18.1 Introduction

Social learning—learning influenced by observation of, or interaction with, other animals (Heyes, 1994; Hoppitt & Laland, 2008)—allows individuals to acquire information, concerning, for instance, the location and quality of food, mates, predators, rivals, and pathways, as well as foraging techniques, vocalizations and a variety of social behavior (Byrne, 2009; Heyes & Galef, 1996; Laland, Atton, & Webster, 2011; Laland & Galef, 2009). Social learning is typically adaptive because it can act as a short‐cut to acquiring optimal, or high‐payoff, behavior, avoiding the relative costs of individual “trial‐and‐error” learning (Rendell et al., 2010). It is therefore unsurprising that a growing body of experimental and observational studies has reported social learning in a wide range of species, including mammals, reptiles, fish, birds, amphibians and insects (Box & Gibson, 1999; Brown & Laland, 2003; Emery, 2006; Ferrari & Chivers, 2008; Leadbeater & Chittka, 2007; Wilkinson, Kuenstner, Mueller, & Huber, 2010). While animals of many species are capable of social learning, there is variation in the extent to which they are reliant on, and perhaps specially adapted for, the exploitation of socially learned information. Specific social learning processes are rarer than the general capability: for instance, chimpanzees are seemingly capable of imitative learning in which a specific action is replicated with high fidelity (Whiten et al., 2007); other species, like nine‐spined sticklebacks, appear to deploy specific social learning strategies, such as payoff‐based copying (Coolen, Ward, Hart, & Laland, 2005). It is therefore interesting to consider why species vary in how much they rely on social learning for survival and reproduction, why some species might possess enhanced social learning skills, in terms of mechanisms or strategies, and what role these abilities might have played in brain evolution. It may not be coincidence that taxa with enlarged forebrains and which are commonly regarded as highly intelligent, such as primates, certain radiations of birds and cetaceans, also happen to be heavily reliant on social learning and traditions.

18.2 Brain Enlargement, Intelligence, and Social Learning

Intelligence can be broadly defined as a cross‐domain measure of cognitive ability in learning, problem‐solving and abstract reasoning, characterized by behavioral flexibility (Jolly, 1966; Reader, Hager, & Laland, 2011; Reader & Laland, 2002). High intelligence in a species may be therefore be suggested by experimental and observational evidence for the flexible use of such behavior as extractive foraging (in the absence of specialized anatomy), food processing, tactical deception, tool use, causal understanding, problem solving, and complex learning (Byrne, 1995; Emery & Clayton, 2004; Huber & Gajdon, 2006; van Schaik, Deaner, & Merrill, 1999; Visalberghi & Tomasello, 1998; Whiten & Byrne, 1997). While the term “intelligence” is often avoided by contemporary animal behavior researchers, who typically address aspects of cognition in a more domain‐specific manner, nonetheless interest in comparative cognition seems to have reached its zenith (Shettleworth, 2010). It seems to be the case that researchers remain interested in intelligent behavior, and its evolutionary origins, but that they are conscious of the difficulties of fair comparison of intelligence across diverse taxa, or suspicious of the notion of general intelligence, leaving them reticent to use the term (Mackintosh, 1988). Although the cognitive capabilities of each species are uniquely adapted to the requirements of their niches, this does not preclude that certain species may justifiably be considered more generally intelligent across domains than others (Reader et al., 2011). For instance, primate genera differ in their performance across diverse laboratory tests of cognition, with great apes consistently outperforming other primates (Deaner, Van Schaik, & Johnson, 2006; Tomasello & Call, 1997). It may be difficult to compare intelligence across species, but the suggestion that all nonhuman vertebrates are equally intelligent (Macphail, 1982) is contradicted by extensive evidence (Byrne, 1995; Deaner et al., 2006; Lefebvre, Reader, & Sol, 2004; Reader & Laland, 2002; Tomasello & Call, 1997). Further, the view that intelligence is a meaningful concept across species does not require the assumption of a scala naturae—the idea that intelligence decreases with phylogenetic distance from humans (Jensen, 1980). On the contrary, recent evidence supports the view that there has been convergent evolution of intelligence in distant taxa, especially in primates, corvids, and toothed whales (Emery & Clayton, 2004; Reader et al., 2011; Rendell & Whitehead, 2001).

Taxa considered to possess high intelligence, such as primates, cetaceans and birds; specifically monkeys, apes, toothed whales, songbirds and parrots; have undergone convergent enlargement of the forebrain, relative to sister taxa, in terms of deviations from expected scaling with whole brain and body size (Alonso, Milner, Ketcham, Cookson, & Rowe, 2004; Barton & Harvey, 2000; Cnotka, Gunturkun, Rehkamper, Gray, & Hunt, 2008; Emery, 2006; Jerison, 1973; Marino, McShea, & Uhen, 2004; Rendell & Whitehead, 2001; Zelenitsky, Therrien, & Kobayashi, 2009). Although birds lack the mammalian neocortex, the avian pallium is structurally similar and lesions to the caudolateral nidopallium result in specific cognitive impairment, suggesting that the avian and mammalian forebrains are analogous (Cnotka et al., 2008; Emery, 2006; Huber & Gajdon, 2006). Despite centuries of interest in the evolution of intelligence, the intuitively appealing notion that brain volume and intelligence are linked commands surprisingly little support, although some compelling evidence does exist in birds and primates (Byrne & Corp, 2004; Deaner et al., 2006; Johnson, Deaner, & Van Schaik, 2002; Lefebvre et al., 2004; Lefebvre, Whittle, Lascaris, & Finkelstein, 1997; Reader et al., 2011; Reader & Laland, 2002; Sol, Duncan, Blackburn, Cassey, & Lefebvre, 2005). However, it remains unclear what cognitive benefits brain size expansion brings. Brain expansion potentially is associated with increases in the number of neurons, local and ‘long range’ connectivity and the number of cortical areas, thereby in a general sense potentially expanding the amount of ‘processing power’ available to an animal (Byrne & Bates, 2007; Changizi & Shimojo, 2005; Striedter, 2005). Striedter (2005, p. 11) notes a general rule of brain evolution that “large equals well connected.”

Trends in vertebrate brain size evolution commonly follow simple allometric scaling rules, with individual brain regions evolving in concert with each other (Finlay & Darlington, 1995). However, certain regions in some taxa appear to vary in size independently of other brain areas, and appear to be associated with evolutionary changes in behavior (Barton & Harvey, 2000; Striedter, 2005). For example, expansion of the primate neocortex (Barton & Harvey, 2000; Byrne & Corp, 2004) is predicted by various social and ecological variables in primates (Alport, 2004; Amici, Aureli, & Call, 2008; Barton, 1996; Dunbar, 1992). Likewise, comparative studies have demonstrated that hippocampus enlargement is associated with food‐caching behavior (Basil, Kamil, Balda, & Fite, 1996; Hampton, Sherry, Shettleworth, Khurgel, & Ivy, 1995; Healy & Krebs, 1992; Krebs, Sherry, & Healy, 1989; Sherry, Vaccarino, Buckenham, & Herz, 1989) and that enlargement of the higher vocal centre is associated with vocal repertoire in birds (Devoogd, Krebs, Healy, & Purvis, 1993; Szekely, Catchpole, Devoogd, Marchl, & Devoogd, 1996). Specific neural expansion in association with adaptive specialization supports the general link between brain size and reliance on specific behavior. It is important to recognize, however, that some variation in brain component size across species may be affected by neural plasticity, reflecting usage within the lifetime; hence within and between species differences in neural structure is not inherently evidence for adaptive specialization (Bolhuis & Macphail, 2001).

Comparative studies show evidence of a relationship between social learning, brain size and intelligence in primates. Across primate species, Reader and Laland (2002) demonstrated that there is a positive correlation between the reported incidence of social learning in a given primate species, corrected for research effort, and both relative and absolute measures of brain volume, in a phylogenetically controlled analysis (Figure 18.1). Biases in research effort were corrected by using, as a corrected frequency, the residuals from a regression of the frequency of social learning against the frequency of published articles on the species in the Zoological Record. Relative brain size was measured as “executive brain ratio”: the ratio of the volume of the neocortex and striatum (executive brain) to the volume of the mesencephalon and medulla (brainstem). However, social learning incidence also correlated positively with further measures of brain size, including absolute executive brain volume and residuals of a plot of executive brain against brainstem volume, although the latter was not statistically significant (Reader & Laland, 2002).

In addition to the relationship between social learning and brain size, Reader and Laland (2002) showed that the corrected incidence of behavioral innovation also co‐varied with measures of brain size across primates, echoing a similar finding in birds (Lefebvre, Whittle, Lascaris, & Finkelstein, 1997). Moreover, the corrected incidence of social learning is positively correlated with corrected frequencies of behavioral innovation and tool use, suggesting that social learning ability could be a component of a cross‐species general intelligence factor, an idea further explored by Reader et al. (2011). In principal component and factor analyses of various naturalistic indicators of behavioral flexibility (including corrected measures of social learning, tool use, innovation, tactical deception, and extractive foraging, in >60 primate species), Reader et al (2011) extracted a single component/factor, which explained over 65% of the variance. In a further analysis, which included three additional socio‐ecological variables (diet breadth, percentage fruit in diet, group size), a major component explained 47% of the variance, on which the five behavioral flexibility measures loaded, plus diet breadth (to a lesser extent); although a second component was also extracted, on which tactical deception, group size, and percentage fruit loaded. These results, which hold when the analyses are conducted at the genus level, when phylogeny is controlled for, and when the apes are removed, strongly imply that aspects of behavioral flexibility co‐vary, and are evocative of the notion of a cross‐species general intelligence (Reader et al., 2011). This statistical association need not imply that the various measures are reliant on the same brain regions or circuits, although this is a possibility.

Support for the association of high general intelligence and brain size expansion comes from the observation that individual species loadings on the principle component (termed g_s), which can be regarded as measures of the general intelligence of the species, are strongly associated with several measures of brain volume, including neocortex/whole brain ratio, executive brain/brainstem ratio, and absolute neocortex size, although not residuals of neocortex on the rest of the brain (Reader et al., 2011). Recent research suggests that relationships between socio‐ecological predictors and brain size vary according to which method of brain measurement (e.g., neocortex/whole brain residuals versus ratios) is employed, although the reasons for this are not well understood (Deaner, Nunn, & Van Schaik, 2000). Further, Reader et al.’s g_s measure was found to correlate positively with measures of performance in laboratory tests of cognition. Finally, when the g_s measure was mapped onto a primate phylogeny, Reader et al. revealed evidence for convergent evolution of enhanced intelligence in four primates groups—great apes, macaques, baboons, and capuchins—precisely those taxa renowned for reliance on social learning and traditions (Figure 18.2; Cambefort, 1981; Ferrari et al., 2006; Hirata, Watanabe, & Kawai, 2001; Huffman, 1984; Kawai, 1965; Perry, 2011; Petit & Thierry, 1993; van Schaik et al., 2003; Watanabe, 1994; Whiten et al., 1999).

In addition to comparative studies, the taxonomic distribution of certain types of social learning seems to support the idea of a relationship between social learning, intelligence, and large brains. Experimental reports of high‐fidelity social learning are concentrated in primates, cetaceans and birds (Hoppitt & Laland, 2008). For example, the ability to learn motor behavior imitatively has been demonstrated in humans, chimpanzees, marmosets and birds, and vocal behavior in cetaceans and birds (Catchpole & Slater, 2008; Hoppitt & Laland, 2008; Janik & Slater, 2000; Rendell & Whitehead, 2001; Voelkl & Huber, 2000). Observational studies and circumstantial accounts of these taxa also attest to their imitative ability, although many such examples have not been systematically or experimentally verified, and it is difficult to extract specific learning processes from observational accounts. However, when specific motor patterns, which are outside of the normal behavioral repertoire or lack apparent utility, are closely reproduced, many researchers infer imitation. In primates, examples include able‐bodied chimpanzees reproducing the idiosyncratic actions of disabled individuals (Byrne, 2009; Hobaiter & Byrne, 2010), and the copying of a variety of human behaviors in orang‐utans (Russon & Galdikas, 1993). Captive dolphins have been reported anecdotally to replicate spontaneously a variety of actions outside of their behavioral range, including the swimming motion of pinnipeds (Tayler & Saayman, 1973) and tool use of humans (Kuczaj, Gory, & Xitco, 1998). Vocal imitation is a common behavior in cetaceans, demonstrated by examples of captive dolphins and killer whales adopting unfamiliar vocal repertoires, coordinated change over time in humpback whale song and duetting in sperm whales (Rendell & Whitehead, 2001). Songbirds and parrots are highly proficient vocal mimics, many of which are able to closely replicate a wide variety of sounds outside their species’ normal vocal range (Emery, 2006; Janik & Slater, 2000). Primates are not known to be especially proficient at vocal imitation, although duetting behavior in gibbons is an exception (Janik & Slater, 2000).

Social learning that results in traditions and “animal culture”—socially transmitted behaviors which endure over many generations and spread throughout a population (Laland & Galef, 2009; Whiten et al., 1999)—is less common than social learning in general, and has been reported in primates, cetaceans, and birds. Studies in which behavioral variation between animal communities is systematically documented have demonstrated considerable ranges of group‐specific behaviors, arguably suggestive of socially learned traditions, including in the domains of foraging, social behavior, tool use and/or vocalizations, in chimpanzees (Whiten et al., 1999), orang‐utans (Bastian, Zweifel, Vogel, Wich, & Van Schaik, 2010; van Schaik et al., 2003), capuchins (Perry, 2011), spider monkeys (Santorelli et al., 2011), Japanese macaques (Leca, Gunst, & Huffman, 2007), killer whales and sperm whales (Rendell & Whitehead, 2001), bowerbirds (Madden, 2008), passerines (Janik & Slater, 2000) and corvids (Bluff, Kacelnik, & Rutz, 2010). However, traditions are by no means exclusive to larger brained species. Fishes too exhibit traditional use of mating sites, schooling sites, resting sites, and pathways through the reef (Helfman & Schultz, 1984; Laland et al., 2011; Warner, 1988, 1990). There is even preliminary evidence for traditions in insects (Donaldson & Grether, 2007; Leadbeater & Chittka, 2007). Therefore, the current taxonomic spread of animal traditions does not support the idea that traditions necessarily require large brains. There is little reason to assume that socially transmitted traditions necessarily require complex social learning mechanisms (such as high‐fidelity copying), since formal theory finds that simpler mechanisms, such as local enhancement, can generate traditions (Franz & Matthews, 2010; Van der Post & Hogeweg, 2008). Furthermore, although group‐specific behaviors are suggestive of socially learned traditions, inference of social learning from observational studies and group‐specific behaviors is contentious in that ecological, genetic and asocial learning factors can rarely be ruled out as explanations of such behavioral variation. Moreover, behavioral variation is likely to result from multiple causes, which are highly correlated (Laland & Janik, 2006; Langergraber et al., 2011). Therefore, it remains unclear whether the existence of traditions and culture in animals across species is associated with especially high‐fidelity copying mechanisms and brain expansion. Some authors have suggested that an association exists between the number of traditions and brain size in animals (Whiten & van Schaik, 2007), but even this is contentious (Laland & Hoppitt, 2003; Laland et al., 2011).

The association of social learning and brain size appears have reached its peak in our own species. Modern humans have both the largest brain size within their order (Striedter, 2005) and a uniquely substantial dependence on culturally acquired behaviors for survival and reproduction (Tomasello, 1999; Whiten, Hinde, Laland, & Stringer, 2011). Humans are especially skilled social learners in that we are capable of copying at exceptionally high fidelity, even as children, when compared to nonhuman apes (Herrmann, Call, Hernandez‐Lloreda, Hare, & Tomasello, 2007; Tennie, Greve, Gretscher, & Call, 2010). For example, “two‐action” studies (Hopper, Flynn, Wood, & Whiten, 2010) have demonstrated that children learn to solve a task in the specific manner that they saw performed by a demonstrator. In transmission chains of 20 children, each method of solving the task was passed from one child to another by social learning with perfect fidelity. When chimpanzees were tested using the same method, the majority, but not all, of the chimpanzees along the transmission chain conformed to the method shown (Whiten et al., 2005). Human adults too show a high level of precision in copying experimental tasks, down to the specific digits used for each part of a task, whereas macaques (Macaca nemestrina) showed little evidence of social transmission in the same experiment (Custance, Prato‐Previde, Spiezio, Rigamonti, & Poli, 2006). However, in a relatively simple task involving a door which could either be lifted or swung open, the results of children and chimpanzees were far more similar, with near perfect transmission fidelity in the chimpanzees (Horner, Whiten, Flynn, & De Waal, 2006). Human children, unlike nonhuman primates, have been shown to copy specific details of the sequence of behavior even when these are causally irrelevant to the goal of the task (Lyons, Young, & Keil, 2007; Whiten, McGuigan, Marshall‐Pescini, & Hopper, 2009b), an effect which is pronounced in adults (McGuigan, Makinson, & Whiten, 2011). In contrast, chimpanzees seem rather to copy only aspects of the solving the task that are physically necessary (Horner & Whiten, 2005; Whiten et al., 2009a).

In addition to exceptionally high‐fidelity copying, humans are seemingly unique in possessing cumulative culture. Cumulative culture is the process by which learned traditions are not only transferred across generations, but increase in complexity, diversity, and efficiency, such that each generation inherits much of the accumulated knowledge of innumerable individuals on which to build further modification and improvement (Sterelny, 2011; Whiten, 2011). Cumulative culture is responsible for the majority of the learned human repertoire, particularly science and technology (Mesoudi, 2011; Tomasello, 1999), and has been instrumental in the evolution of our species’ unique reliance on culture for survival and reproduction. Unsurprisingly, cumulative culture in humans has been demonstrated experimentally, in for example, the steady improvement of paper plane flight distances in a laboratory transmission chain study (Caldwell & Millen, 2010). “Ratcheting” effects—in which a learned behavior is modified—have been clearly observed in humans only (Custance et al., 2006; Hopper et al., 2010). In contrast, chimpanzees showed no ability to learn cumulatively in an artificial “honey dipping” task. In the study, chimpanzees learned how to dip a stick to extract honey from the apparatus from a demonstrator. Subsequently, they observed a demonstrator manipulating the device in a previously unseen manner in order to access a higher quality reward. Despite numerous demonstrations, the chimpanzees did not learn the new and improved method (Marshall‐Pescini & Whiten, 2008).

Related posts:

The Brain Evolved to Guide Action Social Coordination: From Ants to Apes Neural Mechanisms of Communication Neurotransmission—Evolving Systems Neural Development in Invertebrates Nervous System Architecture in Vertebrates

Stay updated, free articles. Join our Telegram channel

Join