Научные отчеты, том 6, Номер статьи: 23323 (2016) Цитировать эту статью
7116 Доступов
26 цитат
24 Альтметрика
Подробности о метриках
Восприятие временных связей между сенсорными событиями является ключевым процессом распознавания динамической среды. Суждение временного порядка (TOJ) и суждение одновременности (SJ) используются для исследования этого процесса восприятия. TOJ и SJ демонстрируют идентичные психометрические параметры. Однако накапливаются психофизические данные, которые отличают TOJ от SJ. Некоторые исследования предположили, что перцептивные процессы для SJ (например, обнаружение последовательности/одновременности) также включены в TOJ, тогда как TOJ требует большего количества процессов (например, определение временного порядка). Другие исследования предложили два независимых процесса для TOJ и SJ. Чтобы выявить различия в нейронной активности, связанной с TOJ и SJ, мы провели функциональную магнитно-резонансную томографию участников во время TOJ и SJ с идентичными тактильными стимулами. TOJ-специфическая активность наблюдалась во многих регионах (например, в левой вентральной и двусторонней дорсальной премоторной коре и левой задней теменной коре), которые перекрывают общую сеть временного прогнозирования для систем восприятия и моторики. Специфическая для SJ активация наблюдалась только в задней части островковой коры. Наши результаты показывают, что TOJ требует больше процессов, чем SJ, и что и TOJ, и SJ реализуют определенные компоненты процесса. Таким образом, нейронные различия между TOJ и SJ сочетают в себе черты, описанные в предыдущих психофизических гипотезах, которые предлагали разные механизмы.
Мы живем в динамичном мире, который меняется со временем. Временные отношения между сенсорными событиями предоставляют ключевую информацию для распознавания мира. Психофизика суждения о временном порядке (TOJ) сенсорных событий широко исследовалась1,2,3,4,5,6. В TOJ участники получают два стимула с определенной асинхронностью начала стимула (SOA) и решают, какой стимул представлен первым (или последним). Доля суждений TOJ как функция SOA обычно представляет собой сигмоидальную кривую, которую можно подогнать к кумулятивной функции Гаусса. Таким образом, психометрическая функция предоставляет два параметра: среднее значение и стандартное отклонение (σ). Среднее значение представляет собой точку субъективной одновременности (PSS), а σ представляет собой временную чувствительность восприятия.
Парадигма TOJ выявила различные психофизические явления. Например, эффект предварительного входа (т.е. влияние внимания на скорость восприятия) был продемонстрирован с использованием TOJ5,7. В другом примере скрещивание рук увеличивает частоту искажений тактильной ТД на руках6,8,9,10. Этот дефицит скрещенных рук предполагает, что мозг обрабатывает пространственное расположение рук, прежде чем временно упорядочить тактильные сигналы от соответствующих рук.
В другом популярном задании — оценке одновременности (SJ) — участники оценивают, предъявляются ли два стимула одновременно или последовательно. Доля суждений SJ представляет собой колоколообразную кривую, которую можно подогнать к гауссовой частотной функции. Таким образом, как и TOJ, SJ также предоставляет среднее значение (PSS) и стандартное отклонение (σ).
Парадигма SJ также использовалась для исследования психофизических явлений. Например, перцептивная адаптация к когерентной задержке между звуковыми и визуальными стимулами была идентифицирована с помощью SJ11. Этот эффект задержки адаптации также наблюдался в аудиовизуальных4,12,13,14,15 и других кросс-модальных16,17 TOJ. Соответственно, экспериментальные результаты TOJ и SJ иногда обсуждались в контексте предположения, что обе задачи отражают одни и те же психологические или нейронные механизмы4,15,18. Некоторые классические исследования предполагали, что восприятие последовательности является необходимым и достаточным условием для правильного восприятия временного порядка, в которых модель перцептивной латентности19 предполагала, что TOJ и SJ основаны на одних и тех же внутренних событиях (обзор у Аллана20).
SJ and SJ > TOJ, details in Supplementary Tables S1 and S2 online). In these contrasts, we excluded the effect of the difference in σ between TOJ and SJ (see fMRI data analysis in Methods) based on the standard interpretation38 and adjustment procedure39,40 for the difference in difficulty between the tasks in the fMRI studies. We used a significance threshold of P < 0.001 uncorrected at the voxel level and P < 0.05 FWE (family-wise error) corrected at the cluster level41./p> SJ contrasts were inclusively masked with the TOJ > rest contrasts (P < 0.05 uncorrected at the voxel level). In this contrast, the σ difference between the tasks was regressed out as a covariate of no interest. Furthermore, the voxels that correlated with the σ differences (P < 0.05 uncorrected at the voxel level) were exclusively masked. dPMC, dorsal premotor cortex; vPMC, ventral premotor cortex; IPL, inferior parietal lobule; SPL, superior parietal lobule./p> TOJ contrasts were inclusively masked with the SJ > rest contrasts (P < 0.05 uncorrected at the voxel level). In this contrast, the σ difference between the tasks was regressed out as a covariate of no interest. Furthermore, the voxels that correlated with the σ differences (P < 0.05 uncorrected at the voxel level) were exclusively masked. IC, insular cortex./p> SJ contrast exhibited greater activation of the left dorsal premotor cortex (dPMC), the left ventral premotor cortex (vPMC), the right dPMC, left PPC (inferior parietal lobule, IPL, and superior parietal lobule, SPL), and the bilateral thalamus (Fig. 3, clusters 1–5 in Table S1). The activation cluster of the PPC extended to the marginal area with the superior temporal gyrus (STG) (cluster 4 in Table S1). In addition, when moderating the significant threshold at the cluster level (P < 0.05 uncorrected) without changing that at the voxel level (P < 0.001 uncorrected), the TOJ > SJ contrast exhibited activation in the right cerebellum (clusters 6 and 8 in Table S1). Moreover, the right cerebellum revealed significant activation in the TOJ > rest contrast but not in the SJ > rest contrast (Supplementary Fig. S1 and Tables S3 and S4 online)./p> TOJ contrast revealed greater activation only in the left posterior insular cortex (IC) (Fig. 4, cluster 1 in Table S2). Using the liberal threshold at the cluster level (P < 0.05 uncorrected), we also observed greater activation in the right posterior IC (cluster 2 in Table S2)./p> SJ contrast revealed stronger activation in the left vPMC, the bilateral dPMC, the left PPC, and the bilateral thalamus. The TOJ-specific activation was thus observed in multiple brain regions, whereas the SJ > TOJ contrast revealed stronger activation in the posterior IC only. Therefore, regarding the first aim of the present study, our fMRI study supported the psychophysical hypothesis that TOJ involves more processes than SJ2,21,24,25,28,29. Regarding the second aim of the present study, our results thus suggest that not only TOJ but also SJ involves its own specific processes. Some psychophysical studies have proposed that TOJ requires more processes than SJ, whereas the processes for SJ are included in those for TOJ2,21,25,28. Other studies have proposed two independent processes for TOJ and SJ3,23,36. Our fMRI results thus exhibit combined features of these two psychophysical proposals. That is, the brain implements independent processes for TOJ (e.g., determination of temporal order) and SJ (e.g., detection of simultaneity/successiveness); however, the brain recruits more regions for TOJ-specific processing. We subsequently discuss the task-specific regions./p> SJ contrast, activation of the PPC extended to the marginal area with the STG (cluster 4 in Table S1), suggesting that the TOJ-specific region overlapped with the TPJ. In addition, using the liberal threshold at the cluster level (P < 0.05 uncorrected), the TOJ > SJ contrast exhibited peak activity in the left anterior IC (cluster 7 in Table S1). These observations suggested that a region near the VFC or a part of the VFC may have been activated. However, these activations were observed in the left hemisphere. The ventral attention network is considered to be lateralised to the right hemisphere31. Thus, the fMRI results do not support our first working hypothesis./p> SJ contrast, greater activation was observed in the left vPMC, the bilateral dPMC, and the left PPC. Using the liberal threshold at the cluster level, greater activation also appeared in the right cerebellum, which exceeded the threshold of the FWE correction for the TOJ > rest contrast but not for the SJ > rest contrast. In addition, the TOJ > SJ contrast exhibited significant activation in the thalamus bilaterally. Thalamus activity was also observed in a previous fMRI study on tactile TOJ using a time-irrespective control task30, and the thalamus plays a role in generating movements and monitoring one's own actions42,43,44. These observations favour our second working hypothesis that the motor control network is activated specifically by TOJs. Notably, the TOJ-specific cerebral activations were primarily distributed in the left hemisphere even though the right vs. left hand factor was counterbalanced in our study (see Tasks in Methods). Based on the consistent involvement of the left premotor and parietal cortices in perceptual temporal expectations, Coull and Nobre45 proposed a general temporal prediction network for perceptual and motor states. Our observation suggests that tactile TOJ involves the general temporal prediction network./p> SJ contrast but not in our SJ > TOJ contrast. The lack of a crossed-arm deficit during SJs is in agreement with our present observation of less activity in the left PPC during SJ than during TOJ./p> TOJ contrast, Fig. 3b in Lux et al. shows that the parietal IC activation includes a relatively anterior area, suggesting that visual SJ also activates a region similar to the posterior IC in our SJ > TOJ contrast. Along with the parietal IC, Lux et al. also detected neural correlates for visual SJ in other regions, such as the left TPJ, the left inferior frontal gyrus (IFG), the left middle frontal gyrus, and the left STG. These regions should contain the regions associated with general temporal processing that is not always specific to SJ. Our observations suggest that the posterior IC is involved in SJ-specific process components, such as detectors3,23 or comparators36, for judgements of simultaneity/successiveness of tactile stimuli./p> SJ contrast. This observation is consistent with a previous report indicating that the bilateral pre-SMAs showed difficulty-dependent activity during a duration discrimination task38. Thus, the pre-SMAs may be a TOJ-specific region, if we abandon the convention that neural correlates of task difficulty should not be recognised as task-related regions./p> SJ contrast may reveal FEF activity, suggesting that visual TOJ shares neural correlates with the control of eye movements. This hypothesis may be supported by psychophysical findings of interactions between visual TOJs and eye movements55,56./p> SJ contrast, activation of the left PPC extended to the boundary with the STG. This observation may suggest that the left TPJ is involved in TOJ, independent of differences in the sensory modality. Thus far, the results and interpretations of the laterality of temporal order processing are complicated. In addition to the accumulation of neural evidence obtained with whole-brain imaging (e.g., fMRI), further investigations using transient disruption techniques, such as transcranial magnetic stimulation (TMS), may help resolve this issue. Woo et al. applied TMS to participants during visual TOJ61. Their results showed that TMS over the right PPC delayed detection of a stimulus presented at the contralateral visual field, whereas TMS over the left PPC did not evoke a significant effect, supporting the involvement of the right PPC in the prior entry effect. Notably, Woo et al. used relatively short TMS-onset delays (50-200 ms from the first stimuli in the visual targets), and the effects of TMS on the right PPC peaked at 50 ms after the presentation of the first visual stimuli. These TMS-onset delays were adequate for testing neural causality of the prior entry effect, whereas the delays might be too short to affect the neural function for the determination of the temporal order of stimuli. In Woo et al., the average values of the TMS effect on the left PPC increased as the TMS-onset delay increased (from 50 to 200 ms), even though the effect did not eventually reach statistical significance. If longer TMS delays (e.g., > 200 ms) are used, a significant effect on TOJs may appear with TMS applied to the left PPC./p> SJ contrast exhibited activation of the left-dominant motor control or general prediction network, similar to the present tactile tasks, whereas there was no activation in the SJ > TOJ contrast. Notably, an fMRI study on motor control demonstrated that a discrete motor task activates more brain regions than a rhythmic motor task66. A similar difference in neural activation between discrete and rhythmic tasks may also occur in TOJ because TOJ-specific activity overlaps the motor control network. The discussions over sensory modalities are merely speculative. Further neural evidence is necessary to resolve these issues./p> SJ; SJ > TOJ; TOJ > rest; and SJ > rest. The contrast images were generated for each subject and then entered into a one-sample t-test to create a random effect SPM. The SPM{T} was transformed into normal distribution units SPM{Z}. Significantly activated voxels were identified using a threshold of P < 0.001 uncorrected at the voxel level (Z = 3.09) and P < 0.05 FWE-corrected at the cluster level41./p> SJ and SJ > TOJ contrasts, we excluded the effect of the σ difference between the tasks, according to the conventions that the behavioural difference between the tasks should be compensated for in the fMRI contrasts39,40. The σ difference for each subject was regressed out as a covariate of no interest in these contrasts. Furthermore, the voxels that were correlated with the σ difference (P < 0.05 uncorrected at the voxel level) were exclusively masked in the present study. In addition, to avoid false activations as a result of deactivations in the contrasting conditions, the TOJ > SJ and SJ > TOJ contrasts were inclusively masked with the TOJ > rest and SJ > rest contrasts (P < 0.05 uncorrected at the voxel level), respectively./p>
Русский
English
Français
Deutsch
Español
Italiano
Português
日本語
한국어