Contingent Negative Variation And Time Estimation

18.2.1 Contingent Negative Variation: Descriptive Data

Walter et al. (1964) reported that when a warning stimulus indicates the presentation of an imperative stimulus after a constant foreperiod (1 or 2 sec), a negative potential shift gradually develops in the EEG over wide areas of the scalp during the interstimulus interval. This slow shift of negative polarity is fully developed whenever a subject is expecting the occurrence of a significant event in the next few seconds, as illustrated in Figure 18.1a. When the imperative stimulus is withdrawn, there is a progressive diminution of the CNV, as shown in Figure 18.1b to e, and when it is reinstated, the CNV is reestablished, as illustrated in Figure 18.1f.

The CNV changes to a biphasic waveform with two negative peaks when fore-periods exceed 3 to 4 sec (Loveless and Sanford, 1974). It is generally agreed that the early component, maximal over the frontal cortex, constitutes an index of the orienting response to the warning stimulus (e.g., Macar and Besson, 1985; Rohrbaugh et al., 1976). The second peak of the biphasic CNV, the terminal CNV, which has a more central distribution over motor areas of the cortex, has been linked to preparatory processes. It has been argued that short intervals in the typical range of 1 to 2 sec would not allow sufficient time for the CNV to develop its biphasic waveform with the initial and terminal CNV components, and that the monophasic waveform represents the consequence of overlapping components (Rohrbaugh and Gaillard, 1983).

The CNV has been associated with several cognitive processes, including expectancy and attention. In particular, it is thought to reflect the preparation or anticipation of a response (Birbaumer et al., 1990; Rockstroh et al., 1993). Several lines of evidence also support the hypothesis that temporal processing is reflected in the CNV (see Brannon and Roitman, this volume; Elbert et al., 1991; Macar and Besson, 1985; McAdam, 1966; Ruchkin et al., 1977). The one aspect of interval timing that has been investigated in CNV research is how this negative shift develops and how

FIGURE 18.1 (a) CNV completely developed between the warning stimulus (S1) and the imperative stimulus (S2). (b to e) Progressive reduction of the CNV after withdrawal of the imperative stimulus. (f) CNV reinstatement after the restoration of the imperative stimulus. (Adapted from Walter, W.G., Cooper, R., Aldridge, V.J., McCallum, W.C., and Winter, A.L., Nature, 203, 380-384, 1964.)

FIGURE 18.1 (a) CNV completely developed between the warning stimulus (S1) and the imperative stimulus (S2). (b to e) Progressive reduction of the CNV after withdrawal of the imperative stimulus. (f) CNV reinstatement after the restoration of the imperative stimulus. (Adapted from Walter, W.G., Cooper, R., Aldridge, V.J., McCallum, W.C., and Winter, A.L., Nature, 203, 380-384, 1964.)

its amplitude and time course (i.e., negativity resolution) vary when participants have to estimate a given interval and press a key by themselves, in the absence of any imperative stimulus. For example, in the Macar et al. (1999) study, results showed that the longer the judgment or the production of a 2.5-sec target interval, the larger the CNV amplitude recorded was at a fronto-central site (electrode FCz). This illustrates that variation in temporal performance is reflected in slow brain potential changes. We will now examine results of some studies concerned with the relations between both CNV amplitude and resolution and the quality of time estimation performance in normal subjects and patients who suffer from time estimation deficits.

18.2.2 CNV Amplitude and Temporal Processing

The CNV has been shown to vary in amplitude depending on the level of training, accuracy, and precision of performance, as well as on mental state or disease. Not surprisingly, the picture of these variations given by the literature is not unequivocal, but relatively complex. McAdam (1966) reported a relationship between the pro gressive increase of the CNV amplitude over trials and the progressive learning of the temporal parameters of the task in the first stages of the experiment. Then the CNV decreased as temporal accuracy approached the maximum level of performance. Macar and Vitton (1980) proposed that the progressive development of the CNV might reflect the elaboration of an internal time reference, corresponding to the target duration. Once this reference is constituted, it may be used more automatically. Some parallels could be drawn with the results of a very recent experiment conducted by my research group (Pfeuty et al., 2003). Subjects had to compare the tempos of two isochronous tone sequences made up of either three or six intervals. CNV amplitude increased during the encoding phase over the frontal part of the scalp. Moreover, when the sequence was composed of six intervals, the results showed that the negativity stopped increasing after the third interval, as illustrated in Figure 18.2. We proposed that this increase in CNV amplitude might reflect the building of a memory trace of the reference interval and that beyond a critical number of intervals, the memory trace would be optimal and time intervals would no longer be estimated.

These data suggest that the level of brain activity may be related to the level of performance, with a relatively low activation corresponding to efficient processing. Some other experimental data corroborate this assumption. Ladanyi and Dubrovsky (1985) found smaller CNVs in accurate than in inaccurate subjects when they were performing a verbal estimation task. Similarly, Casini et al. (1999) showed that the level of CNV activity obtained with correct responses was lower than that obtained with incorrect responses in a task of identification of signal durations. High levels of CNV activity recorded when responses were wrong may reflect an inappropriate attention effort leading to defective encoding. This suggests that good temporal performance could be the result of an efficacious, but economic information-processing mechanism in the brain.

It could be assumed that during practice specific brain structures are progressively selected and interconnected in order to form a reduced network of neurons. Once this network is selected, correct responses may be produced almost automatically without demanding much attention, which would explain the observation of a low level of brain activity in well-practiced subjects. On the one hand, studies using dual-task paradigms have evidenced the crucial role of attention for temporal processing (see Brown, 1997; Fortin, this volume; Pang and McAuley, this volume). On the other hand, relationships between the level of attention and the CNV amplitude have been exemplified by electrophysiological studies outside the time domain. A correlation between the early component of the CNV and the number of checked letters in a letter cancellation test has been found, indicating significant differences in early CNV between low and high performance (Kropp et al., 2001). Similarly, in Filipovic et al. (2001) all segments of the CNV waveform were different in go and no-go conditions. The go CNV displayed a typical pattern of slow rising negativity reflecting the buildup of attentional resources necessary for adequate performance.

The effects of distraction on the CNV have also been investigated in elderly subjects (Michalewski et al., 1980). CNV amplitudes at central and parietal sites were comparable between young and old adults, while amplitude at the frontal sites was consistently reduced in the elderly compared to the young. These data are

0 0

Post a comment