Discussion

The purpose of this study was to evaluate the roles of specific brain regions that may be involved in a temporal discrimination task (e.g., 2 vs. 8 sec) by using ERPs and hippocampal theta spectra. The data from the temporal discrimination task were compared to a simple reaction time control task that did not require animals to make a temporal discrimination. We discuss the brain areas involved in temporal processing with attention to the frontal cortex, hippocampus, and cerebellum in order to evaluate the information-processing model proposed by Gibbon et al. (1984). The ERPs recorded from the three regions following stimulus onset consisted of N1, P2, and P3 components, as shown in Figure 13.3. There were no significant differences in N1 components. There was a main effect of task in the P2 component amplitude that was larger for the T-task than for the C-task. Interestingly, the hippocampus and the cerebellum seem to show almost the same amplitude and latency. These results suggest that the cerebellum is involved in temporal processing of durations in the seconds range. Meck et al. (Mattel and Meck, 2000; Matell et al., this volume; Meck, 1996; Meck and Benson, 2002) have proposed that timing for durations in the seconds-to-minutes range is subserved by a different system, involving frontal-striatum circuits. Little attention has been given to the relation between the cerebellar cortex and these frontal-striatal circuits, although there are projections from the cerebellum to the basal ganglia (e.g., Middleton and Strick, 1994). It is debatable whether the cerebellum and frontal-striatal circuits are independent timing systems or work in conjunction with each other (Diedrichsen et al., this volume; Ivry, 1996; Ivry and Richardson, 2002).

The P2 components in the hippocampus were prominent during the T-task, and the P2 latency was almost the same time for the hippocampus and the cerebellum. This result suggested that the information processing related to working memory might be carried out simultaneously in the course of the task and be observed concurrently with the activation of the clock mechanism. The idea here is that certain aspects of memory processing may be required for the initialization of working memory at the beginning of a trial. The P2 latency was significantly longer for the recordings in the frontal cortex than for the hippocampus and the cerebellum. This result showed that the information processing in the frontal cortex was delayed and might involve the integration of temporal processing as represented by the comparison stage (see Meck, this volume; Meck et al., 1987).

The P3 components were observed in all three brain regions at the time of stimulus onset. We used a time window from 300 to 600 msec in positive peak to define the P3 component. The P3 components are relatively long-latency ERPs and have commonly been employed to investigate cognitive processing in a variety of tasks. This component could be elicited by task-relevant orienting stimuli. The tone onset indicates the opportunity for reward in both tasks. There were no significant differences in the P3 components at the onset of the signal for both tasks. In contrast, the ERPs following stimulus offset consisted of the observed positive potential for the frontal cortex and the hippocampus. The mean amplitude of the positive potential had a tendency to be larger for the T-task than for the C-task. This result suggests that the frontal cortex mediates the decision/comparison stage of temporal processing. This component could not be thought of as the same as the onset P3 component. The P3-like component is typically elicited by an "oddball task" in rats (e.g., Jodo et al., 1995). The onset P3 most likely reflects the detection of the reward value of the signal, whereas the offset P3 likely reflects decision making about the duration of the signal. The onset P3 amplitudes were larger than the offset P3 amplitudes. It is important to investigate these offset ERPs in spite of their lower amplitude. In future studies, we propose that it would be better to design experiments in which the period defined by the onset starting stimulus to the onset ending stimulus is used to evaluate the processing of stimulus duration.

The hippocampal theta power increased more in the T-task than in the C-task at the onset of stimuli. The peak frequency shifted toward a higher range in the T-task than in the C-task, as shown in Figure 13.4. These results suggest that the hippocampus was initialized for temporal processing by the resetting of working memory at stimulus onset. According to the information-processing model outlined above, the pacemaker will generate pulses that go through a switch into an accumulator. On a new trial, the timing process may need to initialize or reset working memory before transferring pulses from the accumulator to the hippocampus. At this point, the hippocampus continues working to process the current temporal memories that may reflect hippocampal theta activity (e.g., Meck, 2000b). It is clear that frontal activation related to temporal processing at the beginning of the interval is also observed. This is the first study to investigate the correlations between EEG and proposed interval timing mechanisms in animals (for a description of related work in humans, see Pouthas, this volume). It will be necessary to gather more EEG data to obtain a better understanding of the relations between the neuronal activity in various brain structures and temporal discrimination processes.

0 0

Post a comment