladysmith, Arid: Kalkrand → Upington, Upington → Kalkrand).

Chimeric colony studies were carried out under standardized laboratory conditions at the University of California Santa Barbara. Twenty intact S. dumicola colonies were collected form each site and transported to laboratory in January 2017. Colonies were dissected out and the prosoma width, mass, and boldness of each colony constituent were determined over a 72 hr period. We then constructed chimeric colonies (n = 96) of the same behavioral composition (19 spiders with a boldness index of 0, and 1 spider with a boldness index between 1 and 600). Colonies were then subjected to foraging aggressiveness assays identical to those detailed above.

To determine if the shy spiders from arid sites are socially susceptible to any indication of boldness in their group, we produced a third set of interspecific-chimeric colonies. Here we constructed colonies of 19 shy S. dumicola from two arid sites (Kalkrand, Namibia and Upington South Africa, total n = 48) and two wet sites (Rundu, Namibia and Ladysmith, South Africa, total n = 37) with a single heterospecific congener Stegodyphus mimosarum. Like S. dumicola, S. mimosarum exhibits temporally stable individual differences in boldness that are linked with individual social roles [38]. This species has independently evolved sociality and shared a common ancestor with S. dumicola several million years ago [39].

Web Recruitment Study

We designed a final study to determine 1) whether shy individuals from arid sites are uniquely drawn to the cues of conspecific on the capture web, and 2) whether food restriction could cause shy individuals from wet sites to become more socially susceptible (see Figure S1). To do this, we created colonies of 19 shy spiders or 19 bold spiders from two arid (Kalkrand, Namibia and Upington South Africa) and two wet (Rundu, Namibia and Ladysmith, South Africa) sites (total N = 33). We then subjected these colonies to two different sensory stimuli at three time points (in a fully factorial design), where each colony was tested in each of the six different treatment combinations. To test responsiveness to conspecifics we introduced a bold conspecific taken from the same source colony to the experimental colony’s capture web. To test change over time in responsiveness to prey, we introduced a domestic cricket on the capture web. Colonies were presented with each of these stimuli in a randomized order one, two, and five days after an ad libitum feeding event and we recorded the number of spiders that responded to each of these stimuli. The ordering of prey-response treatments (1, 2 or 5 days post-feeding) was likewise alternated among focal colonies.

Timeline

  • Colony Establishment in the Field: November-December 2016

  • Colony Performance Assessment: February 2017, July 2017

  • Chimeric Colony Experiment (Lab): February-April 2017

  • Web Recruitment Studies: April-May 2017

QUANTIFICATION AND STATISTICAL ANALYSIS

Our statistical analyses were performed in JMP 12.0. In brief, we constructed a variety of GLMs and LMMs to evaluate associations between group composition and various behavioral and colony performance response variables (see Data S1 for full model descriptions). Statistical significance was evaluated using an α = 0.05. Sample sizes were determined by the availability of spiders of specific behavioral tendencies and general feasibility based on personnel limitations. No replicates were excluded from our statistical analyses. The distribution of residuals and the nature of our response variables (e.g., binary versus continuous) were used to determine our model structures.

DATA AND SOFTWARE AVAILABILITY

Behavioral and colony performance data summarized in this paper are available in the Dryad Digital Repository: https://doi.org/10. 5061/dryad.7p090.

Supplementary Material

Supplemental1

Supplemental2

Supplemental1

Click here to view.(366K, pdf)

Supplemental2

Click here to view.(378K, docx)

Acknowledgments

We are indebted to J.W. Jolles, S.E. Bengston, R. Wollman, C.N. Keiser, J.E. Smith, D.R. Farine, A. Gardner, and two anonymous reviewers for comments on prior versions of this manuscript. Funding was provided by NSF IOS grants 1352705 and 1455895 to JNP, 1456010 to N.P.-W. and NIH grant GM115509 to N.P.-W. and J.N.P.

Department of Ecology, Evolution and Marne Biology, University of California, Santa Barbara, Santa Barbara, CA 93106, USA
Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona, Tucson, AZ 85721, USA
Department of Biological Sciences, University of Pittsburgh, Pittsburgh, PA 37906, USA
Department of Ecology and Evolutionary Biology, University of California, Los Angeles, Los Angeles, CA 90095, USA
Correspondence: moc.liamg@sisponelega
These authors contributed equally
Lead Contact

SUMMARY

Particularly socially influential individuals are present in many groups [18], but it is unclear whether their emergence is determined by their social influence versus the social susceptibility of others [9]. The social spider Stegodyphus dumicola shows regional variation in apparent leader-follower dynamics. We use this variation to evaluate the relative contributions of leader social influence versus follower social susceptibility in driving this social order. Using chimeric colonies that combine potential leaders and followers, we discover that leader-follower dynamics emerge from the site-specific social susceptibility of followers. We further show that the presence of leaders increases colony survival in environments where leader-follower dynamics occur. Thus, leadership is driven by the “social susceptibility” of the population majority, rather than the social influence of key group members.

SUMMARY
Highlights

Footnotes

SUPPLEMENTAL INFORMATION

Supplemental Information includes two figures and one data file and can be found with this article online at https://doi.org/10.1016/j.cub.2017.11.038.

AUTHOR CONTRIBUTIONS

J.N.P., C.M.W., J.L.L.L., G.T.C., and B.L.M. all contributed to study design, data collection, writing, and analyses. A.K. contributed to writing, data analyses, and figure production. N.P.-W. contributed to study design, writing, and data analyses.
Footnotes

References

  • 1. Beauchamp GIndividual differences in activity and exploration influence leadership in pairs of foraging zebra finches. Behaviour. 2000;137:301–314.[PubMed][Google Scholar]
  • 2. Fischhoff IR, Sundaresan SR, Cordingley J, Larkin HM, Sellier MJ, Rubenstein DISocial relationships and reproductive state influence leadership roles in movements of plains zebra, Equus burchellii. Anim Behav. 2007;73:825–831.[PubMed][Google Scholar]
  • 3. Flack A, Pettit B, Freeman R, Guilford T, Biro DWhat are leaders made of? The Role of individual experience in determining leader-follower relations in homing pigeons. Anim Behav. 2012;83:703–709.[PubMed][Google Scholar]
  • 4. Guttridge TL, Gruber SH, DiBattista JD, Feldheim KA, Croft DP, Krause S, Krause JAssortative interactions and leadership in a free-ranging population of juvenile lemon shark Negaprion brevirostris. Mar Ecol Prog Ser. 2011;423:235–245.[PubMed][Google Scholar]
  • 5. Jolles JW, Fleetwood-Wilson A, Nakayama S, Stumpe MC, Johnstone RA, Manica AThe role of previous social experience on risk-taking and leadership in three-spined sticklebacks. Behav Ecol. 2014;25:1395–1401.[PubMed][Google Scholar]
  • 6. Krause J, Reeves P, Hoare DPositioning behaviour in roach shoals: the role of body length and nutritional state. Behaviour. 1998;135:1031–1039.[PubMed][Google Scholar]
  • 7. McClure M, Ralph M, Despland EGroup leadership depends on energetic state in a nomadic collective foraging caterpillar. Behav Ecol Sociobiol. 2011;65:1573–1579.[PubMed][Google Scholar]
  • 8. Conradt L, Roper TJConsensus decision making in animals. Trends Ecol Evol. 2005;20:449–456.[PubMed][Google Scholar]
  • 9. King AJFollow me! I’m a leader if you do; I’m a failed initiator if you don’t? Behav Processes. 2010;84:671–674.[PubMed][Google Scholar]
  • 10. Sumana A, Sona CKey relocation leaders in an Indian queenless ant. Behav Processes. 2013;97:84–89.[PubMed][Google Scholar]
  • 11. Modlmeier AP, Keiser CN, Watters JV, Sih A, Pruitt JNThe keystone individual concept: an ecological and evolutionary overview. Anim Behav. 2014;89:53–62.[PubMed][Google Scholar]
  • 12. Smith JE, Gavrilets S, Mulder MB, Hooper PL, El Mouden C, Nettle D, Hauert C, Hill K, Perry S, Pusey AE, et al Leadership in mammalian societies: emergence, distribution, power, and payoff. Trends Ecol Evol. 2016;31:54–66.[PubMed][Google Scholar]
  • 13. Couzin ID, Krause J, Franks NR, Levin SAEffective leadership and decision-making in animal groups on the move. Nature. 2005;433:513–516.[PubMed][Google Scholar]
  • 14. Robson SK, Traniello JFA. Key individuals and the organisation of labor in ants. In: Deneubourg DC, Pasteels JM, editors. Information Processing in Social Insects. Birkhauser Verlag; 1999. pp. 239–259. [PubMed]
  • 15. McComb K, Shannon G, Durant SM, Sayialel K, Slotow R, Poole J, Moss CLeadership in elephants: the adaptive value of age. Proc Biol Sci. 2011;278:3270–3276.[Google Scholar]
  • 16. Brent LJN, Franks DW, Foster EA, Balcomb KC, Cant MA, Croft DPEcological knowledge, leadership, and the evolution of menopause in killer whales. Curr Biol. 2015;25:746–750.[PubMed][Google Scholar]
  • 17. Gilby IC, Machanda ZP, Mjungu DC, Rosen J, Muller MN, Pusey AE, Wrangham RW‘Impact hunters’ catalyse cooperative hunting in two wild chimpanzee communities. Philos Trans R Soc Lond B Biol Sci. 2015;370:20150005.[Google Scholar]
  • 18. Harcourt JL, Ang TZ, Sweetman G, Johnstone RA, Manica ASocial feedback and the emergence of leaders and followers. Curr Biol. 2009;19:248–252.[PubMed][Google Scholar]
  • 19. Van Vugt M, Hogan R, Kaiser RBLeadership, followership, and evolution: some lessons from the past. Am Psychol. 2008;63:182–196.[PubMed][Google Scholar]
  • 20. King AJ, Johnson DDP, Van Vugt MThe origins and evolution of leadership. Curr Biol. 2009;19:R911–R916.[PubMed][Google Scholar]
  • 21. Kurvers R, Eijkelenkamp B, van Oers K, van Lith B, van Wieren SE, Ydenberg RC, Prins HHTPersonality differences explain leadership in barnacle geese. Anim Behav. 2009;78:447–453.[PubMed][Google Scholar]
  • 22. Aplin LM, Farine DR, Mann RP, Sheldon BCIndividual-level personality influences social foraging and collective behaviour in wild birds. Proc Biol Sci. 2014;281:20141016.[Google Scholar]
  • 23. Rands SA, Cowlishaw G, Pettifor RA, Rowcliffe JM, Johnstone RASpontaneous emergence of leaders and followers in foraging pairs. Nature. 2003;423:432–434.[PubMed][Google Scholar]
  • 24. Rands SA, Cowlishaw G, Pettifor RA, Rowcliffe JM, Johnstone RAThe emergence of leaders and followers in foraging pairs when the qualities of individuals differ. BMC Evol Biol. 2008;8:51.[Google Scholar]
  • 25. Lee PC, Moss CJWild female African elephants (Loxodonta africana) exhibit personality traits of leadership and social integration. J Comp Psychol. 2012;126:224–232.[PubMed][Google Scholar]
  • 26. Wickler W, Seibt UPedogenetic sociogenesis via the sibling-route and some consequences for Stegodyphus spiders. Ethology. 1993;95:1–18.[PubMed][Google Scholar]
  • 27. Aviles L. Causes and Consequences of Cooperation and Permanent Sociality in Spiders. In: Choe C, Crespi B, editors. The Evolution of Social Behaviour in Insects and Arachnids J. Cambridge University Press; 1997. pp. 476–498. [PubMed]
  • 28. Wright CM, Keiser CN, Pruitt JNPersonality and morphology shape task participation, collective foraging and escape behaviour in the social spider Stegodyphus dumicola. Anim Behav. 2015;105:47–54.[PubMed][Google Scholar]
  • 29. Wright CM, Keiser CN, Pruitt JNColony personality composition alters colony-level plasticity and magnitude of defensive behaviour in a social spider. Anim Behav. 2016;115:175–183.[PubMed][Google Scholar]
  • 30. Pruitt JN, Keiser CNThe personality types of key catalytic individuals shape colonies’ collective behaviour and success. Anim Behav. 2014;93:87–95.[PubMed][Google Scholar]
  • 31. Grinsted L, Pruitt JN, Settepani V, Bilde TIndividual personalities shape task differentiation in a social spider. Proc Biol Sci. 2013;280:20131407.[Google Scholar]
  • 32. Keiser CN, Jones DK, Modlmeier AP, Pruitt JNExploring the effects of individual traits and within-colony variation on task differentiation and collective behavior in a desert social spider. Behav Ecol Sociobiol. 2014;68:839–850.[PubMed][Google Scholar]
  • 33. Lichtenstein JLL, Wright CM, Luscuski LP, Montgomery GA, Pinter-wollman N, Pruitt JNParticipation in cooperative prey capture and the benefits gained from it are associated with individual personality. Curr Zool. 2017;63:561–567.[Google Scholar]
  • 34. Pruitt JN, Pinter-Wollman NThe legacy effects of keystone individuals on collective behaviour scale to how long they remain within a group. Proc Biol Sci. 2015;282:89–96.[Google Scholar]
  • 35. Pruitt JN, Wright CM, Keiser CN, DeMarco AE, Grobis MM, Pinter-Wollman NThe Achilles’ heel hypothesis: misinformed keystone individuals impair collective learning and reduce group success. Proc Biol Sci. 2016;283:20152888.[Google Scholar]
  • 36. Riechert SEThe evolution of behavioral phenotypes: lessons learned from divergent spider populations. Adv Study Behav. 1993;22:103–134.[PubMed][Google Scholar]
  • 37. Hodge MA, Uetz GWA comparison of agonistic behavior of colonial web-building spiders from desert and tropical habitats. Anim Behav. 1995;50:963–972.[PubMed][Google Scholar]
  • 38. Laskowski KL, Pruitt JNEvidence of social niche construction: persistent and repeated social interactions generate stronger personalities in a social spider. Proc Biol Sci. 2014;281:20133166.[Google Scholar]
  • 39. Johannesen J, Lubin Y, Smith DR, Bilde T, Schneider JMThe age and evolution of sociality in Stegodyphus spiders: a molecular phylogenetic perspective. Proc Biol Sci. 2007;274:231–237.[Google Scholar]
  • 40. Smith JE, Estrada JR, Richards HR, Dawes SE, Mitsos K, Holekamp KECollective movements, leadership and consensus costs at reunions in spotted hyaenas. Anim Behav. 2015;105:187–200.[PubMed][Google Scholar]
  • 41. Laskowski KL, Bell AMStrong personalities, not social niches, drive individual differences in social behaviours in sticklebacks. Anim Behav. 2014;90:287–295.[Google Scholar]
  • 42. Laskowski KL, Montiglio PO, Pruitt JNIndividual and group performance suffers from social niche disruption. Am Nat. 2016;187:776–785.[PubMed][Google Scholar]
  • 43. Modlmeier AP, Laskowski KL, DeMarco AE, Coleman A, Zhao K, Brittingham HA, McDermott DR, Pruitt JNPersistent social interactions beget more pronounced personalities in a desert-dwelling social spider. Biol Lett. 2014;10:2014–2019.[Google Scholar]
  • 44. Schneider JM, Bilde TBenefits of cooperation with genetic kin in a subsocial spider. Proc Natl Acad Sci USA. 2008;105:10843–10846.[Google Scholar]
  • 45. Wilson DS, Coleman K, Clark AB, Biederman LShy bold continuum in pumpkinseed sunfish (Lepomis gibbosus): an ecological study of a psychological trait. J Comp Psychol. 1993;107:250–260.[PubMed][Google Scholar]
  • 46. Keiser CN, Pruitt JNPersonality composition is more important than group size in determining collective foraging behaviour in the wild. Proc Biol Sci. 2014;281:20141424.[Google Scholar]
  • 47. Pinter-Wollman N, Keiser CN, Wollman R, Pruitt JNThe effect of keystone individuals on collective outcomes can be mediated through interactions or behavioral persistence. Am Nat. 2016;188:240–252.[Google Scholar]
  • 48. Keiser CN, Modlmeier AP, Singh N, Jones DK, Pruitt JNExploring how a shift in the physical environment shapes individual and group behavior across two social contexts. Ethology. 2014;120:825–833.[PubMed][Google Scholar]
  • 49. Keiser CN, Wright CM, Pruitt JNIncreased bacterial load can reduce or negate the effects of keystone individuals on group collective behaviour. Anim Behav. 2016;114:211–218.[PubMed][Google Scholar]
  • 50. Pruitt JN, Grinsted L, Settepani VLinking levels of personality: personalities of the ‘average’ and ‘most extreme’ group members predict colony-level personality. Anim Behav. 2013;86:391–399.[PubMed][Google Scholar]
  • 51. Keiser CN, Wright CM, Pruitt JNWarring arthropod societies: social spider colonies can delay annihilation by predatory ants via reduced apparency and increased group size. Behav Processes. 2015;119:14–21.[PubMed][Google Scholar]
  • 52. Henschel JRPredation on social and solitary individuals of the spider Stegodyphus dumicola (Araneae, Eresidae) J Arachnol. 1998;26:61–69.[PubMed][Google Scholar]
Collaboration tool especially designed for Life Science professionals.Drag-and-drop any entity to your messages.