The first aim of the current study was to investigate the effects of the distance manipulation on sentence comprehension in normal individuals (NI) and persons with aphasia (PWA). Consistent with Gibson’s (1998; 2000) locality theory, when distance was manipulated by varying the syntactic dependency (subject-verb: SV and filler-gap: FG) and type of modifier (No, prepositional-phrase: PP, and relative-clause: RC), NI demonstrated an increase in the number of errors and response times to the yes/no questions as the distance increases. NI also exhibited a relatively systematic increase of reading times on the verb (RT-V) and response times (RT) as a function of distance manipulation except for the most complex condition (FG-RC) in which 77% of NI performed at chance-level. More than 60% of normal individuals performed at chance-level in sentences with FG-dependency. Consistent with the previous literature on syntactic comprehension in aging, older adults showed decreased performance on the filler-gap computations.

PWA generated more errors in the FG- than SV-dependency. However, their sentence comprehension was not affected by the manipulation of the modifiers. Their RT-V data were difficult to interpret due to the very limited observations for FG-dependency conditions after chance-level performers were excluded from the analyses. One can argue that the high rate of chance-level performance in PWA, especially in the FG conditions, is consistent with the specific impairments hypotheses. However, chance-level performance was observed in majority of individuals with aphasia and even in normal individuals, which was not predicted by those hypotheses. The current results were more consistent with resource-related hypotheses which suggested that sentence comprehension deficits will manifest themselves regardless of the type of aphasia when their capacity is taxed to be exceeded.

When distance-based integration cost was held constant between the two dependencies, FG-dependency generated more errors across the groups. However, the RT-V was not significantly different between the two conditions in older adults. These results are consistent with the locality theory. Considering the longer RT-V differences between FG- and SV-dependency in older adults than younger adults, the non-significant group results might be due to the large variability of older adults’ performance and the limited sample size.