How are a bilingual's two languages organized in the brain? A major focus of my research has been to establish this using both behavioral and neuroimaging methods. My research in this area has sought to elucidate two questions. First, what differences, if any, exist between the cerebral representation of each language. Second, what brain areas are involved in a bilingual's ability to switch between languages. Language switching was explored by presenting participants with stimuli in one language (blocked condition) or presenting runs of stimuli in which response language switches from one stimulus to the next (mixed condition). Studies investigating these issues have yielded the following results:

• Results from neuroimaging work with bilinguals suggests that the two languages do reveal different patterns of neural activity.
• In a picture naming task early bilinguals show increased activity in an area just anterior to Broca’s area suggesting that naming pictures in a less fluent language involves more effortful selection among semantic competitors. However, late bilinguals show increased activity in the inferior temporal lobe and the insula suggesting that these participats were having more difficulty making the link between the image and the articulatory code (Hernandez, Dapretto, Mazziotta & Bookheimer, in prep).
• Increased neural activity in a second language depends on the amount of orthographic overlap and the concreteness of the item during a semantic judgment task. Specifically, abstract items with low orthographic overlap show increased activity in brain areas known to be involved in semantic retrieval (Hernandez, Kotz & Friederici, in prep).
• Highly fluent early Spanish-English bilinguals show significant increases in the amount of activation in the dorsolateral prefrontal cortex when switching between languages relative to a single language condition (Hernandez, Martinez, & Kohnert, 2000).
• Older adult bilinguals show a substantial increase in language switching costs relative to young adults. These costs are largest when the switching is predictable, suggesting that the effect has to do with the frequency of switching (Hernandez & Kohnert, 1999). This increased cost diminishes across childhood, is stable in young and middle adulthood, and increases after age 60.

My research on the brain bases of bilingualism suggests that differences in neural activity between two languages are driven by fluency. In addition, the amount of orthorgraphic and representational overlap (i.e. concreteness) modulate how much additional neural activity is needed in a second language. Finally, results from my work suggest that the frontal lobes are intimately involved in having to switch between languages. The results from this work add to the current literature in two significant ways. First, my findings suggest that differential impairment of a language could be due to damage to areas involved in semantic retrieval or executive control. Second, these studies suggest that differences between languages are not absolute and may depend on factors such as amount of orthographic or semantic overlap.