Region of interest analyses.

Semantic memory representations are generally well maintained in aging, whereas semantic control is thought to be more affected. To explain this phenomenon, this study tested the predictions of the Compensation-Related Utilization of Neural Circuits Hypothesis (CRUNCH), focusing on task demands in aging as a possible framework. The CRUNCH effect would manifest itself in semantic tasks through a compensatory increase in neural activation in semantic control network regions but only up to a certain threshold of task demands. This study compares 39 younger (20–35 years old) with 39 older participants (60–75 years old) in a triad-based semantic judgment task performed in an fMRI scanner while manipulating task demand levels (low versus high) through semantic distance. In line with the CRUNCH predictions, differences in neurofunctional activation and behavioral performance (accuracy and response times) were expected in younger versus older participants in the low- versus high-demand conditions, which should be manifested in semantic control Regions of Interest (ROIs). Our older participants had intact behavioral performance, as proposed in the literature for semantic memory tasks (maintained accuracy and slower response times (RTs)). Age-invariant behavioral performance in the older group compared to the younger one is necessary to test the CRUNCH predictions. The older adults were also characterized by high cognitive reserve, as our neuropsychological tests showed. Our behavioral results confirmed that our task successfully manipulated task demands: error rates, RTs and perceived difficulty increased with increasing task demands in both age groups. We did not find an interaction between age group and task demand, or a statistically significant difference in activation between the low- and high-demand conditions for either RTs or accuracy. As for brain activation, we did not find the expected age group by task demand interaction, or a significant main effect of task demand. Overall, our results are compatible with some neural activation in the semantic network and the semantic control network, largely in frontotemporoparietal regions. ROI analyses demonstrated significant effects (but no interactions) of task demand in the left and right inferior frontal gyrus, the left posterior middle temporal gyrus, the posterior inferior temporal gyrus and the prefrontal gyrus. Overall, our test did not confirm the CRUNCH predictions.

语义记忆表征通常在衰老过程中保持完好，而语义控制则被认为更易受衰老影响。为解释这一现象，本研究检验了神经回路补偿相关利用假说（Compensation-Related Utilization of Neural Circuits Hypothesis, CRUNCH）的预测，将衰老中的任务需求作为潜在研究框架。该假说提出，CRUNCH效应会通过语义控制脑网络区域的神经激活代偿性增强，在语义任务中显现，但仅在达到特定任务需求阈值前有效。本研究共纳入39名年轻参与者（20~35岁）与39名年长参与者（60~75岁），让其在功能磁共振成像（fMRI）扫描仪内完成基于三联刺激的语义判断任务，并通过语义距离操控任务需求水平（低需求与高需求）。按照CRUNCH假说的预测，在低需求与高需求条件下，年轻与年长参与者的神经功能激活及行为表现（准确率与反应时）差异应体现在语义控制感兴趣区（Regions of Interest, ROIs）中。正如已有语义记忆任务相关研究所提出的，本研究的年长参与者行为表现完好：准确率保持稳定，但反应时（RTs）更长。要检验CRUNCH假说的预测，年长组相较于年轻组需保持行为表现不变，这一点是必要前提。此外，神经心理学测试结果显示，本次研究的年长成人具备较高的认知储备。行为学结果证实本任务成功操控了任务需求：两个年龄组的错误率、反应时及主观感知难度均随任务需求提升而增加。本研究未发现年龄组与任务需求间存在交互效应，也未发现高、低需求条件下，参与者的反应时或准确率存在统计学显著差异。就大脑激活而言，本研究未发现预期的年龄组×任务需求交互效应，也未观察到任务需求存在显著主效应。总体而言，本研究结果与语义网络及语义控制网络的部分神经激活特征相符，激活区域主要集中在额颞顶叶区域。对感兴趣区的分析显示，左侧及右侧额下回、左侧后中颞回、后颞下回及前额回均存在显著的任务需求主效应（但未发现交互效应）。综上，本研究未验证CRUNCH假说的预测。