La recherche sur l'apprentissage opérant chez le nourrisson : un parallèle avec l'état général de la question chez l'humain

Abstract

Summary : Operant conditioning in infants : a parallel wiih the general situation in humans. Research on infants' operant behaviours has followed the methodological model devised to study behaviours of other species, in dissimilar contexts. The need to identify discrete and repetitive responses reinforced by discrete stimuli in order to meet wiih the demands of a response rate analysis, greatly limits the range of possible target-behaviours. Moreover, the control variables elaborated for the human contexts have little in common with those used with animais. In the first case, the variables are part of an ecological reinforcement paradigm, appropriate for exploration and problem solving settings. The temporary and differential reinforcing qualities of ecological stimuli have favoured the study of infants' perceptual and cognitive processes, without actually increasing our knowledge about learning processes per se. Therefore, it appears preferable to depart from the sole dependency on response rate changes as index of learning, and to study other behaviours that can be measured and analysed differently. Still, it is important to work, beforehand or in concomitance, on a conceptual frame-work that will go beyond the simple stage of post-hoc explanation. Research could be enlarged by the study of contingency conditions in a less usual perspective, particularly in terms of an analysis of the sufficiency and necessity of behaviours, and also of conjugale and synchrohous reinforce-ment. We examine the question of intra-individual behavioural variability, while distinguishing induced variability front reinforced variability. Thèse less studied concepts seem crucial to the opérant model. Key words : human opérant behaviour, opérant conditioning, contingency.