Studies of lexical processing have relied heavily on adult ratings of word learning age or age of acquisition, which have been shown to be strongly predictive of processing speed. This study reports a set of objective norms derived in a large-scale study of British children's naming of 297 pictured objects (including 232 from the Snodgrass & Vanderwart, 1980, set). In addition, data were obtained on measures of rated age of acquisition, rated frequency, imageability, object familiarity, picture-name agreement, and name agreement. We discuss the relationship between the objective measure and adult ratings of word learning age. Objective measures should be used when available, but where not, our data suggest that adult ratings provide a reliable and valid measure of real word learning age.

Early learned words are recognized and produced faster than later learned words. The authors showed that such age of acquisition effects are a natural property of connectionist models trained by backpropagation when patterns are introduced at different points into training and learning of early and late patterns is cumulative and interleaved. Analysis of hidden unit activations indicated that the age of acquisition effect reflects a gradual reduction in network plasticity and a consequent failure to differentiate late items as effectively as early ones. Further simulations examined the effects of vocabulary size, learning rate, sparseness of coding, use of a modified learning algorithm, loss of early items, acquisition of very late items, and lesioning the network. The relationship between age of acquisition and word frequency was explored, including analyses of how the relative influence of these factors is modulated by introducing weight decay.

Six experiments are reported that contrasted the effects of frequency and age of acquisition on written word recognition. Age of acquisition affected word-naming speed when frequency was controlled (Experiment 1), but there was no effect of frequency when age of acquisition was controlled (Experiment 2). Experiments 3 and 4 found an effect of age of acquisition upon immediate but not delayed naming speed, but no frequency effect on either immediate or delayed naming once age of acquisition was controlled. Independent effects of frequency and age of acquisition were observed in the lexical decision task (Experiments 5 and 6). Implications for theoretical accounts of word recognition and the possible roles of age of acquisition and frequency in word recognition are discussed.The frequency with which a word occurs in the language is widely believed to affect the ease which that word can be recognized and responded to, with common, high-frequency words being recognized more rapidly and/or more accurately than less common, low-frequency words. This belief has been incorporated into cognitive models of word recognition which regard the capacity to explain frequency effects as an important aspect of a model's adequacy (see Monsell, 1991, for a review of both evidence and theories). We argue in this article that for several aspects of visual word recognition, this belief in the importance of word frequency is mistaken. We suggest that it has arisen because^f a consistent failure to disconfound the possible effects of word frequency from effects of other factors which correlate with frequency, most notably the age at which words are learned. High-frequency words tend to be learned earlier in life than low-frequency words, so that sets of words selected as being of high or low frequency of occurrence tend also to be sets of words which are early-or late-acquired, respectively. Despite evidence that age of acquisition (AOA) affects visual word recognition (see Brown & Watson, 1987;Gilhooly & Watson, 1981), most studies of word frequency have failed to control for it. Evidence from the existing

No abstract

Previous research on the effects of age of acquisition on lexical processing has relied on adult estimates of the age at which children learn words. The authors report 2 experiments in which effects of age of acquisition on lexical retrieval are demonstrated using real age-of-acquisition norms. In Experiment 1, real age of acquisition emerged as a powerful predictor of adult object-naming speed. There were also significant effects of visual complexity, word frequency, and name agreement. Similar results were obtained in reanalyses of data from 2 other studies of object naming. In Experiment 2, real age of acquisition affected immediate but not delayed object-naming speed. The authors conclude that age-of-acquisition effects are real and suggest that age of acquisition influences the speed with which spoken word forms can be retrieved from the phonological lexicon.