OBJECTIVESThe appearance of 'twin language' has been highlighted as a reason for delayed language development in twins. 'Twin language' is a unique language understandable only within the pair, and not by their mother or others. The purpose of this study was to examine and clarify the factors affecting the appearance of 'twin language'.

METHODSA mailed questionnaires survey was conducted in 2733 mothers of twins. Of them, 1395 mothers returned the questionnaires. The core questionnaire asked for data on birth weight, age at first spoken word, whether the twins were as alike as two peas in a pod, household members and non-verbal play. Logistic regression analysis was used in this study.

RESULTSOut of the 1395 pairs included in this analysis, 598 pairs (42.9%) showed the appearance of a 'twin language'. When the 598 pairs were divided by whether the twin pair was exactly alike or not, there were 112 opposite sex pairs, 105 not-alike male pairs, 106 not-alike female pairs, 129 exactly alike male pairs and 140 exactly alike female pairs. Namely, 38.4% of the opposite sex pairs, 40.4% of the not-alike male pairs, 39.3% of the not-alike female pairs, 47.6% of the exactly alike male pairs and 48.4% in the exactly alike female pairs had a twin language. By multivariate logistic regression analysis controlling for twins' age, it was found that 'twin language' was significantly more frequent in exactly alike twin pairs, pairs with non-verbal play, and pairs with fewer older siblings.

CONCLUSIONSThese findings suggest that three factors (whether the twin pair is exactly alike or not, older siblings, non-verbal play) affect the appearance of 'twin language'.