Anderson Britt, Soliman Sherif, O'Malley Shannon, Danckert James, Besner Derek
Department of Psychology, University of Waterloo Waterloo, ON, Canada.
Psychology Department, McMaster University Hamilton, ON, Canada.
Front Psychol. 2015 Mar 27;6:321. doi: 10.3389/fpsyg.2015.00321. eCollection 2015.
Drawing on theoretical and computational work with the localist dual route reading model and results from behavioral studies, Besner et al. (2011) proposed that the ability to perform tasks that require overriding stimulus-specific defaults (e.g., semantics when naming Arabic numerals, and phonology when evaluating the parity of number words) necessitate the ability to modulate the strength of connections between cognitive modules for lexical representation, semantics, and phonology on a task- and stimulus-specific basis. We used functional magnetic resonance imaging to evaluate this account by assessing changes in functional connectivity while participants performed tasks that did and did not require such stimulus-task default overrides. The occipital region showing the greatest modulation of BOLD signal strength for the two stimulus types was used as the seed region for Granger causality mapping (GCM). Our GCM analysis revealed a region of rostromedial frontal cortex with a crossover interaction. When participants performed tasks that required overriding stimulus type defaults (i.e., parity judgments of number words and naming Arabic numerals) functional connectivity between the occipital region and rostromedial frontal cortex was present. Statistically significant functional connectivity was absent when the tasks were the default for the stimulus type (i.e., parity judgments of Arabic numerals and reading number words). This frontal region (BA 10) has previously been shown to be involved in goal-directed behavior and maintenance of a specific task set. We conclude that overriding stimulus-task defaults requires a modulation of connection strengths between cognitive modules and that the override mechanism predicted from cognitive theory is instantiated by frontal modulation of neural activity of brain regions specialized for sensory processing.
基于与局部主义双通路阅读模型相关的理论和计算工作以及行为研究结果，贝斯纳等人（2011年）提出，执行需要推翻特定刺激默认设置的任务（例如，给阿拉伯数字命名时涉及语义，评估数字单词的奇偶性时涉及语音）的能力，需要具备在特定任务和刺激基础上调节词汇表征、语义和语音等认知模块之间连接强度的能力。我们使用功能磁共振成像来评估这一观点，通过在参与者执行需要和不需要这种刺激任务默认设置推翻的任务时，评估功能连接的变化。枕叶区域对两种刺激类型显示出最大的血氧水平依赖（BOLD）信号强度调节，被用作格兰杰因果关系映射（GCM）的种子区域。我们的GCM分析揭示了前额叶内侧皮质的一个区域存在交叉交互作用。当参与者执行需要推翻刺激类型默认设置的任务（即数字单词的奇偶性判断和给阿拉伯数字命名）时，枕叶区域和前额叶内侧皮质之间存在功能连接。当任务是刺激类型的默认设置时（即阿拉伯数字的奇偶性判断和阅读数字单词），则不存在统计学上显著的功能连接。这个额叶区域（BA 10）先前已被证明与目标导向行为和特定任务集的维持有关。我们得出结论，推翻刺激任务默认设置需要调节认知模块之间的连接强度，并且认知理论预测的推翻机制是由专门用于感觉处理的脑区神经活动的额叶调节来实现的。
