Our long-term knowledge of language requires us to store information about the words of our language - what they sound like (phonology), what they mean (semantics), and how they are combined to construct utterances (syntax). This array of information is made available to us when we hear or see words. When native speakers of English, for example, hear the word walk, they will recognize it as a phonological form they have encountered before, and will recover both its meaning and its syntactic characteristics (as a verb or a noun). Importantly many English words – and even more so in other languages - are more complex than the simple form walk, as in the inflectional variants walks, walking, walked. The question then is: Does each of these complex forms has its own stored representation, or are they analyzed into their stems (e.g., {walk}) and their respective suffixes (e.g., {~s}, {~ing}, and {~ed})? Similar questions arise for the representation of complex derivational forms, such as walker or walkable, as well as the issue of whether semantic factors play a role in the choice of decomposed or full-form representations.
I have been interested in these questions in the context of Arabic where every content word is morphologically complex consisting of a consonantal root, which conveys semantic information, and a word pattern which acts as a mold that provides morpho-syntactic and phonological information. For example the word [katam] hide, consists of the root {ktm} and the word pattern {faEal}, where the letters "f, E, l" are placeholders for the first, second and third consonant of the root respectively. We (Marslen-Wilson and myself) have shown that Arabic roots and word patterns plays a pivotal role in guiding the process of accessing the lexicon both from auditory and visual input. Furthermore, in collaboration with Friedemann Pulvemüller, Olaf Hauk and Yry Shtyrov, we have established that both the root and the word pattern elicit distinct MisMatch Negativity responses, but with different time courses and different topographies. The root MMN has a symmetric fronto-central distribution, whereas the word pattern MMN lateralizes to the left, indicating involvement of left perisylvian areas. This is consistent with a dual systems approach contrasting bihemispheric lexico-semantic processes with a language-specific Left Hemisphere peri-sylvian system, preferentially activated by grammatical morphemes such as the word-pattern.
We are currently working, in collaboration with Dr S, Boujraf in Fez, Morocco and Dr M. Elalaoui-Fares, Rabat, Morocco, on a project using the voxel-based lesion-symptom mapping technique which relies on correlating signal intensity across the entire brain of focal-lesion patients with a behavioural response of some sort (we are using priming scores and lexical decision latencies) to identify which regions are critical for various tasks. Our goal is to determine which domain of knowledge (morphology, phonology, semantics) is affected depending on the extent and site of the lesion.