To distinguish between the dual and single system views, I will investigate the brain bases of irregular and regular word transformations - in which lexicon and grammar can be contrasted, while other factors are held constant. For example, English irregular past tense forms cannot be predicted from the sounds of their stems (e.g., sing-sang, bring-brought), and therefore must be memorized in the lexicon, whereas regular past tenses are formed according to a predictable pattern (verb + -ed; e.g., look-looked), and therefore could be generated by a grammatical rule. A dual-system (but not single-system) view predicts links between irregulars (lexicon) and temporal cortex, and between regulars (grammar) and frontal cortex. This prediction is supported by preliminary studies, which have revealed double dissociations between the use of irregulars and regulars in patients with either temporal-lobe damage (posterior aphasics) or frontal-lobe damage (anterior aphasics).

The domain-specific/domain-general controversy will be addressed by investigating whether the lexicon/grammar distinction is tied to a fundamental distinction between two well-studied brain systems. My colleagues and I have posited that the memorization and use of words is subserved by temporal-lobe circuits previously implicated in the learning and use of fact and event knowledge, whereas the acquisition and use of grammatical rules is subserved by frontal/basal-ganglia circuits previously implicated in the learning and expression of motor, perceptual, and cognitive "skills," such as riding a bicycle. This hypothesis is supported by double dissociations between the production of irregulars and regulars in preliminary studies of patients with either damage to the temporal-lobe system and fact use deficits (patients with Alzheimer's disease), or damage to the frontal/basal-ganglia system and skill use deficits (patients with Parkinson's disease).

I propose to confirm and broaden these findings. Patients with either temporal-lobe or frontal/basal-ganglia damage will be given tasks probing the use of facts and skills, and the production and recognition of irregular and regular past tense inflection (sleep-slept, slip-slipped), plural inflection (mouse-mice, bee-bees), and derivational morphology (solemn-solemnity, eager-eagerness). My specific objectives are to investigate two hypotheses by testing for double dissociations:

Hypothesis 1: Lexical memory is linked to the temporal-lobe system. The left temporal-lobe damage (but relative sparing of frontal/basal-ganglia structures) in posterior aphasia and Alzheimer's disease is predicted to lead to greater difficulty using irregulars than regulars, and more errors with facts than skills.

Hypothesis 2: Grammatical rules are linked to the frontal/basal-ganglia system. The left frontal/basal-ganglia damage (but relative sparing of temporal lobe regions) in anterior aphasia and Parkinson's disease is predicted to lead to greater difficulty using regulars than irregulars, and more errors with skills than facts.

Significance: If the two well-studied brain systems play a role in lexicon and grammar, then data from the numerous investigations of the neural, computational, and developmental underpinnings of the two systems may be pertinent to language.