Fig. 3From: Syntactic chunking reveals a core syntactic representation of multi-digit numbers, which is generative and automaticDigit error rate in the 6-word numbers in Experiment 1 for each serial position of the number words. In condition D, the error rate was the lowest for the first two words and increased thereafter (except the last word)—a standard pattern of primacy effects in free recall tasks. Conditions A and B did not show this pattern: In these conditions, the error rate did not show a consistent pattern. The word “thousand” is not plotted here, as its digit error rate is undefined. The letters H, T, and O denote a Hundreds, Tens, and Ones word, respectivelyBack to article page