GKT范式在确定犯罪嫌疑人中的应用：一项ERP测谎研究

刑事犯罪已成为当今各国面临的普遍性社会问题,因此急需可靠有效的测谎技术以适应当前打击犯罪的严峻形势。本研究采用犯罪知识测试范式（Guilty knowledge test, GKT）,以犯罪嫌疑人为对象,以真实犯罪现场为素材,探究真实环境下与测谎探测相关的脑电指标（P300、CNV）的有效性和可靠性。行为数据显示,目标刺激的反应正确率显著小于探测和无关刺激,探测刺激的反应时显著小于目标刺激和无关刺激。脑电数据显示,相较于目标刺激和无关刺激,探测刺激诱发的P300波幅最大且P300的检出率为92.5%,而在CNV指标上目标刺激诱发的波幅最大, 无关刺激次之,探测刺激最小。结果表明,与CNV相比,P300能够有效检测被试所隐藏的犯罪细节信息,且检出率较高,稳定性较好,是一种较为可靠的测谎指标。     

Effects of anticipation certainty on preparatory brain activity and anticipatory postural adjustments associated with voluntary unilateral arm movement while standing

We examined the effects of anticipation certainty concerning which voluntary movement is required in response to a stimulus while standing on preparatory brain activity and anticipatory postural adjustments (APAs). Ten right-handed adults abducted their left or right arm rapidly in response to a visual imperative stimulus, based on the type of stimulus. A warning cue, which did or did not contain information about the side of arm abduction, was presented 2000ms before the imperative stimulus. Preparatory brain activity before arm abduction was quantified by the mean amplitude of the contingent negative variation 100ms before the imperative stimulus (late CNV amplitude). Compared with the low anticipation condition, in the high anticipation condition the following results were obtained only in the case of right arm abduction: (1) larger late CNV amplitude, (2) earlier postural muscle activities with respect to the focal muscle of arm abduction, and (3) smaller peak displacement of center of pressure during the abduction. These findings suggest that high anticipation of voluntary movement of dominant arm to a stimulus while standing influences preparatory brain activity before the movement, resulting in earlier APAs and thus smaller disturbance of postural equilibrium during the movement.     

Dissociation of event-related potentials indexing arousal and semantic cohesion during emotional word encoding

Event-related potential (ERP) studies have shown that emotional stimuli elicit greater amplitude late positive-polarity potentials (LPPs) than neutral stimuli. This effect has been attributed to arousal, but emotional stimuli are also more semantically coherent than uncategorized neutral stimuli. ERPs were recorded during encoding of positive, negative, uncategorized neutral, and categorized neutral words. Differences in LPP amplitude elicited by emotional versus uncategorized neutral stimuli were evident from 450 to 1000 ms. From 450 to 700 ms, LPP effects at midline and right hemisphere frontal electrodes indexed arousal, whereas LPP effects at left hemisphere centro-parietal electrodes indexed semantic cohesion. This dissociation helps specify the processes underlying emotional stimulus encoding, and suggests the need to control for semantic cohesion in emotional information processing studies.     

汉语普通话声调感知的老年化效应：来自ERP的证据

研究考察汉语普通话老年人前注意阶段声调感知状况, 探究是否存在领域特殊的老年化。运用事件相关电位技术, 采用被动oddball范式诱发MMN回应, 控制领域一般性因素的影响。结果显示涉及范畴变化的声调和非语音音调诱发MMN强度衰减, 不涉及范畴变化的声调诱发MMN强度未衰退。研究结果表明在前注意阶段, 在特定汉语普通话声调范畴知识加工能力上存在领域特殊的衰退, 而不涉及母语音位知识的声调的感知存在领域特殊的一定程度的保留, 这一保留与时间维度上补偿机制的调用有关。受补偿机制调节, 语言加工呈现出衰退或保留等不同的老年化进程。     

正性情绪诱导下的自我参照加工：来自ERPs的证据

采用事件相关电位考察正性情绪对自我参照加工程度效应的影响。实验采用图片启动范式, 先呈现情绪图片, 然后再呈现自我参照刺激。实验发现, 在P2上, 中性情绪条件比正性情绪条件激发了更大的波幅, 高自我相关的刺激比其它刺激诱发了更短的P2潜伏期; 在N2上, 高自我相关名字比中等自我相关名字和非自我相关名字诱发了更小的N2的波幅和更长潜伏期; 在P3上, 高自我相关名字比中等自我相关名字和非自我相关名字诱发了更大的P3波幅, 中等自我相关名字比非自我相关名字诱发了更大的P3波幅。实验结果表明, 人类大脑对正性情绪刺激的加工可能是不敏感的。无论在正性情绪启动还是中性启动的影响下, 自我参照加工都能展现出稳定的特征, 而且高自我相关的刺激会得到更为深入和精细的加工, 表现出自我参照加工的程度效应。     

The role of levels of processing in disentangling the ERP signatures of conscious visual processing

We aimed to distinguish electrophysiological signatures of visual awareness from other task-related processes through manipulating the level of processing of visual stimuli. During an event-related EEG experiment, 36 subjects performed either color (low-level condition) or magnitude (high-level condition) evaluations of masked digits. Participants also assessed subjective visibility of each stimulus using the Perceptual Awareness Scale (PAS). Mean amplitude of the components of interest was analyzed (VAN − 140–240 ms; LP − 380–480 ms) with weighted regression mixed model. In the VAN component time window the mean amplitude correlated with PAS rating in both conditions. Mean amplitude in the LP time window correlated with PAS ratings in the high-level condition, but not in the low-level condition. Our results support the temporal unfolding of ERP makers of conscious processing, with an early component reflecting the initial perceptual experience and a late component being a correlate of the conscious experience of non-perceptual information.     

The role of transparency in perceptual grouping and pattern recognition.

The shortest stimulus exposure time for which transparency can be seen was examined. In the first experiment, overlapping digits were presented for 120 ms and the luminance in the overlapping regions was varied. Subjects reported, in separate blocks of trials, either the apparent transparency of the digits or the identity of the digits. When the luminance was set so that one set of digits appeared to be seen through the other, recognition of the digits was high. When the luminance in the overlapping regions did not produce impressions of transparency, digit recognition was low. In the second experiment, digit identification at several stimulus durations was compared between stimuli that had luminance that was valid for transparency and stimuli that had invalid luminance. Performance was found to be higher in the valid luminance condition than in the invalid condition after as little as 60 ms exposure duration. This result suggests that the impression of transparency requires only relatively short exposure durations.     

Tonal frequency shifts and gaps in acoustic stimulation as reflex-modifying events

When a relatively weak signal, such as a mild tone, precedes an intense reflex-eliciting stimulus by an appropriate interval (about 100 msec), the amplitude of the elicited reaction is often reduced. It was found that in student volunteers a brief gap in a steady pure tone that occurred 150 msec prior to a mild tap to the glabella (the flat region between the eyebrows) could inhibit the eyeblink elicited by the tap. It was also found that a shift in tonal frequency across a gap in a tone was more inhibitory than a gap with no frequency shift, but it was no more inhibitory than the onset of the short second tone alone. The final study determined the minimum amount of frequency shift required to produce an additional inhibitory effect above that of a gap alone. The findings are discussed in terms of various aspects of sensory processing.     

Modulation of the acoustic startle reflex in humans in the absence of anticipatory changes in the middle ear reflex

If a weak tone precedes an intense tone, then the acoustic startle eyeblink reflex elicited by the stronger stimulus is inhibited. It has been suggested that the leading stimulus gives rise to a protective middle ear reflex that attenuates the effective intensity of the second. This hypothesis was tested and disproved. In seven subjects intense tone bursts sufficient to elicit both intratympanic and eyeblink responses were presented sometimes alone and sometimes preceded at various lead times (25 to 400 msec) by a weak tone. The weak tone inhibited the amplitude of the eye blink to the strong tone, maximally at intervals of 100 to 200 msec, but was never seen to produce any of the anticipatory impedance changes that would be characteristic of middle ear reflex activity during the interval between the two stimuli.     

汉字刺激质量和频率在N400上的交互

采用事件相关电位,运用无启动的同一字判断（靶字延迟反应）范式,以大学生为被试探讨汉字刺激质量和频率是否具有交互作用。行为结果显示,刺激质量和频率在反应时上存在交互作用。脑电结果显示,重度模糊诱发了更大的N250和N400,且刺激质量和频率在N400上存在交互作用。结论是,汉字刺激质量和频率在反应时和N400上存在交互作用,表明在汉字识别中存在串级加工,支持IA模型。     

Blindness to response-compatible stimuli in the psychological refractory period paradigm

This study investigated the conditions under which the processing in a speeded response task interferes with concurrent processing in a visual encoding task. Three experiments used a dual-task paradigm, in which a speeded left or right response to at one was combined with the identification of a masked left-or right pointing arrow following the tone with variable SOA. Two additional experiments tested the impact of the presentation of pure tone on visual encoding.There were four major findings. First, an unspecific decrease in identification accuracy was observed with decreasing SOA. Second, a blindness to response-compatible stimuli was observed with speeded responses. Third, a specific interference was found between low- and high-pitched tones and left- or right-pointing arrows. Fourth, the specific tone-arrow interference modulated the specific responsearrow interference when the task allowed both to occur simultaneously. The present findings, which suggest both procedural and structural interference between response preparation and stimulus encoding, are discussed in terms of a two-stage model of action planning.     

Temporal integration of acoustic and cutaneous stimuli shown in the blink reflex

Temporal integration of pairs of brief blink-eliciting acoustic and cutaneous stimuli was investigated to determine if there was integration of stimuli from different modalities. Reflexes elicited by a tone burst or by a brief electrical shock to the supraorbital nerve followed by a second tone burst or shock at short stimulus onset asynchronies (SOAs) were larger and faster than control reflexes elicited by a single stimulus identical to the lead stimulus of the stimulus pairs. Reflex amplitude was augmented at longer SOAs where there was no effect on latency. Temporal integration was evident for all stimulus pairs, showing that it is due, at least in part, to processes that occur outside specific sensory pathways. Heterogeneous stimulus pairs produced greater reflex enhancement than did homogeneous stimulus pairs. This finding was examined further in Experiment 2, which showed that reflex enhancement with pairs of acoustic pulses was unaffected by the frequency of the second stimulus, suggesting that sensory masking was not acting to suppress reflex expression with acoustic pulse pairs. Integration of reflexogenic acoustic stimuli shown in the blink reflex is restricted to shorter intervals than is integration of acoustic stimuli shown by psychophysical procedures, suggesting that the two methods reflect different aspects of stimulus processing. Integration of reflexogenic stimuli may result from summation of activity associated more directly with reflex expression than with perceptual awareness.     

P300和CNV在GKT的延时反应范式中测谎效果的分离

在杀人游戏情境中,设置无反馈和有反馈两个实验组,采用GKT（guilt knowledge test）的延时反应范式,考察P300和CNV两种ERP成分的测谎效果。结果发现：刺激呈现后大约450ms诱发了一个正性ERP成分（P300）,大约800ms诱发了一个负性ERP成分（CNV）。在有、无反馈两种实验范式下,P300和CNV两种测谎指标的变化情况发生了分离——有无反馈不影响P300峰值大小;而在有反馈条件下,“杀手”对于犯罪细节词语（探测刺激）的CNV明显增大。使用Bootstrap波幅差异分析法,分别计算P300和CNV在有、无反馈情况下的检出率,结果发现：两个指标的检出率在有、无反馈两种实验范式下的变化情况也发生了分离。这些结果表明：P300和CNV两个指标具有不同的测谎功能,这可能和它们所反映的不同欺骗心理过程有关。     

Accessory stimuli modulate effects of nonconscious priming

In a recent study, it was shown that subliminal priming (SP) effects are affected by the temporal predictability of a stimulus onset. If predictability is not given, SP effects are eliminated (Naccache, Blandin, & Dehaene, 2002). In two experiments, we investigated how different levels of preparation for target processing affect SP effects. For this purpose, an accessory tone stimulus was presented at different times prior to a subliminal priming task. The results demonstrate a clear modulation of the SP effects at different foreperiod intervals. Relative to conditions without an accessory stimulus, SP effects were smaller for short foreperiod intervals of the accessory stimulus, and larger for long foreperiod intervals. The results suggest that the presentation of an accessory stimulus facilitates response activation processes because of the participants' enhanced level of preparation for stimulus processing.     

Neurophysiological evidence for disturbances of conflict processing in patients with schizophrenia

Deficits in cognition are a hallmark of schizophrenia. In the present study, the authors investigated the effects of schizophrenia on the neural correlates of conflict processing in a single-trial version of the Stroop task by using event-related brain potentials. Relative to matched controls, patients with schizophrenia showed increased Stroop interference in response time, but this effect was eliminated when the effect of response slowing was controlled. In controls, conflict processing was associated with a negative wave peaking between 400 and 500 ms (N450) and conflict sustained potential (SP) peaking between 600 and 800 ms after stimulus onset. In patients with schizophrenia, the amplitude of the N450 was significantly attenuated and the conflict SP was absent. These results provide evidence for the existence of altered neural processes associated with conflict processing that may be associated with dysfunction of the anterior cingulate and prefrontal cortex in patients with schizophrenia.     

Task integration as a factor in secondary-task effects on sequence learning

In several studies it has been found that implicit sequence learning is impaired by the concurrent performance of a secondary task. In most studies the task was to count high-pitched tones when high-pitched and low-pitched tones were presented in a random sequence. In this paper the hypothesis is examined that dual-task interference in the particular combination of tasks results from task integration, in particular from the learning of an integrated visual-auditory sequence in which every second (auditory) element is random. Instead of the tone-counting task a similar go/no-go task was used in which foot responses to high-pitched tones were performed. In Exp. 1 the sequence of tones was random in one condition, but in two other conditions repeated tone sequences of 5 and 6 elements were combined with visual sequences of 6 elements. Under dual-task test conditions, implicit learning of the visual as well as the auditory sequences was better with the auditory sequence of 6 elements than of 5 elements, while under single-task test conditions the nature of the tone sequence had no effect. In Exp. 2 the superior implicit learning with the 6-element sequence was replicated with different test procedures in which either the visual or auditory sequence was changed to random or in which the two sequences remained intact but were shifted by one element relative to each other. Randomization of the visual or auditory sequences not only impaired visual or auditory RT, respectively, but also impaired RT to stimuli in the other modality, and this cross-modal effect was almost as strong as the intra-modal effect of randomization. Finally, in Exp. 3 it was shown that integrated visual-auditory sequences are learned only when responses to both of them are required, but not when the tones can be neglected. These results are consistent with a conception of implicit learning as (at least partly) a basic and nonselective type of learning of all potentially behaviorally relevant relations between stimuli in the environment and one's own actions.     

Anticipatory attention during the sleep onset period

To examine whether anticipatory attention or expectancy is a cognitive process that is automatic or requires conscious control, we employed a paired-stimulus event-related potential (ERP) paradigm during the transition to sleep. The slow negative ERP wave observed between two successive stimuli, the Contingent Negative Variation (CNV), reflects attention and expectancy to the second stimulus. Thirteen good sleepers were instructed to respond to the second stimulus in a pair during waking sessions. In a non-response paradigm modified for sleep, participants then fell asleep while tones played. As expected, N1 decreased and P2 increased in amplitude systematically with the loss of consciousness at sleep onset; the CNV was increasingly more positive. Sleep onset latency was correlated with the amplitude of the CNV. The systematic attenuation of the CNV waveform at sleep onset and its absence in sleep indicates that anticipatory attention requires endogenous conscious control.     

不同时距加工机制的比较:来自ERP的证据(Ⅰ)

采用事件相关电位的方法分析了不同时距加工机制的差异。实验采用时间两分任务,要求被试形成短(400ms)和长(1600ms)的时间判断标准,然后对7个刺激系列进行短、长的判断。结果表明时间知觉过程可以诱发出一种类似于CNV的慢性负波。对于400ms的短时距加工存在显著的半球效应,右半球的波幅显著高于左半球。1600ms的时距加工没有表现出明显的半球效应。不同时距的加工存在显著的差异。1600ms比400ms加工诱发出更大的CNV波幅。     

Are crossmodal correspondences relative or absolute? Sequential effects on speeded classification

Crossmodal correspondences have often been demonstrated using congruency effects between pairs of stimuli in different sensory modalities that vary along separate dimensions. To date, however, it is still unclear the extent to which these correspondences are relative versus absolute in nature: that is, whether they result from pre-defined values that rigidly link the two dimensions or rather result from flexible values related to the previous occurrence of the crossmodal stimuli. Here, we investigated this issue in a speeded classification task featuring the correspondence between auditory pitch and visual size (e.g., congruent correspondence between high pitch/small disc and low pitch/large disc). Participants classified the size of the visual stimuli (large vs. small) while hearing concurrent high- or low-pitched task-irrelevant sounds. On some trials, visual stimuli were paired instead with “intermediate” pitch, that could be interpreted differently according to the auditory stimulus on the preceding trial (i.e., as “lower” following the presentation of a high pitch tone, but as “higher” following the presentation of a low pitch tone). Performance on sequence-congruent trials (e.g., when a small disc paired with the intermediate-pitched tone was preceded by a low pitch tone) was compared to sequence-incongruent trials (e.g., when a small disc paired with the intermediate-pitch tone was by a high-pitched tone). The results revealed faster classification responses on sequence-congruent than on sequence-incongruent trials. This demonstrates that the effect of the pitch/size correspondence is relative in nature, and subjected to trial-by-trial interpretation of the stimulus pair.     

The startle response and auditory temporal summation in neonates

The present study assessed temporal summation of transient and sustained stimuli in the startle eyeblink response system in neonates during quiet sleep. Subjects received 100-dB(A), fast-rising broadband noise bursts of two types: (a) single stimuli varying in duration from 20 to 100 ms and (b) pairs of 3-ms bursts presented at interpulse intervals corresponding to the single-stimulus durations. In addition, a single 3-ms pulse was used as an anchor point for both stimulus types. For startle amplitude, single stimuli were more effective than were paired stimuli, but the temporal summation functions were similar for the two types of stimuli. Response amplitude increased as stimulus duration/interval increased to 50 ms, but not beyond. For startle probability, temporal summation was similar for single and paired stimuli at 20 ms. Pairs of pulses were equally effective at 20, 35, and 50 ms, beyond which the second pulse was not effective. Increasing the duration of single stimuli from 20 to 35 ms resulted in increased probability, illustrating a contribution of sustained summation beyond that of transient summation. Response latency was generally greater for paired than for single stimuli. The results suggest that temporal summation of brief stimuli is deficient in the neonate. These data were compared to adult data from an analogous study, and suggest that the transient system is immature in infants, and that this immaturity is expressed differently by startle amplitude, probability, and latency.