Encephalization has not been as well studied in invertebrates as in vertebrates. Some structures have, however, been thought to play equivalent roles to the ones that the forebrain plays in mammals and birds. Invertebrates often cited for their cognitive skills are the hymenoptera on the one hand and the octopus and cuttlefish on the other (see Cognition in Invertebrates). Hymenoptera have the most complex social behaviors of all insects. Octopus and cuttlefish are at the extremes of the habitat complexity distribution proposed by Hanlon and Messenger (1996, figure 3.9). The same intraclass logic we have applied earlier to birds and mammals can thus be applied to the groups that hymenoptera and octopus belong to, insects and cephalopods. In these classes, the mushroom bodies and vertical lobes, respectively, are the brain structures most often mentioned in studies of encephalization.
In insects, the mushroom bodies have long been seen as the higher centers that might be the substrate of cognition (see Strausfeld et al., 1998, and Farris, 2005, for reviews). They control sensory integration, learning, and memory, and, according to Farris (2005), are convergent equivalents of the mammalian cortex. Their crucial role in memory is evidenced by Drosophila mutants that lack both the vertical lobes of the mushroom bodies and long-term (but not short-term) memory (Pascual and Preat, 2001). The insect taxa (ants, honeybees, and wasps) that have evolved complex societies with division of labor, as well as altruistic reproduction and nest defense, have enlarged mushroom bodies (Howse, 1974; Gronenberg et al., 1996; Ehmer and Hoy, 2000). Diet might be as important as social life in determining insect mushroom body size. Mares et al. (2005) found that honeybees (Apis mellifera) do not have larger mushroom bodies than does the bumblebee Bombus impatiens, as one would have predicted from the much more complex social life of honeybees. B. impatiens is a dietary generalist, however, which raises the intriguing possibility that specialized species of Bombus might have smaller mushroom bodies than either B. impatiens or A. mellifera. Farris and Roberts (2005) compared 11 generalist and specialist scarab beetle species and found sharp differences in mushroom body size and structure associated with dietary differences. Generalist (e.g., phytophagous) beetles have larger and more convoluted mushroom bodies featuring double calyces, whereas specialist species (e.g., dung beetles) have smaller mushroom bodies with single calyces. Ontogenetic changes, both natural and experimentally manipulated, that make honeybees switch from larval care to the much more complex task (e.g., learning and dance communication of flower patches, swarming; Seeley and Burhman, 1999) of foraging outside the hive are accompanied by an increase in the Kenyon cells of the mushroom bodies (Withers et al., 1993). Similar results have also been reported for carpenter ants (Gronenberg et al., 1996)
Compared to other classes, the relative brain size of cephalopods is between that of fish and reptiles on the one hand and birds and mammals on the other (Packard, 1972). Within the 800 or so cephalopod species, there is a large degree of variation in learning performance, brain size, and vertical lobe size. The vertical lobe is the area of the cephalopod brain that Nixon and Young (2003) describe as the modulator for the systems that guide visual and tactile responses. Nixon and Young (2003) list the relative size of 14 brain areas in 63 species (see also Wirz, 1959). The data in their table 2.6 are expressed as fractions of total brain size. As Hanlon and Messenger (1996) point out, the two genera that are most often mentioned as intelligent, octopus and cuttlefish (Sepia), do not have the largest vertical lobes according to this fraction estimate. However, when we use the more usual technique of regressing either whole brain size or vertical lobe size against body size (in this case, mantle length, given for 49 of the species), the two species are 1.5 (octopus) and 2.5 (cuttlefish) standard deviations above the mean cephalopod regression line. The third cephalopod whose nervous system has been intensively studied, the squid Loligo, places around 2 standard deviations above the line.
Many studies have been conducted on associative learning in octopus by Young and his colleagues (Wells, 1966). Three features of avian and primate cognition, innovation, social learning, and improvement over successive learning reversals, have been described in the field by Norman (1999) and in the lab by Fiorito and Scotto (1992) and Mackintosh and
Mackintosh (1964). Octopus in Indonesia forage for complimentary fragments of coconut shells thrown by humans in shallow water, using them as portable dens (Norman, 1999). Octopus (Fiorito et al., 1998) can also solve the kind of innovative food-finding problem that passerines, but not doves, readily succeed at (Webster and Lefebvre, 2001; see however, Bouchard, 2002 for pigeons). Finally, octopus that observe a trained conspecific attack a white stimulus (instead of the normally preferred dark stimulus), will also attack the white stimulus when tested alone after the observation sessions (Fiorito and Scotto, 1992). Lesions of the vertical lobe, which are known to affect associative learning in octopus, also affect observational learning, but only over short time intervals (Fiorito and Chichery, 1995). The octopus vertical lobe seems to show evidence of convergent evolution with vertebrate learning mechanisms, with long-term potentiation of glutaminergic synaptic field potentials (Hochner et al., 2003). In cuttlefish, the vertical lobe also appears to be involved in learning. In particular, Dickel et al. (2001) show a striking similarity between ontogenetic increases in the relative size of the vertical lobe (but not of other areas) and improvements in learning.
Was this article helpful?