表徵特性對工作記憶為基注意力攫取的影響

The influence of representation on working memory-driven attentional capture

郭郡羽(Chun-Yu Kuo)

指導教授 ： 葉怡玉

共同指導教授 ： 趙軒甫

國立臺灣大學/理學院/心理學研究所/博士(2014年)

https://doi.org/10.6342/NTU.2014.00669

摘要

本論文透過「工作記憶為基的注意力攫取」現象探討工作記憶與注意力兩大心理學建構的互動。工作記憶為基的注意力攫取顯示：與工作記憶內容相符的訊息再次出現時，會獲得競爭優勢並被優先選擇。此現象在過去研究中已被反覆驗證，並作為工作記憶與注意力緊密互動的支持證據。然而，工作記憶中表徵特性所扮演的角色卻尚未被探討。本論文遵循傳統的雙作業派典要求參與者序列進行儲存、注意力作業及記憶相關作業，檢驗當記憶訊息再次出現在注意力作業中時，對注意力的影響。參與者在每個嘗試的最初需記憶一個訊息，並在嘗試的最末進行與該記憶表徵相關的作業。記憶訊息到記憶相關作業間，參與者需進行一個與記憶作業無關的運動方向偵測作業，以檢驗當記憶訊息再次出現時對注意力分派的影響。在運動方向偵測作業中，其中一個刺激會向上或向下移動，而其他刺激靜止，參與者需要判斷移動刺激的運動方向。其中，儲存於工作記憶裡的表徵可能與注意力作業中的移動目標物相符（有效情境）、與靜止的干擾物相符（無效情境）、或不出現於注意力作業的畫面中（中性情境）。當有效情境的反應時間顯著快於中性情境的反應時間或當無效情境的反應時間顯著慢於中性情境的反應時間時，此結果代表儲存於工作記憶的表徵攫取注意力。本論文的三個系列研究中分別關心注意力標籤表徵、既有表徵及表徵共伴性所帶來的表徵特性差異對工作記憶為基注意力攫取效果的影響。研究一操控儲存於工作記憶中的顏色表徵為之後記憶相關作業裡的目標物或干擾物而改變表徵的注意力標籤。結果發現若工作記憶中的內容為後續作業的目標物，被賦予目標物標籤，將優先攫取注意力；若工作記憶中的內容為後續作業的干擾物，被賦予干擾物標籤，則無法攫取注意力。研究二探討與工作記憶關係密切的長期記憶如何調節工作記憶為基的注意力攫取。參與者需記憶由字母組成的真字、假字或非字，並進行運動方向偵測作業。結果發現：儲存於工作記憶裡的字母共享既有表徵時（真字或可發音假字），單一字母再現時可透過分散激發攫取注意力；若儲存字母間彼此獨立，無共享的既有表徵時（非字），單獨字母將彼此削弱，此時，需完整視覺表徵再現方能攫取注意力。研究三進一步透過表徵共伴性的操弄，檢驗情境脈絡所誘發的表徵差異如何影響工作記憶為基的注意力攫取。研究三以唸名作業取代記憶再認作業，避免因使用特殊策略而傾向在注意力作業中利用工作記憶中的內容來強化記憶表徵。參與者唸出應記憶項目的名稱後，進行運動方向判斷作業。並跨實驗操弄工作記憶中的訊息對注意力作業目標物的預期性來檢驗情境脈絡的影響。結果發現工作記憶為基的注意力攫取為原始設定模式，當無特殊控制歷程介入時，與工作記憶內容相符的訊息將會攫取注意力；若透過共伴性建立了認知設定，則可透過策略化控制歷程避免注意力受到工作記憶內容的攫取。本論文在三個系列研究中透過觀察工作記憶為基的注意力攫取及其限制情境，與工作記憶內容相符的訊息再現非必然攫取注意力，尚受到不同表徵特性的影響，更精緻化的展現了注意力與工作記憶在不同情境裡的互動。

關鍵字

工作記憶 ； 選擇注意力 ； 認知控制 ； 注意力標籤 ； 既有表徵 ； 共伴性

並列摘要

The relationship between working memory and selective attention has been demonstrated by numerous studies which showed that the content of working memory biases attention. This effect, working memory-driven attentional capture, has been well demonstrated in the literature. However, few studies have investigated how the characteristics of the representation maintained in working memory may influence the capturing effect. This doctoral dissertation aims to address this issue across three studies using a dual-task paradigm that combines a memory related task and an attentional selection task. The participants in all studies were required to maintain an item in working memory for a later memory-related test while performing a motion judgment task. In the motion judgment task, the participants were required to judge the direction of a moving target. The memorized item was identical to the moving target in the valid condition; the memorized item was identical to the static distractor in the invalid condition; and the memorized item was not presented in the display of the motion judgment task in the neutral condition. The capturing effect was reflected by faster performance in the valid condition than in the neutral condition, or slower performance in the invalid condition than in the neutral condition. Study 1 focused on the effect of attentional tag associated with the information maintained in working memory. The memory related task showed two colored notched circles after the selection task. Attentional tag was manipulated by assigning the memorized color to be the to-be-judged and the to-be-ignored color in the target and distractor block, respectively. The results showed that the attentional tag modulated working memory-driven attentional capture. The color tagged as a target captured attention and reduced the reaction time for the motion judgment task, and the color tagged as a distractor did not capture attention. The goal of Study 2 was to examine how pre-existing representation in long-term memory modulated the capture effect driven by the individual component of the stimulus stored in working memory. The participants were required to remember letters which formed a word, pronounceable pseudoword, or nonword and then performed a motion judgment task. The results showed that the individual letters of a word or pronounceable pseudoword captured attention whereas letters of a nonword did not. In the nonword condition, reappearance of the complete visual representation is necessary for inducing the capturing effect. In Study 3, a naming task was used to replace the memory recognition task for eliminating potential strategic perceptual resampling, which may have played a role in several previous studies. The probability that the verbalized item is identical to the selection targets, the prime validity, was manipulated across experiments. The results showed a large capturing effect when the validity was higher than the chance level, but the capturing effect was not observed when the prime validity was lower than the chance level. The results suggested that contingency associated with the representation also modulated working memory-driven attentional capture. The content of working memory appears to capture attention as a default mode when there is no reason to avoid the influence of primes, and it is subject to strategic control based on the cognitive set. Taken together, the content of working memory does not automatically capture attention. The characteristics of the representation (the attentional tag, accessibility to pre-existing representation, and contingency related to attentional target) can modulate the working memory-driven attentional capture effect. Working memory and attention interact in a flexible manner.

並列關鍵字

working memory ； selective attention ； cognitive control ； attentional tag ； pre-existing representation ； contingency

參考文獻

Broadbent, D. E., Broadbent, M. H. P. (1981). Recency effects in visual memory. The Quarterly Journal of Experimental Psychology, 33, 1-15.

Ahmed, L. & de Fockert, J. W. (2012). Working memory load can both improve and impair selective attention: evidence from the Navon paradigm. Attention, Perception, & Psychophysics, 74, 1397-1405.

Anderson, B. A., Laurent, P. A., & Yantis, S. (2011). Value-driven attentional capture. Proceedings of the National Academy of Sciences of the United States of America, 108, 10367-10371.

Anllo-Vento, L., Luck, S. J., & Hillyard, S. A. (1998). Spatiotemporal dynamics of attention to color: Evidence from human electrophysiology. Human Brain Mapping, 6, 216-238.

Ansorge, U., Kiss, M., Worschech, F., & Eimer, M. (2011). The initial stage of visual selection is controlled by top-down task set: new ERP evidence. Attention, Perception, & Psychophysics, 73, 113-122.