Reduced Syntactic Processing Efficiency in Older Adults During Sentence Comprehension

Abstract

Researchers have frequently reported an age-related decline in semantic processing during sentence comprehension. However, it remains unclear whether syntactic processing also declines or whether it remains constant as people age. In the present study, 26 younger adults and 20 older adults were recruited and matched in terms of working memory, general intelligence, verbal intelligence and fluency. They were then asked to make semantic acceptability judgments while completing a Chinese sentence reading task. The behavioral results revealed that the older adults had significantly lower accuracy on measures of semantic and syntactic processing compared to younger adults. Event-related potential (ERP) results showed that during semantic processing, older adults had a significantly reduced amplitude and delayed peak latency of the N400 compared to the younger adults. During syntactic processing, older adults also showed delayed peak latency of the P600 relative to younger adults. Moreover, while P600 amplitude was comparable between the two age groups, larger P600 amplitude was associated with worse performance only in the older adults. Together, the behavioral and ERP data suggest that there is an age-related decline in both semantic and syntactic processing, with a trend toward lower efficiency in syntactic ability.

Keywords: ERP; P600; aging; neural efficiency; syntactic processing.

FIGURE 1

Grand averaged ERPs evoked by the critical words at four representative electrodes. (A) Grand averaged ERPs in younger adults. (B) Grand averaged ERPs in older adults. The solid line represents congruent sentences (CON), the dashed line represents sentences with semantic violation (SEM), and the broken line represents sentences with both semantic and syntactic violations (SEM + SYN). Waveforms are time-locked to the onset of the critical words. Negative is plotted upward.

FIGURE 2

Semantic effect in younger and older groups. (A) Grand averaged ERPs evoked by the critical words in CON and sentences with semantic violation (SEM). (B) Topographic distribution of the semantic effect (SEM – CON) in younger and older groups from 300 to 800 ms after the critical word onset. Waveforms are time-locked to the onset of the critical words. Negative is plotted upward.

FIGURE 3

Syntactic effect in younger and older groups. (A) Grand averaged ERPs evoked by the critical words in sentences with semantic violation (SEM) and sentences with semantic and syntactic violation (SEM + SYN). (B) Topographic distribution of the semantic effect (SEM + SYN – SEM) in younger and older groups from 300 to 800 ms after the critical word onset. Waveforms are time-locked to the onset of the critical words. Negative is plotted upward.

FIGURE 4

Re-normalized grand averaged ERPs difference between conditions based on the N400 time window for semantic effect and syntactic effect in both groups. (A) Grand averaged ERPs difference for semantic effect (SEM – CON) in both younger and older adults. (B) Grand averaged ERPs difference for syntactic effect (SEM + SYN – SEM) in both younger and older adults. Re-normalization based on difference in the N400 time window. The solid line represents younger adults and the dashed line represents older adults. Waveforms are time-locked to the onset of the critical words. Negative is plotted upward.