The performance of chronic schizophrenic probands (n=21), their first-degree schizotypal (22) and non-schizotypal (19) relatives, and normal controls (24), was measured in two associative learning paradigms, latent inhibition and the Kamin blocking effect. These paradigms assess the effects on learning of initial exposure to other learning contingencies. The normal subjects showed latent inhibition (retarded learning of an association between a burst of white noise and a visually displayed counter increment, if the subject had first been pre-exposed to the white noise without any other consequence) and Kamin blocking (retarded learning of an association between two visual stimuli, if the conditioned stimulus was presented simultaneously with a second, already conditioned stimulus). The schizophrenic probands and both the schizotypal and non-schizotypal relatives were severely impaired in basic associative learning, performing much worse than the normal subjects in the control conditions (i.e. those lacking stimulus pre-exposure of any kind) of both the latent inhibition and the Kamin paradigms and also showed a loss of the normal latent inhibition and Kamin blocking effects. The performance of the three clinically defined groups was statisitically indistinguishable. These findings contrast with previous reports of the performance of normal subjects classified as schizotypal by questionnaire, who are not impaired in basic associative learning, and are particularly fast to learn after stimulus pre-exposure. The results question the assumption that high schizotypy, as assessed by questionnaire, is like schizotypy in schizophrenic kin. The severe impairment in basic associative learning in schizophrenic patients and their kin warrants further investigation.