doi: 10.3819/CCBR.2018.130010 Volume 13, 2018 Comparative Cognition & Behavior Reviews

Where the Standard Approach in Comparative Neuroscience Fails and Where It Works: General Intelligence and Brain Asymmetries Davide Serpico Elisa Frasnelli Department of Philosophy and Educational Sciences School of Life Sciences University of Turin University of Lincoln Department of Antiquity, Philosophy, and History University of Genoa

Although brain size and the concept of intelligence have been extensively used in comparative neuroscience to study cognition and its evolution, such coarse-grained traits may not be informative enough about important aspects of neurocognitive systems. By taking into account the different evolutionary trajectories and the selection pressures on neurophysiology across species, Logan and colleagues suggest that the cognitive abilities of an organism should be investigated by considering the fine-grained and species-specific phenotypic traits that characterize it. In such a way, we would avoid adopting human-oriented, coarse-grained traits, typical of the standard approach in cognitive neuroscience. We argue that this standard approach can fail in some cases, but can, however, work in others, by discussing two major topics in contemporary neuroscience as examples: general intelligence and brain asymmetries.

Keywords: general intelligence, brain asymmetries, comparative neuroscience

What do coarse-grained and taxon-neutral traits, of a given species. This would help to avoid reification, such as brain size and intelligence, tell us about neuro- defined by the authors as mistaking an operationalized cognitive systems? What characteristics should a behav- target of measurement with a real, causally meaningful ioral proxy have to allow us to properly compare cogni- object. tive abilities across different taxa? In their article Although we agree with this conclusion, there are “Beyond Brain Size: Uncovering the Neural Correlates two aspects analyzed by Logan and colleagues that we of Behavioral and Cognitive Specialization,” Logan think may deserve further consideration. The first is the and colleagues address these questions by reviewing common tendency in comparative research to assess the literature on the relationship between brain size human-oriented phenotypic traits in nonhuman species; and cognition. In the light of empirical and theoretical the second is the tendency, which is also widespread, to considerations, the authors suggest that, to understand adopt coarse-grained traits, such as brain size, instead how brain and cognition evolve, comparative biologists of fine-grained and more informative traits. should focus on fine-grained, taxon-specific phenotypic Let us start by considering the first one. A general traits within the relevant ecological and adaptive context premise of Logan and colleagues’ article is that

ISSN: 1911-4745 95 96 Serpico and Frasnelli comparative behavioral research is often characterized in neuroscience: general intelligence and brain asymme- by a sort of “anthropocentric perspective.” However, as tries. These two examples elucidate where the coarse- the authors argue, human behavioral traits—and the grained and human-oriented comparative approach fails related neural bases—are not necessarily shared by other and where instead it can be appropriate and helpful. species. This argument can be understood as a criticism The authors’ argument against anthropocentric against the so-called Great Chain of Being, a secular comparative research sounds effective in the case of ideology that assumes a natural hierarchy of organisms general intelligence and its putative underlying mecha- with humankind at the top and then, successively, lower nism, namely, the g factor. General intelligence repre- animals from primates to bacteria (see, e.g., Sternberg, sents a psychological trait assessed by psychometric IQ 2017). Logan and colleagues clarify how misleading this tests that generally recruit linguistic, mathematical, logi- interpretation of the tree of life is by highlighting the cal, and spatial abilities. The g factor, instead, is often importance of the ecological and adaptive context of understood as a domain-general cognitive mechanism a species: The ability of the organisms to achieve their accounting for both individuals’ performance in tests goals should be evaluated within the range of challenges (i.e., the intelligent behavior) and individual intellectual they face within their natural environment. Therefore, differences within populations (see Burkart, Schubiger, phenotypic traits should be identified by accounting for & van Schaik, 2017; Serpico, 2017; Sternberg & Grigore- what is actually “meaningful” for a given species rather nko, 2002). General intelligence is the subject of heated than what is meaningful for humans. controversy in regards to two aspects tackled by Logan The second central point of the target article and colleagues with respect to brain size, that is, the concerns the use of general, coarse-grained phenotypic granularity problem and the anthropocentric perspec- traits and proxy measures to study cognition. Brain tive widespread in comparative research. size represents, in the authors’ view, a noninforma- First, exactly like brain size, general intelligence tive measure of neurocognitive systems. Indeed, such represents a coarse-grained evaluation of a cognitive a broad measure cannot disentangle from one another system, regardless of any detail about its structural important aspects such as the dimensions of specific and functional composition. However, many schol- brain areas, the neuron density, and the connectivity ars have argued that intelligence, rather than reflect- patterns between neurons, which may be more informa- ing a single neurocognitive mechanism, is composed tive than brain size about the properties of a neurocogni- of several distinct and autonomous—but not necessar- tive system. Hence, it is not surprising that correlational ily independent—cognitive­ mechanisms (see Hamp- analyses of brain size tend to produce spurious associ- shire, Highfield, Parkin, & Owen, 2012; Serpico, 2017; ations at both the intraspecific and interspecific levels Sternberg & Grigorenko, 2002; Van der Maas et al., (reviewed by Logan et al.). 2006). Logan and colleagues’ argument about brain We recognize that these criticisms could rule out size is in line with the view that general intelligence, misleading assumptions in comparative behavioral like other high-level psychological constructs, does not research. However, although the authors’ arguments represent an informative measure of neurocognitive seem to be applicable to most research about high-level systems (see Craver, 2009, who discussed the problem of psychological constructs, we believe they may miss the subtyping complex psychological traits into lower-level mark in respect with structural features of nervous characteristics). systems and the related behavioral manifestations. To Second, we think that general intelligence is exactly explain our concerns about the approach proposed by that sort of human-oriented trait that Logan and Logan and colleagues, we use as examples two topics colleagues criticize as erroneously generalized to other species. Although the g factor likely plays a role, if any, in human cognition only (indeed, g accounts for Author Note: Davide Serpico, Department of Philosophy and Ed- the population variance in a battery of IQ tests), many ucational Sciences, University of Turin, Via Verdi, 8, Turin, Italy. authors have tried to assess its role in other species. For instance, in their review about the evolution of general Correspondence concerning this article should be addressed to intelligence, Burkart and colleagues (2017) took the g Davide Serpico at [email protected]. factor as a domain-general neurocognitive mechanism Acknowledgments: We thank Anna Wilkinson for inviting us to shared by humans, primates, birds, and rodents. The contribute a commentary on this paper. arguments provided by Logan and colleagues against

COMPARATIVE COGNITION & BEHAVIOR REVIEWS General Intelligence and Brain Asymmetries 97 the anthropocentric viewpoint in comparative studies us to uncover the evolution of behavioral and neural seem to us well suited to describe, and possibly to rule asymmetries, and their developmental mechanisms out, this sort of reification of general intelligence. as well. By contrast, the arguments provided by the authors The take-home message is that the approach seem to be less suited to be applied to structural prop- proposed by Logan and colleagues might be promis- erties of nervous systems. This is the case of brain ing for behavioral traits reflecting higher-level cogni- and behavioral asymmetries, where a sort of human- tive aspects such as general intelligence but be poten- oriented, coarse-grained approach seems to be promis- tially fragile for behavioral traits reflecting lower-level ing and successful. structural properties of the nervous system. As we have Traditionally, lateralization (i.e., the different func- argued, the disagreement about general intelligence tional specialization of the right and left sides of the relies on its generality and on its dubious value in char- nervous system) was considered a uniquely human char- acterizing the neurocognitive system of nonhuman acteristic, related to handedness and linguistic func- species. By contrast, what we know about lateralization tions (McManus, 1999). Over the past decades, thanks points at the (partial) adequacy of a human-oriented, to research conducted in comparative psychology, neuro- coarse-grained approach in comparative research. science, and developmental biology, we have realized We do not mean to assume an anthropocentric view- that lateralization is a widespread phenomenon among point. Nevertheless, there are good reasons to think that vertebrates (Rogers, Vallortigara, & Andrew, 2013). some structural aspects of nervous systems—and their Moreover, there is emerging evidence of behavioral and behavioral correlates—are highly conserved across brain asymmetries in invertebrates, suggesting that later- different taxa, regardless of their granularity. Whether alization is a feature of simpler as well as more complex the conservation of this kind of traits is due to natu- neurocognitive systems (reviewed by ­Frasnelli, 2013). ral selection, to the modularity of the related genetic In most vertebrates, for example, the right hemi- processes, or to some sort of developmental constraint sphere is involved in responding to unexpected and novel in brain morphology is an empirical question that future stimuli (e.g., predators) and in interacting with conspe- research must address. cifics, and the left hemisphere is specialized in less complex and repetitive tasks (e.g., behavioral routines). References Lateralization also characterizes learning and memory both in vertebrates (e.g., in birds; see Andrew, 1999; Andrew, R. J. (1999). The differential roles of Cipolla-Neto, Horn, & McCabe, 1982; Clayton, 1993) right and left sides of the brain in memory and in invertebrates. For instance, honeybees present formation. Behavioural Brain Research, 98, 289–295. an asymmetrical use of the right and left antennae in doi:10.1016/S0166-4328(98)00095-3 learning and recalling olfactory memories: recall of short-term memory is implemented by the right side, Burkart, J. M., Schubiger, M. N., & van Schaik, C. whereas recall of long-term memory is possible through P. (2017). The evolution of general intelligence. the left side (see Letzkus et al., 2006; Rogers & Vallorti- Behavioral and Brain Sciences, 40, 1–24. gara, 2008; for a review, see Frasnelli et al., 2014). This doi:10.1017/S0140525X16000959 suggests that the shifts between the two sides of the brain from recently acquired information to more inte- Cipolla-Neto, J., Horn, G., & McCabe, B. J. (1982). grated and complete long-term records might constitute Hemispheric asymmetry and imprinting: The a considerable advantage for both arthropods and verte- effect of sequential lesions to the hyperstriatum brates. Thus, mechanisms controlling such shifts may ventrale. Experimental Brain Research, 48, 22–27. have evolved, perhaps independently, in both phyla (see doi:10.1007/BF00239569 Frasnelli, Vallortigara, & Rogers, 2012). In sum, despite differences between the nervous Clayton, N. (1993). Lateralization and unilateral systems of humans and other animals, there is good transfer of spatial memory in marsh tits. Journal of evidence that lateralization represents a structural Comparative Physiology A: Neuroethology, Sensory, feature implemented by similar mechanisms across Neural, and Behavioral Physiology, 171, 799–806. different species and taxa (see Rogers et al., 2013). Thus, doi:10.1007/BF00213076 thanks to these similarities, animal models have allowed

VOLUME 13, 2018 98 Serpico and Frasnelli

Craver, C. (2009). Mechanisms and natural kinds. Rogers, L. J., & Vallortigara, G. (2008). From antenna Philosophical Psychology, 22, 575–594. to antenna: Lateral shift of olfactory memory doi:10.1080/09515080903238930 recall by honeybees. PLoS One, 3(6), e2340. doi:10.1371/journal.pone.0002340 Frasnelli, E. (2013). Brain and behavioral lateralization in invertebrates. Frontiers in Psychology, 4, 939. Rogers, L. J., Vallortigara, G., & Andrew, R. J. doi:10.3389/fpsyg.2013.00939 (2013). Divided brains: The biology and behaviour of brain asymmetries. New York, NY: Cambridge Frasnelli, E., Haase, A., Rigosi, E., Anfora, G., Rogers, University Press. L. J., & Vallortigara, G. (2014). The bee as a model to investigate brain and behavioural asymmetries. Serpico, D. (2018). What kind of kind is intelligence? Insects, 5, 120–138. Philosophical Psychology, 31(2), 232–252. doi:10.3390/insects5010120 doi:10.1080/09515089.2017.1401706

Frasnelli, E., Vallortigara, G., & Rogers, L. J. (2012). Sternberg, R. J. (2017). It’s time to move beyond Left–right asymmetries of behaviour and the “Great Chain of Being.” Behavioral nervous system in invertebrates. Neuroscience and Brain Sciences, 40, 45–46. & Biobehavioral Reviews, 36, 1273–1291. doi:10.1017/S0140525X16001783 doi:10.1016/j.neubiorev.2012.02.006 Sternberg, R. J., & Grigorenko, E. (Eds.). (2002). The Hampshire, A., Highfield, R. R., Parkin, B. L., general factor of intelligence. How general is it? & Owen, A. M. (2012). Fractionating human London, England: Psychology Press. intelligence. Neuron, 76, 1225–1237. doi:10.1016/j.neuron.2012.06.022 Van Der Maas, H. L., Dolan, C. V., Grasman, R. P., Wicherts, J. M., Huizenga, H. M., & Raijmakers, Letzkus, P., Ribi, W. A., Wood, J. T., Zhu, H., Zhang, M. E. (2006). A dynamical model of general S. W., & Srinivasan, M. V. (2006). Lateralization of intelligence: The positive manifold of intelligence olfaction in the honeybee Apis mellifera. Current by mutualism. Psychological Review, 113, 842–861. Biology, 16, 1471–1476. doi:10.1016/j.cub.2006.05.060 doi:10.1037/0033-295X.113.4.842

McManus, I. C. (1999). Handedness, cerebral lateralization, and the evolution of language. In M. C. Corballis & S. E. G. Lea (Eds.), The descent of mind: Psychological perspective on hominid evolution (pp. 194–217). New York, NY: Oxford University Press. doi:10.1093/acprof:oso/9780192632593.003.0011

COMPARATIVE COGNITION & BEHAVIOR REVIEWS