Crusio, Wim E. (1997 In Press) Neuropsychological inference using a microphrenological approach does not need a locality assumption. Behavioral and Brain Sciences.

Neuropsychological inference using a microphrenological approach does not need a locality assumption

Wim E. Crusio

Electronic mail: crusio@citi2.fr

Farah's (1994) target article provides an admirable overview of the problems connected with the use of the locality assumption, one that more or less equalizes the function of a lesioned structure with the defects exhibited by the damaged brain and is almost always invoked to interpret the results of lesion studies. In an elegant way, Farah provides evidence that this reasoning may lead to false conclusions. Besides convincing, her treatment is also constructive, in that she provides alternative hypotheses that may explain the data.

One remarkable feature of the target article (including most of the commentaries) and, indeed, of much of neuropsychology, is the comparatively sparse use of information from other branches of neuroscience. This is all the more striking because I think that neuroanatomical and neurophysiological data provide massive support for Farah's thesis. For example, the mammalian hippocampus is a very distinct structure anatomically and might easily give rise to a modular interpretation of the brain. Yet anatomical and physiological evidence clearly indicates the presence of reciprocal connections with, for example, the entorhinal cortex. Accordingly these regions interact (Jones 1993), the physiological state of the one modulating that of the other1. These data make it impossible to assume that the functioning of the hippocampus would remain unchanged after entorhinal damage or vice versa.

Although Farah shows considerable creativity in proposing alternative explanations for a number of lesion-induced defects, no attempt is made to devise research strategies that would not rest on the locality assumption. In the field of neurobehavioral genetics such an approach already appears to exist: using genetic methods exploiting naturally occurring individual differences as a tool for understanding brain function. No brain is like another2 and every individual behaves differently. The assumption that there is a link between the variability of the brain and individual talents and propensities appears quite plausible. This approach differs from the usual one in neuropsychology in two important aspects. First, no subjects are studied that, by accident or by design, have damaged brains. Rather, all subjects will fall within the range of normal, nonpathological variation. Second, instead of comparing a damaged group with normal controls, we study a whole range of subjects and try to correlate variation at the behavioral level with that at the neuronal level. This strategy is reminiscent of the phrenological approach propagated by Franz Josef Gall (1743-1826); Lipp has coined the name "microphrenology" for it (Lipp et al. 1989). It appears that, as long as variation in one neuronal structure is independent of that in another, there will be no need for a locality assumption to interpret results of experiments carried out along these lines. (Thus, contrary to Gall's own strong support for locality; cf. van Gelder 1994). In combination with methods from the field of behavior genetics (Crusio 1992), this strategy yields a very powerful approach. For example, to "magnify" individual differences, we may study animals from different inbred strains and look for correlations between the means obtained for different variables (see Crusio et al. 1993 and references therein for some illustrative examples). Alternatively, genetic correlations may be used to help clarify brain-behavior relationships (Crusio 1993)3.

With the advent of noninvasive brain imaging methods such as MRI (magnetic resonance imaging) and PET (positron emission tomography), this approach is becoming increasingly feasible for use with human beings and some interesting results are already being obtained (e.g., Squire et al. 1992; see also Posner's commentary). Some of these techniques might be fruitfully applied to specific problems mentioned by Farah. For example, if we subjected a number of healthy volunteers to some test involving their knowledge of living and nonliving things and simultaneously assessed their brain activity with PET, we would expect to see a selective activation of the temporal lobe. By appropriately manipulating test-items, we might subsequently detect whether such changes correlate with the living/nonliving dichotomy proposed by Warrington and Shallice (1984) or with the visual/functional dichotomy proposed by Farah. In Warrington and Shallice's model, using knowledge of living things and knowledge of nonliving things would activate different regions. Farah's model would predict that using visual information as opposed to other functional information preferentially activates different regions. If the locality assumption were correct, these differential activations would be exclusive, i.e. only one region would be activated according to the property of the information needed. If the locality assumption were false, the activation would be expected in more than one region at a time but, according to the type of semantic memory implied, one region would be more strongly activated than the others.

Kosslyn and Intriligator (1992) have warned us of the perils of "sitting on a one-legged stool" and advised neuropsychologists to use a three-legged one: like Farah, they advocate using behavioral data, computational modeling and neural constraints to formulate and test theories. I suggest that a four-legged stool will be even more stout: let us add the study of individual differences in brain and behavior to the neuropsychological chair.

Acknowledgments

The preparation of this commentary benefitted from support by the CNRS (URA 1294), UFR Biomédicale (Université Paris V René Descartes), DRED, and the Fondation pour la Recherche Médicale.

Notes

1. Another example, the functional overlap between the hippocampus and the superior colliculus, has been discussed in detail by Foreman and Stevens (1987). With "functional overlap", these authors meant that "effective functioning of structure A is dependent on information handling in structure B, or vice versa, that the involvement of both structures is necessary for efficient performance of a particular behaviour" (p. 102).
2. This heritable variation of the brain is another aspect that neuropsychologists (and many neuroscientists alike) tend to ignore, most likely to their own peril. For example, Donovick et al. (1981) and Fanelli et al. (1983) reported widely divergent behavioral effects of septal lesions in mice, depending on which particular inbred strain was being used.
3. The defects shown by neurological mutants can be regarded as the neurobehavioral-genetic parallel of lesions. Such animals may solve another practical problem inherent in many lesion studies: almost no lesion is limited to only one single brain structure, not even in controlled animal studies. Animals with an intact dorsal commissure of the fornix but without a corpus callosum are almost impossible to obtain by means of a surgical intervention. However, appropriate crosses between the inbred mouse strains BALB/cWah1 and 129/ReJ, for example, may produce just such animals (Wahlsten and Schalomon, in press). Still, neurological mutations are seldomly limited in their effects to one brain structure only, so this approach constitutes more a (welcome) addition rather than a real alternative to the lesion technique.

References

Barber, R. P., Vaughn, J. E., Wimer, R. E. & Wimer, C. C. (1974) Genetically-associated variations in the distribution of dentate granule cell synapses upon the pyramidal cell dendrites in mouse hippocampus. Journal of Comparative Neurology 156:417-434.

Crusio, W. E. (1992) Quantitative Genetics. In: Techniques for the Genetic Analysis of Brain and Behavior: Focus on the Mouse. Techniques in the Behavioral and Neural Sciences, Volume 8, eds. D. Goldowitz, D. Wahlsten & R. Wimer. Elsevier.

Crusio, W. E. (1993) Bi- and multivariate analyses of diallel crosses: A tool for the genetic dissection of neurobehavioral phenotypes. Behavior Genetics 23:59-67.

Crusio, W. E., Schwegler, H. & Brust, I. (1993) Covariations between hippocampal mossy fibres and working and reference memory in spatial and non-spatial radial maze tasks in mice. European Journal of Neuroscience 5:1413-1420.

Donovick, P. J., Burright, R. G., Fanelli, R. J. & Engellenner, W. J. (1981) Septal lesions and avoidance behavior: Genetic, neurochemical and behavioral considerations. Physiology and Behavior 26:495-507.

Fanelli, R. J., Burright, R. J. & Donovick, P. J. (1983) A multivariate approach to the analysis of genetic and septal lesion effects on maze performance in mice. Behavioral Neuroscience 97:354-369.

Farah, M. J. (1994) Neuropsychological inference with an interactive brain: A critique of the "locality" assumption. Behavioral and Brain Sciences 17:43-104.

Foreman, N. & Stevens, R. (1987) Relationships between the superior colliculus and hippocampus: Neural and behavioral considerations. Behavioral and Brain Sciences 17:101-152.

Jones, R. S. G. (1993) Entorhinal-hippocampal connections: a speculative view of their function. Trends in Neurosciences 16:58-64.

Kosslyn, S. M. & Intriligator, J. M. (1992) Is cognitive neuropsychology possible? The perils of sitting on a one-legged stool. Journal of Cognitive Neuroscience 4:96-106.

Lipp, H.-P., Schwegler, H., Crusio, W. E., Wolfer, D., Leisinger-Trigona, M.-C., Heimrich, B. & Driscoll, P. (1989) Using genetically-defined rodent strains for the identification of hippocampal traits relevant for two-way avoidance learning: A non-invasive approach. Experientia 45:845-859.

Posner, M. I. (1994) Local and distributed processes in attentional orienting. Behavioral and Brain Sciences 17:78-79.

Squire, L. R., Ojemann, J. G., Miezin, F.M., Petersen, S. E., Videen, T. O. & Raichle, M. E. (1992) Activation of the hippocampus in normal humans: A functional anatomical study of memory. Proceedings of the National Academy of Sciences, USA 89:1837-1841.

van Gelder, T. (1994) Playing Flourens to Fodor's Gall. Behavioral and Brain Sciences 17:84.

Wahlsten, D. & Schalomon, P. M. (1994) A new hybrid mouse model for agenesis of the corpus callosum. Behavioural Brain Research (in press).