时间知觉的注意调节:一项ERP研究

Attentional Modulation of Time Perception:An ERP Study

采用双任务实验范式探讨时间知觉注意调节的动态过程。通过指导语使被试按照比例将注意分配到声音刺激的时间属性与音调属性上,形成对时间因素不同注意程度的五种注意条件,同时记录反应时、错误率和事件相关电位数据。对峰值分析发现,P2注意条件主效应不显著,但多重比较时T(只注意时间)与P(只注意音调)条件差异显著;对CNV平均波幅分析,发现注意条件主效应显著,随着分配在时间属性上的注意的减少,其波幅逐渐降低;对T与P条件的差异波地形图分析发现,波幅最大的区域位于额叶、中央区和顶叶;该差异波的偶极子定位于辅助运动区(额上回、额中回)和顶下小叶。以上结果表明,P2阶段存在时间信息加工;CNV反映时间知觉的注意调节,时间知觉中存在控制加工应该以CNV的出现为指标;辅助运动区可能为时间知觉的核心成分,具有跨通道效应;顶下小叶也与对时间的注意有关,但并不是特异性的。本研究不支持Lewis提出的"自动"与"认知控制"计时系统理论中"秒"为两系统分界点以及辅助运动区只属于"自动计时系统"的观点。

Previous researches have indicated that attention can modulate subjective time perception： the less the attention distributed to duration, the more the chances of time being misperceived. Most brain mechanism studies have focused on the comparison between temporal and nontemporal conditions. These studies were confined to the attention shift between temporal and nontemporal attributes, and did not refer to the process of attentional modulation of time perception. A recent research studied the cerebral areas concerned with the attentional modulation of time perception using fimctional magnetic resonance imaging （fMRI）. The study employed a dual - task paradigm in which attention was distributed differently between the time and color of the visual stimulus. However, the dynamic process of attentional modulation of time perception was still unclear because of the low temporal resolution of the fMRI. In contrast, this experiment used event - related potentials （ERPs） to investigate this process. Nineteen volunteers participated in this experiment. All subjects performed a training session at least two days before the formal experiment. An attentional cue directed the subjects to selectively attend to （1） the stimulus time （T condition）, （2） the time rather than the pitch （Tp condition）, （3） both parameters equally （tp condition）, （4） the pitch rather than the time （tP condition）, or （5） the pitch （P condition）. Effective attentional allocation was obtained by varying the relative likelihood of trials involving time distinguishing or those involving pitch distinguishing. The reaction time, error rate, and event - related potentials were recorded simultaneously by comuuters during the experiment. The electroencephalogram （EEG） was recorded from 64 scalp locations of the extended 10 20 system, using an electrode cap （Brain Product Inc.） with Ag/AgC1 electrodes. Linked mastoids were used as the reference, and a ground electrode （GND） was placed on the medial aspect of the frontal lobes （on the middle line, between FZ and FCz）. An estimation of the dipolar sources of the T - P difference waveform was performed using the Brain Electrical Source Analysis program （BESA version 5.0）, which is based on the four - shell ellipsoidal head model. Further, a principal component analysis （PCA） was employed in the interval from 300 to 540 ms in order to estimate the minimum number of dipoles. Finally, a genetic algorithm was used to determine the location of the dipoles. Behavioral data showed that the subjects allocated their attention to the time and pitch based on the attentional cues. The ERP data indicated the following： （1）the main effect of the attentional conditions in the amplitude of P2 was not significant, but the difference between the T and P conditions was significant in the multiple comparison;（2）the main effect of the attentional conditions in the average amplitude of the contingent negative variation （CNV） was significant from the start latency time to 540ms, and an increase in the attention to time augmented the average amplitude of the CNV; （3）the map of the T - P difference waveform showed that the maximal amplitude of the difference waveform was located on the areas of the frontal, center, and parietal lobes; and （4） the dipoles of the T - P difference waveform were located on the supplementary motor area （superior frontal gyms, middle frontal gyms） and the inferior parietal lobule. These results have the following implications： （1） the time information process exists in P2 phase; （2） CNV reflects the attentional modulation of time perception, which is an index of the existing controlled process; （3） the supplementary motor area may play a key role in time perception with cross - modal effect; and （4） the inferior parietal lobule is also related to the attention of time, albeit time attention is not the only attribute. However, these results are contradictory to Lewis＇ notion that （1） a ＂second＂ is the boundary between an ＂automatic＂ and a ＂cognitively controlled＂ timing system and （2） the supplementary motor area belongs only to the ＂automatic timing system.＂