Extended spider cognition

Abstract

There is a tension between the conception of cognition as a central nervous system (CNS) process and a view of cognition as extending towards the body or the contiguous environment. The centralised conception requires large or complex nervous systems to cope with complex environments. Conversely, the extended conception involves the outsourcing of information processing to the body or environment, thus making fewer demands on the processing power of the CNS. The evolution of extended cognition should be particularly favoured among small, generalist predators such as spiders, and here, we review the literature to evaluate the fit of empirical data with these contrasting models of cognition. Spiders do not seem to be cognitively limited, displaying a large diversity of learning processes, from habituation to contextual learning, including a sense of numerosity. To tease apart the central from the extended cognition, we apply the mutual manipulability criterion, testing the existence of reciprocal causal links between the putative elements of the system. We conclude that the web threads and configurations are integral parts of the cognitive systems. The extension of cognition to the web helps to explain some puzzling features of spider behaviour and seems to promote evolvability within the group, enhancing innovation through cognitive connectivity to variable habitat features. Graded changes in relative brain size could also be explained by outsourcing information processing to environmental features. More generally, niche-constructed structures emerge as prime candidates for extending animal cognition, generating the selective pressures that help to shape the evolving cognitive system.

Keywords: Evolvability; Extended cognition; Modular cognition; Niche construction; Web building.

Fig. 1

Cycle of actions necessary to build the current segment of the adhesive spiral. Steps and processes within the cycle (a), with the illustration of some of the behaviours involved (b adapted from Eberhard and Wcislo 2011). The cycle (blue arrows in a) begins and ends with the fixation of the current adhesive spiral segment (blue box). The spider fix (the current segment) over the current radius (R_n, a; spider behaviour displayed at b, top figure), and then in the next radius (R_n+1, a; spider behaviour displayed at b, bottom figure). The spider performs successive actions (large blue arrows, a), while assessing the position of some rapidly changing cues (coloured balls, A and B). Slowly and rapidly changing cues are stored, compared to each other (to obtain distances and rates of change) and then integrated (continuous and thin blue lines, a) to determine the position of the next adhesive segment fixation (in R_n+1). When confronted with conflicting cues, the spider may ignore some cues (inner-loop sensing), and proceed to a shortcut, an alternative routine (dotted blue line, a). These cycles are repeated until the completion of the capture area

Fig. 2

Matched filters provide high-level information to the CNS, but do not extend cognition. Parasitoid wasps use the curvature (angle BCD) of the host insect egg to determine the number of progeny allocated to it. The angle between head and scapus (angle BCD) is correlated with the radius of the egg. Figure adapted from Wehner (1987)

Fig. 3

Graded changes in brain–body allometry (grey dots at the basal plane, positive Central cognition-body size correlation) do not follow from Haller’s rule. A possible explanation for the stepped correlation across taxonomic groups (each of the three distinct clouds of points) would be the relative amount of non-brain based information processing (negative correlation between Extended cognition and Body size, grey dots at the right plane). Relatively higher degrees of extended cognition would help animals to maintain performance at much smaller body sizes. The correlation between cognition (both central and extended) and body size is represented by the red dots