Habitat choice and complex decision making in a trap-building predator

Behavioral Ecology The official journal of the ISBE International Society for Behavioral Ecology Behavioral Ecology (2016), 27(5), 1491–1498. doi:10.1093/beheco/arw071 Original Article Shay Adar, Roi Dor, and Inon Scharf Department of Zoology, Faculty of Life Sciences, Tel Aviv University, 69978 Tel Aviv, Israel Received 30 June 2015; revised 10 April 2016; accepted 12 April 2016; Advance Access publication 11 May 2016. The selection of a suitable habitat may maximize the survival and reproduction chances of individuals. We examined the habitat choice of wormlions, small sand-dwelling insects that construct pit-traps to capture prey. Unlike the well-studied antlions, which employ a similar hunting strategy, wormlion behavior has not been thoroughly studied. We first determined whether wormlions prefer 1) shaded versus lit, 2) deep versus shallow sand, and 3) surface-obstructed versus clear microhabitats. The first option of each of the three pairs was highly preferred. In unfavorable microhabitats, larvae were also less likely to construct a pit, mainly in shallow sand, and moved over longer distances. Next, we sought to determine whether the choice is additive (taking more than a single factor into account), hierarchical (ranking the factors according to an order of preference), or based on a threshold rule (accepting a microhabitat that crosses a quality threshold). Combining shade with one of 3 environmental factors – surface obstacles, shallow sand, or high conspecific density – led to lower proportions of wormlions choosing shade. In other words, a combination of shade and any factor was no longer preferred over lit and otherwise favorable microhabitats. We thus conclude that the wormlions’ microhabitat choice is additive, taking more than a single factor into account, and that favorable conditions of any sort are traded-off against one another. Key words: density, habitat selection, photokinesis, psammophile, trap-building predators, Vermileonidae, Wormlion. INTRODUCTION Animals usually select habitats, oviposition sites, or mating partners by comparing multiple available alternatives and choosing one according to the costs and benefits associated with each option. This decision is often simplified by either reducing the number of alternatives considered, the number of attributes considered for each alternative, or the decision accuracy (Körding 2007; Chittka et al. 2009). A decision can involve 1) a process of elimination of alternatives that do not qualify according to a specific attribute, and choosing the first that suffices, that is, a threshold rule; or 2) a comparative process in which alternatives are scored according to certain attributes and then the best one is chosen, that is, a best-ofN method (see Payne et al. 1993 and Franks et al. 2003 for more precise and elaborate definitions of decision types). The latter type of decision can be performed in a strictly hierarchical manner, meaning that certain attributes are always prioritized over others. Alternatively, several attributes can be simultaneously scored and given different or equal weights in an additive manner. An example of a hierarchical choice is that of nest relocation in cavity-dwelling ants. Nests are chosen according to attributes that differ in their importance: darkness is more important than nest height, which is more important than entrance width (Franks et al. 2003). Desert Address correspondence to I. Scharf. E-mail: . © The Author 2016. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: spiders, in contrast, employ a threshold rule, and prefer a specific range of temperatures and high prey availability. They do not rank the sites according to all attributes, but seem to stop searching once the first adequate site has been found, that is, either high prey availability or a favorable temperature (Riechert 1985). While foraging, different environmental factors, such as safety and food availability, are traded-off with one another and a compromise has to be made (e.g., Verdolin 2006; Berger-Tal et al. 2010). Other factors affecting foraging decisions are also relevant, such as habitat quality and conspecific density (Clobert et al. 2009). For instance, the web-site tenacity of orb-weaving spiders is influenced both by abiotic factors, such as humidity and the physical structure of the microhabitat, and biotic factors, mainly conspecific density, which creates direct and indirect competition (Janetos 1986; Ward and Lubin 1993; Samu et al. 1996). In addition, past experience can affect habitat choice (Davis and Stamps 2004; Segura et al. 2007). Animals that have developed in low-quality habitats tend to be less selective later as adults when choosing a new habitat, compared with animals that have developed in rich habitats (Davis 2008). Moreover, previous success (e.g., in mating or foraging) is an important factor that affects an animal’s choice of whether to return to a particular habitat (Switzer 1997; Barron 2001). Pit-building antlions and wormlions are examples of relatively sedentary, sit-and-wait larvae and mobile dispersive flying adults (Scharf et al. 2011). Owing to the limited mobility of the larvae, Habitat choice and complex decision making in a trap-building predator 1492 METHODS Animal collection We collected 170 wormlion larvae (mass of 6.72 ± 3.11 mg, mean ± 1 SD) from 2 adjacent sandy patches in north Tel Aviv, west of Tel Aviv University (32°6′33.0″N 34°47′50″E) in winter 2014 and 2015. Log-transformed body mass and length are tightly correlated (r = 0.947, n = 100, P < 0.001; mean length of wormlions collected during the same season ±1 SD: 10.58 ± 1.83 mm). In the laboratory the larvae were placed in individual plastic cups (4.5 cm diameter) filled with 3 cm-deep sand. Prior to each experiment, each wormlion larva was fed with a small flour beetle (Tribolium castaneum) larva, and weighed using an analytical balance (accuracy of 0.1 mg). It was then left for 3 days in order to standardize hunger level. Body mass was measured in order to examine whether it affects habitat choice and movement distance (see below). Sand was collected from a nearby area and filtered to remove foreign objects (plant matter and small stones). Traces in the sand that the wormlions left while moving during the experiments described below were marked and movement distances were measured using the software ImageJ (Abramoff et al. 2004). A piece of millimetric paper was included in the photos to serve as a scale. Experiments were conducted at room temperature (fluctuations of ±1°C) and relative humidity inside and outside the trays was similar (54.6 ± 2.9%). The room was constantly lit by ceiling fluorescent lights (25.63 ± 10.9 LUX in the shade, 104 ± 21.2 LUX in the light, measured with a Sper Scientific 840020 Dual-Scale Light Meter). Microhabitat selection I: light versus (...truncated)