Chunks are not enough: The insufficiency of feature frequency-based explanations of artificial grammar learning.
Philip Anthony Higham · Canadian Journal of Experimental Psychology/Revue canadienne de psychologie expérimentale · 1997
Tw o experiments tested chunk frequency expla- nations o f artificial gramma r learning which hold that classification performance i s dependent on some metric derived from the frequency with which certain features occur withi n th e lette r strin g stimuli . Experimen t 1 revealed that classification performance was affected b y close graphemic similarity between specific training (e.g., MXRVXT) and test strings (e.g., MXRMXT), despite the fact that similar strings did not contain frequently occurring features (e.g., bigrams or trigrams). This effect was repli- cated in Experiment 2a and Experiment 2b demonstrated that substituting letters to make the consonant string s pronounceable (e.g. , substituting X , R, and T , in th e consonant strin g MXRMX T with Y , A, I , to produc e MYAMYl) affected classification performance, despite the fact that objective measures of feature frequency were not altered. It is argued that models of classification that focus entirely on the frequency o f features within the literal stimulus are insufficient, and that some allowance must be made for how the stimulus is encoded.