Retrieval inhibition and interference in cognitive arithmetic

Campbell and Clark (1989) presented evidence that answers to simple multiplication problems are temporarily inhibited after they have been retrieved. According to the network-interference model, inhibition of incorrect, associatively related responses should facilitate retrieval. This prediction was confirmed by new analyses of the Campbell and Clark data: When successive multiplication problems were related (i.e., 'neighbours' in the same time-tables), second-problem response times were faster and accuracy was higher than when successive problems were. unrelated. In another expeirment, subjects received pairs of multiplication problems in which the second problem was presented 500 ms after the response to the first. Control trials involved a subtraction probe followed by a multiplication target. Experimental trials confirmed the predicted benefits of inhibition of related answers, and controls showed that experimental effects were not due to a simple effect of operand repetition. Campbell et Clark (1989) ont presente des evidences qui expliquent comment des problemes de multiplication simple sont temporairement inhibes apres avoir ete recouvres. Selon le modele d'interference de reseaux, l'inhibition de reponses incorrectes associatvement reliees devrait faciliter le recouvrement. Cette prediction a ete confirmee par de nouvelles analyses des donnees de Campbell et Clark: lorsque des problemes de multiplication successifs etaient relies ('voisins' dans la meme table de multiplication), les temps de reponses du deuxieme probleme etaient plus rapides et la precision plus grande que dans les cas ou les problemes successifs n'etaient pas relies. Dans une autre experience, les sujets recevaient des paires de problemes de multiplication dans lesquels le deuxieme probleme etait presente 500 ms apres la reponse au premier. Les essais de controle impliquait une sonde de soustraction suivi d'une cible de multiplication. Les essais experimentaux ont confirme les effets benefiques predits de l'inhibition des reponses reliees et les controles ont montre que les effets experimentaux n'etaient pas dus a un simple effet de repetition operante.