大鼠學習抑制型躲避學習作業前後杏仁核神經元活性的變化

Abstract

本研究利用細胞外單一神經元活性記錄，來探討杏仁核在兩種形式的抑制型躲避學習作業中（熱板踏下式抑制型躲避學習作業及傳統穿越式抑制型躲避學習作業）所扮演的角色。簡言之，大鼠分為實驗組（n＝8）和控制組（n＝5），實驗組大鼠在制約前，共進入作業環境三次，然後在制約後一天再進行測試。控制組和實驗組的差別在於控制組的大鼠其制約階段是在一個和原作業環境全然不同的地方進行。我們成功的利用嫌惡刺激，讓大鼠學會了這兩種作業。透過神經元反應的記錄顯示，隨著大鼠進入作業環境的次數增加，杏仁核細胞放電的頻率隨之下降，而在制約之後，放電的頻率在實驗組中再度上升，但在控制組中則沒有這種現象，故杏仁核可能藉由放電的強度來傳遞訊息。初級體感覺皮層的細胞在此兩種作業中扮演的角色並不一致，在熱板踏下式抑制型躲避學習作業中，此部位細胞的放電和學習歷程無關，但在傳統穿越式抑制型躲避學習作業中，放電頻率則表現出和杏仁核細胞相似的模式。神經系統可能藉由參與於某一事件的神經區域間同步放電的增加來記錄訊息，但在本實驗中，並沒有觀察到杏仁核和初級體感覺皮層的細胞在制約後同步放電增加的現象。抑制型躲避學習作業牽涉到工具制約和古典制約，從本研究及杏仁核各核區毀除實驗得到的結果，熱板踏下式抑制型躲避學習作業可能主要是藉由工具制約來達成，而古典制約則是學習傳統穿越式抑制型躲避學習作業很重要的因素之一。我們並沒有得到支持性的證據顯示學習抑制型躲避學習作業時，右側杏仁核比左側杏仁核更重要。有些研究顯示杏仁核可能有神經元分工的現象，但本研究中沒有得到決定性的支持證據。5-12Hz的慢腦波（theta rhythm band）是大鼠清醒且警戒時杏仁核主要的慢腦波成份，於本實驗中，大鼠由等待箱進入作業環境時，此成份明顯的增加，但實驗組大鼠在制約成功後，於等待箱中時此成份即明顯出現，表示實驗組大鼠於制約成功後，在測試時，即使是在等待箱中，依然保持警戒的狀態。因此，綜合從本研究得到的結果及前人相關的研究，提出雖然杏仁核不一定是長期記憶儲存的所在，但至少本核區在制約後一天仍密切參與於行為調控中，調控的機制可能是利用整體放電的增加。大鼠可能是利用不同的策略來學習此研究中所使用的兩種抑制型躲避學習作業，於熱板踏下式抑制型躲避學習作業中，工具制約的成分主導了學習，而於傳統抑制型躲避學習作業中，古典制約的成分是很重要的學習因素。

This study examined the functional role of the amygdala in two versions of inhibitory avoidance tasks, the step-down avoidance motivated by heat and the step-through avoidance motivated by electric shock, with single-unit recording during task sessions. Long-Evans rats successfully acquired these two conditioning paradigms. Rats were divided into the experimental group (n = 8) and the control group (n = 5). Briefly, rats in the experimental group were allowed to adapt the task environment three times, then received the conditioning session, and followed by the test session. Rats in the control group received the unconditioned noxious stimuli in an environment distinct from the original task apparatus. Changes in the neuronal response showed that as the animal got used to the task environment from the first-entry session to the habituated session (the third-entry), neuronal firing rates in the amygdala decreased. After the conditioning session, firing rates increased again in the experimental group, but not in the control group. These results imply that the amygdala may code information by neuronal firing rates. The role of the primary somatosensory cortex in these two tasks was unclear. Data obtained from the experimental group showed that in the step-down avoidance task, firing rates decreased from the first-entry session to the habituated session, and showed further decrement in the test session. This result implies that the primary somatosensory cortex may not be involved in the step-down conditioning process. However, in the step-through avoidance task, firing rates decreased from the first-entry session to the habituated session and increased again in the test session, which implies that the primary somatosensory cortex may be involved in the step-through conditioning process. We failed to find changes in the synchronization among neurons after the conditioning session between the primary somatosensory cortex and the amygdala. Inhibitory avoidance learning may engage both the operant component and the classical component. The present data combined with the lesion data obtained in our lab suggest that the operant component dominated in the step-down avoidance task, and the classical component is an important factor in learning the step-through avoidance task. Our experiment design failed to yield supporting evidence for more involvement of the right amygdala than the left one. It is possible that the neurons in the amygdala are functionally segregated, but this notion needs further examination. We found out that the theta rhythm dominates as the rats entered the task apparatus from the waiting cage, and rats in the experimental group showed dominant theta rhythm in both the waiting cage and the task apparatus after conditioning. Since theta rhythm correlates with awake and alert state in the amygdala, these results imply that the rats in the experimental group may stay alert even in the waiting cage after the conditioning session. From previous related studies, although the amygdala may not be the storage site of long-term memory in this kind of inhibitory avoidance learning, we suggest that the amygdala could pivot avoidance behavior through increase in population firing rate, at least within one day after the conditioning session. Further, the animal may adopt different strategies in acquiring the two inhibitory avoidance tasks: learning in the step-down avoidance task was coped with more active behavior than in the step-through avoidance task.