For a long time people have speculated about what makes some human brains “brainier” than others; in the present target article Jung & Haier (J&H) go a bold step beyond speculation. As I think the authors do too, I have a hard time believing that we will find one singular “structural basis of intelligence”: Considering the multitude of characteristics that, together, make up intelligence, I would rather expect several collaborating (and competing?) network nodes, with strengths in one (or more) explaining individual performance advantages. Therefore, the proposed model is a step in the right direction, allowing for different emphases within its framework. Considering the numerous studies that contributed to this review, it seems to me that intelligence research using neuroimaging methods has come to a crossroad: Based on the available data, a broad pattern seems to emerge, and now more sophisticated approaches seem warranted, including an exact characterization of the cognitive functions that are under scrutiny.
In that respect, I tend to agree with Thorndike (Reference Thorndike1921) that intelligence is “that which intelligence tests measure.” Correlates of global parameters are likely to give us only so much insight, as the various cognitive functions underlying performance in such tests will likely preclude finding distinct nodes in the to-be-suspected networks. Instead, the correlates of defined and more “pure” cognitive functions as assessed in different ways would seem to be more promising in defining distinct parts within the complex network that must be suspected to underlie “intelligence.” Neuroimaging studies of “schizophrenia” may serve as a reminder: there is practically no brain region that has not been implicated in the neurobiology underlying schizophrenia (Honea et al. Reference Honea, Crow, Passingham and Mackay2005; McCarley et al. Reference McCarley, Wible, Frumin, Hirayasu, Levitt, Fischer and Shenton1999). The explanation for this (apart from different methodological approaches) is likely that schizophrenia, in the neurobiological sense, is not one disease but may rather be a common final path of several disorders, or the result of a variable combination of disturbances caused by (at least) “two hits” (Rapoport et al. Reference Rapoport, Addington, Frangou and Psych2005). At the least, the clinical variability between subjects severely decreases the chances to find similarities, as the consequently increased variability will make the detection of subtle abnormalities much less likely.
The same may and will be true for finding the neuroanatomical substrate of intelligence, broadly defined: no one region can be expected to underlie such a complex cognitive function, and the challenge for the next years will be to more closely define the different nodes while not forgetting their contribution to the underlying network. Along these lines, we may find that the differences between the nodes tell us more than the similarities.
The omission of results from complementary imaging methods, although necessary for this format, somewhat limits the scope of the model: Processing speed and neuronal regions driving each other on the millisecond time scale are likely important aspects contributing to the efficiency of fast decision-making processes. These are more accurately described using methods such as magnetoencephalography or electroencephalography, as done recently (Thatcher et al. Reference Thatcher, North and Biver2007). The lower spatial resolution of these methods is somewhat compensated by the very high time resolution, thereby almost ideally complementing dynamic MRI methods. As J&H pointed out, multi-modal imaging approaches are likely to have the highest yield when assessing such a complex interplay.
There is one point where I would like to urge caution: I would be very careful in drawing inferences on normal neuronal organization from studies on children with severe epilepsy. The pediatric brain has been shown to be highly adaptable when overcoming early neuronal insults (Krägeloh-Mann Reference Krägeloh-Mann2004), and our own group could show the enormous adaptive potential with regard to language (Staudt et al. Reference Staudt, Lidzba, Grodd, Wildgruber, Erb and Krägeloh-Mann2002), as well as the limitations of this potential with regard to the perception of biological motion (Pavlova et al. Reference Pavlova, Sokolov, Birbaumer and Krägeloh-Mann2006) and visuospatial functions (Lidzba et al. Reference Lidzba, Staudt, Wilke and Krägeloh-Mann2006). Considering this plasticity (which has also been shown in the context of epilepsy; Gleissner et al. Reference Gleissner, Sassen, Schramm, Elger and Helmstaedter2005; Yuan et al. Reference Yuan, Szaflarski, Schmithorst, Schapiro, Byars, Strawsburg and Holland2006) it is difficult to establish a cause-and-effect chain: If subjects with severe epilepsy profit when the epileptogenic region is removed, it could be due to the interfering effect of this region on other regions (which are consequently released from the interference and therefore function better). Alternatively, the interfering region itself could be an important network player in the healthy state and might have lost this functionality in favor of another region that has since assumed this role. Or, the epileptogenic region is still a malfunctioning part of the network and only its removal prompts a reorganization, or… Without knowing the sequence of events, the effect of such long-term interference in a more-adaptive neuronal system (i.e., the pediatric brain) is difficult to impossible to predict.
This is not to discount the importance of lesion studies, however. The findings from acute brain lesions in adults, as in the case of the historic missile wound studies, may well be used to supplement and inform the results obtained from healthy individuals (or special patient populations), and may contribute important information (e.g., it was mainly the case of the famous “Monsieur Tan” that lead Paul Broca to formulate and articulate his theory over specific language centers in the brain; cf. Broca Reference Broca1861).
Winston Churchill is said to have coined the famous “a riddle, wrapped in a mystery” quote (concerning Russia; supposedly he even added “inside an enigma”). Right now, the field of intelligence research may have moved towards stripping away parts of the mystery, but the riddle inside is still unsolved. Overall, I believe this to be an important contribution, hopefully sparking interesting and fruitful discussions within the intelligence research community.
