Yoshida et al (Reference Yoshida, Marks and Craggs1999) demonstrate the need for further longitudinal observational studies of the mother-infant relationship when mothers have schizophrenia that began during pregnancy. It is important to disentangle the impact on the infant of the mother-infant relationship and the environment she provides for her child from those genetic factors which place the infant at risk.

Neurobiological features are suggested by reports that: variations in maternal care in the rat promote hippocampal synaptogenesis and spatial learning and memory through systems known to mediate experience-dependent neural development (Reference Liu, Diorio and FrancisLiu et al, 2000); schizophrenia is a disorder of developmentally reduced synaptic connectivity (Reference McGlashan and HoffmanMcGlashan & Hoffman, 2000); and selective deficits in early-stage sensory processing in schizophrenia are due to a failure to support the entrainment of intrinsic gamma-frequency oscillations (30-50 Hz or broader, centred on 40 Hz) (Reference Friedman and CoatsFriedman & Coats, 2000) involved in processes associated with encoding into sensory memory both at the cellular level (synaptic potentiation) and at the cognitive level (Reference Haenschel, Baldeweg and CroftHaenschel et al, 2000).

This hypothesis is supported by short-term laboratory experience demonstrating that adult female speech production is sufficient to influence infant speech production occurring in the silent intervals between the adult vocalisations of the order of 3 seconds. This is linked with increased coherence of electroencephalograph gamma-band activity associated with the execution of more complex tasks (Reference Friedman and CoatsFriedman & Coats, 2000); language discrimination by human newborns may be influenced by hearing rhythmic aspects of speech while in the womb, a period in development during which exposure may have a more profound impact on the organisation of the brain than does learning after birth (Reference Ramus, Hauser and MillerRamus et al, 2000).

These findings prompt the possibility of prevention of neurocognitive defects (at least those of a sensory and perceptual nature) by establishing effective cortical oscillations, starting during pregnancy as suggested by Yoshida et al.