| 摘要: | 前額葉眼動區（FEF）與準備眼動、產生眼動、控制眼動有重要關聯。過去研究假設FEF處理視覺訊息的方式是透過「顯著性地圖」（Salience map），來決定眼動的目標。用以結合所見資訊與大腦決策，決定眼動目標。其表徵整合過後的視覺訊息，包括視知覺訊息（即由下往上的bottom-up歷程），和內在產生的視覺目標（即由上往下的top-down歷程）。因此本實驗利用視覺搜尋作業來探討由下往上訊息如何受干擾物的性質與特徵影響，以及由上往下的歷程中，前額葉眼動區扮演的角色。
本系列研究利用兩組實驗進行研究，一組操弄目標物與干擾物的相似性，一組操弄干擾物出現的時態，並使用跨顱磁刺激（TMS）與電刺激（tDCS）探究腦區與眼動行為之間的因果。所使用的行為指標是眼動反應時間與眼動彎曲率。前者代表眼動準備所需的時間，而後者選擇眼動該動往目標物或干擾物的競爭結果。當眼動成功的移往目標物，也就是成功的抑制動往干擾物的傾向，在神經元裡是對干擾物的反應抑制（distractor suppression），這樣的壓制程度多寡，便影響到眼動彎曲的程度。
第一組實驗，欲探究由下往上與由上往下兩歷程，在顯著性地圖的影響，進而反應在眼動行為。透過目標物與干擾物可能顏色相同或是形狀相同，我們發現當顏色相同的情況，眼動所需花費的反應時間較長，眼動軌跡遠離干擾物的幅度也較大；當施打tDCS在FEF後，反應時間一致的降低，但相同顏色的情況，遠離干擾物的幅度更大了。這說明了在顯著性地圖上，顏色比起形狀更有顯著性，也再證實FEF與顯著性地圖的關係。
第二組實驗，欲探究干擾物在目標物之前出現的不同時間點影響，進而影響對干擾物的反應抑制程度。實驗結果發現干擾物提前出現至少70毫秒，可以使得眼動往目標物的反應時間大幅降低，但眼動的彎曲程度並不受影響；TMS伴隨著干擾物出現時間施打後，TMS施打在FEF的情況，使得眼動遠離干擾物的情況減緩，也就是彎曲的較傾向干擾物，這意味著降低了干擾物的顯著程度，進而減少對干擾物的壓抑情況，但並沒有干擾眼動的反應時間。
第二組實驗同時探討後頂葉皮質區（PPC），過去研究指出PPC與FEF在視覺搜尋作業上的功能類似。實驗結果發現，PPC受TMS影響後增長了眼動反應時間，但眼動彎曲率不受影響。這意味著PPC與FEF在眼動準備歷程中有相似但不同的功能。PPC可能主要負責由下往上的訊息處理；而FEF則著重在由上往下的過程。
本碩論除了再次驗證了先前在FEF的顯著性地圖對眼動的影響，並提供了有力證據說明顏色比起形狀更有顯著性；同時透過操弄干擾物出現的時間，影響對干擾物的反應抑制程度，比較FEF與PPC的功能──前者著重處理由上往下的內在產生的視覺目標，後者則是處理由下往上的視知覺訊息。
The functional roles of the Frontal Eye Field (FEF) are preparation for saccades, saccade triggering, and controlling eye movements. Saccade goals are based on a salience map. This salience map is a convergence of both top-down (i.e. the decision for selection) and bottom-up (i.e. target and distractor features) processes. The current study examines the role of the FEF in these processes through the effects of transcranial direct current stimulation (tDCS) and transcranial magnetic stimulation (TMS). The distractor-related modulation of saccade latencies and trajectories are used as indicators. The saccade latencies are used to represent the saccade preparation time and the saccade trajectories represents the target selection and distractor suppression processes.
In the first series of experiments, the aim of the research was investigation of both top-down and bottom up processes and their relationship to the saliency map. The manipulation of target-distractor similarity and tDCS were used. In experiment 1-1, participants made reflexive saccades to a blue square target appearing with an accompanying distractor. The distractor was either a blue diamond or a green square, both of which share only one feature with the target. Behavioral results showed that the saccade latency was longer when the distractor was of the same color, implying that color has higher salience than shape.
In experiment 1-2, participants made voluntary saccades to peripheral locations specified by a central arrow-cue. The target was either a blue square or green diamond and was indicated on each trial. The distractors were the same as experiment 1-1, but the role of same color or same shape were defined by the target in each trial. The behavioral results showed no differences between these conditions. However, when tDCS was applied over participant’s rFEF, the saccades made curved more away from the distractor in the same color condition, and curved away less in same shape condition. This not only indicated the color has higher salience than shape, but also demonstrated a relationship between the saliency map hypothesis and FEF.
In experiment 2, the aim of the research was examining the distractor suppression role in FEF and comparing this with posterior parietal cortex (PPC). In this case, the appearance time of distractors was manipulated. The distractor and the TMS preceded the target by various lengths of time, i.e. short (70 ms), middle (150 ms), and long (300 ms). TMS was applied over either FEF or PPC, two critical areas for saccade generation. The results indicated that saccade latencies were significantly longer when distractors appeared slightly ahead of a target across all TMS conditions, but TMS over PPC increased saccade latency in all time conditions. Additionally, the initial slope of saccade curvatures decreased in the FEF TMS condition for all distractor time offsets. These results show that distractor suppression requires more than 70ms and also that the contingency between saccade latency and curvature was dissociated. Most importantly, evidence is provided for PPC and FEF involvement in saccade generation, but that this involvement is different, particularly in the case of distractor suppression.
The current study examined the degree to which distractor saliency modulates saccade curvature, and provides evidence that color has higher saliency than shape in this context. Increasing FEF activity by tDCS stimulation demonstrated its role in forming a saliency map for eye movement generation. Above all, differing roles of FEF and PPC are illustrated by the effects TMS in distractor suppression during the pre-saccadic period. FEF seems to have a role in top-down processes, and PPC a role in bottom-up processes. |
