Setting the midpoint of sentences: The role of the left hemisphere

Abstract

The human brain has a remarkable capacity to focus processing resources based on the features and the relevance of the task at hand. The two cerebral hemispheres contribute differentially to this capacity, with the left hemisphere linguistic and right hemisphere visuo-spatial abilities each offering unique contributions. For example, previous research has established that healthy participants set the subjective mid-point of written sentences more leftwards of center, compared to unpronounceable letter strings or simple lines. Remarkably, patients with right hemisphere damage exhibiting unilateral spatial neglect also show this pattern, even though, as well known in the literature, they tend toward a rightward- bias for non-linguistic stimuli. This evidence suggests that the leftward bias for sentential material is due to linguistic, mainly left-hemisphere mediated processes, which are largely unimpaired in right brain-damaged patients, and intact in heathy participants. To test this hypothesis, we compared sentence bisection performance to that of letter strings and simple lines in left brain-damaged patients (with and without aphasia). If the larger leftward bias in the bisection of sentential material is based on linguistic processes, then the left brain-damaged patients should show a reduction or absence of a leftward bias in sentence bisection. We tested twenty-four left brain-damaged patients (12 with aphasia and 12 without aphasia), and 24 age-matched elderly participants (patients and controls were all right-handed). Participants were asked to bisect 240 stimuli, comprising: (i) affirmative and interrogative clauses, (ii) sentences with lexical and syntactic violations, (iii) letter strings and (iv) simple lines. As predicted, neurologically intact participants showed larger leftward biases in bisecting written readable sentences compared to strings of letters. In contrast, the left hemispheredamaged patients (both with and without aphasia) showed no differences in bisecting sentences and letter strings or lines. These findings indicate that the larger leftward bias exhibited by healthy participants in the bisection of sentences is likely due to ortho-phonological coding taking place implicitly during the bisection task. This ortho-phonological coding is impaired with left brain damage – also in absence of apparent aphasia – leading to the left hemispheredamaged patients showing a reduced leftward bias in sentence bisection. These findings support the hypothesis that the leftward bias in the bisection of written sentences is the result of ortho-phonological influences rather than visual-spatial biases.