以事件相關腦電位探討刺激出現前記憶登錄活動中清晰記憶與熟悉感歷程的分離

An ERP Study of the Dissociation between Recollection- and Familiarity-Related Prestimulus Encoding Activities

近期研究指出記憶形成會受到學習項目出現前的神經活動影響。本研究探討此「刺激前之隨後記憶效果」（prestimulus subsequent memory effect）是否因為隨後的再認記憶表現係基於清晰記憶（recollection）或是熟悉感（familiarity）歷程而有不同型態。實驗區分為學習及來源記憶測驗階段。在學習階段，受試者依據提示對隨後出現的刺激項目做兩種判斷作業中的一種。在隨後的來源記憶測驗階段，受試者針對曾在學習階段出現的項目進行再認，並在來源記憶判斷中指出在學習階段中，對該項目所做的判斷作業為何。學習階段的事件相關腦電位以提示出現的時間為始，根據之後是否正確再認，以及再認後是否指出正確來源做區分。結果顯示，相較於隨後被遺忘的項目，正確接受項目的刺激前提示在左前區域有較為負向的波形。將正確接受項目區分為來源記憶正確與不正確的接受項目時，相較於來源記憶不正確項目，來源記憶正確項目在前方腦區有較為負向的波形；而相較於遺忘項目，來源記憶不正確項目出現前的提示則引發廣泛分布的正向波。根據雙重歷程理論，來源正確的接受項目同時涉及清晰記憶與熟悉感歷程，而來源不正確的接受項目僅依賴熟悉感歷程作反應。本實驗的結果顯示刺激出現前的記憶活動，可依隨後的記憶牽涉到清晰記憶或是熟悉感歷程加以分離。

Recent findings in long-term memory suggest that successful encoding is modulated by the neural activities prior to encountering the to-be-encoded items. The current study examined whether this “prestimulus subsequent memory effect” dissociates between recollection- and familiarity-related encoding. Participants first engaged in a study phase, during which they made one of two types of judgments to concrete nouns. The judgment to be made for each word was signaled by a cue presented prior to the nouns. In the following test phase, participants differentiated studied words from unstudied ones. They were also asked to identify the type of judgment made to the identified words at study. ERPs elicited by the task cues during the study phase were sorted according to whether the following study items were recognized and their sources (i.e., judgment type during encoding) correctly identified in the subsequent recognition test. The cue-eliciting ERPs were found to be more negative-going over the left frontal area when the study items were correctly identified (hit trials) in comparison to those that were incorrectly rejected (miss trials). Importantly, when the hit trials were further segregated according to their source judgment accuracy, source-correct hit trials gave rise to a frontally distributed negativity in comparison to source-incorrect ones. Source-incorrect hit trials, on the other hand, were associated with a widespread positivity when compared with miss trials. On the basis that source correct hits involve both recollection- and familiarity-based recognition whereas source incorrect hits involve familiarity only, we argue that the current findings reflect dissociable prestimulus subsequent memory effects for subsequent recollection- and familiarity-related encoding activities.