The reduced (48%) incidence of schizophrenia in monozygotic twins when the co-twin develops schizophrenia[1] has long remained a puzzle and a challenge.  Attempts have been made to explain the phenomenon by the use of the concept of incomplete penetrance and a possible postzygotic, mitotic crossover that could account for the discordance.  A study conducted by Polymeropoulos et al[2] claims “no genetic marker discordances were identified between the co-twins, still leaving the possibility that a tissue-specific, mitotic crossover may have occurred in one of the discordant twins.”

A more promising explanation is suggested by experimental observations of neuronal dendritic development in the sea slugs Aplesias (similar development also is known to occur in the daphnid and the fruit fly).  In identical twins of such slugs, their giant neurons are located initially in spatial configurations identical to each other in their respective ganglia.  Furthermore, it has been observed that this identical initial configuration of the developing dendrites in twin cohorts of Aplesias soon diverges into different franctal patterns (ie, similar in branching ratios on any scale but not identical in spatial coordinates). [3] Such divergence can only be due to random, environmental, perturbing influences after birth.  The final outcome of the fractal deployment of dendrites within the boundaries and constraints of the initial genetic algorithm thus can be considered to be ther result of the sum of such random influences in each twin.

It is plausible to assume that this phenomenon occurs in the developing neuronal systems of all higher organisms, including humans.  This implies a different final dendritic structural arrangement for each human co-twin and hence a different functional reality, even though they began with identical genetic programming.

If such randomized dendritic deployment expresses itself as a sufficiently faulty functional abnormality, schizophrenia occurs in at least one co-twin.  The probability is that the other co-twin will be protected approximately 50% of the time from also expressing a full-blown, clinically identifiable schizophrenia.  Nevertheless, as any experienced clinician will attest, subtle signs of peculiar thought and behavioral patterns will still be found in the other co-twin (ie, a kind of “background noise” reflecting their common genetic origin).

It is our belief that this phenomenon may shed some light on the influences of detailed dendritic structure on brain function.  The gross dendritic structure is no doubt genetically programmed.  The peculiar thought and behavioral patterns identifiable in both twins reflect a common heritage close to the threshold of instability and, hence, schizophrenia.  The “toss-of-the-coin” associated with the occurrence of lack of occurrence of schizophrenia in nearly 50% of the co-twins of twins with schizophrenia proves that twins reflects a final level of sensitivity to “faulty wiring” of genetic origin.

Structural defects in the brains of schizophrenics, shown by magnetic resonance imaging, have been well established.  In a study by Weinberger et al,[4] the affected twin had a smaller anterior pes hippocampus and less regional cerebral blood flow in the dorsolateral prefrontal cortex during performance of a particular task (Wisconsin card sorting test).  The detected abnormalities on magnetic resonance imaging may be the macroscopic consequence of such fractal divergence of the dendrite patterns as they unfold differently (similarly but not identically) for each of the co-twins as I have outlined.