中心視野にテクスチャパタンを呈示する直前に高コントラストのマスク刺激を短時間呈示すると, 実際には何も呈示されていない周辺視野にもテクスチャパタンが広がり, テクスチャが視野全体を覆うように知覚される(充填)。筆者が発見したこの「マスク誘導充填(mask induced filling-out)」について, マスクと刺激の呈示位置関係を主な変数として生起要因を検討する実験を行った。
実験では, 中心視野のみにテクスチャを呈示する 「錯視刺激」と実際に周辺視野までテクスチャで覆った「全視野刺激」のいずれかをPC画面上に呈示した。実験参加者の課題は2種類の刺激のどちらが呈示されたかを二肢強制選択法で回答することであった。この際「錯視刺激」を「全視野刺激」と取り違えて報告する誤答率を錯視生起の指標とした。このパラダイムを用い, (実験1)刺激の大きさ, (実験2)マスクとテクスチャの呈示位置のずれ, (実験3)マスクとテクスチャの呈示眼の異同, の3変数を独立変数とした実験を行った。
この結果, 半径わずか視角2度の刺激を0.5秒呈示しただけで網膜偏心度視角20度の周辺視野まで充填が生じ, この現象が高速かつ広範囲に生じることがわかった(実験1)。また充填はマスクとテクスチャの輪郭間にある程度の隙間があっても生じるが, 刺激輪郭の網膜偏心度が小さくなると生じ難いことから, 充填の生起にはやはりテクスチャ面の輪郭をマスクすることが重要であることが示された(実験2)。さらに充填はマスクとテクスチャを左右別々の眼に呈示しても生じること, しかしその生起率は同眼呈示時よりも低くなることが示され, この現象の生起に両眼統合の前後2水準の処理が関わっていることが示唆された(実験3)。
本現象を1分間のデモンストレーションとしてまとめた動画は, 本年度の国際錯視コンテスト(Best Illusion of the Year 2017)に於いてTop10ファイナリストに選出された。
When a high-contrast mask stimulus is shown for a short time just before presenting the texture pattern in the central visual field, the texture pattern spreads to the peripheral, and the texture is perceived to cover the entire field of view. Regarding this "mask induced filling-out" discovered by the author, a series of experiments were conducted to examine the mechanism of illusion, especially by manipulating the relationship between the position of the mask and the texture stimulus.
In the experiment, the "illusion stimulus" in which texture was presented only in the central visual field, or the "full-field stimulus" in which the texture actually covered entire visual field, was presented on the PC screen. Participants answered which of the two types of stimuli was presented by the two-alternative-forced-choice method. The ratio of incorrect answer that mistaken "illusion stimulus" as "full-field stimulus" was used as an index of illusion occurrence. With this paradigm, we conducted three experiments with following three variables ; Experiment 1 : the stimulus size, Experiment 2 : gap between the mask and the texture, Experiment 3 : difference of presented eye of the mask and the texture.
As a result, I found that the texture stimulus which radius was only 2 degree in visual angle and presented only for 0.5 sec could produce the illusion, suggesting that the illusion occurs quite rapidly and extensively (Experiment 1). Also, the illusion disappeared when there was a wide (>2 deg visual angle) gap between the contours of the mask and the texture (Experiment 2). The results suggest that masking the contour of the textured surface is important for the occurrence of the illusion. Furthermore, the illusion occurred even if the mask and the texture stimulus were presented to the different eyes, but that the occurrence rate was lower when they were presented to the different eyes than when they were presented to the same eyes (Experiment 3). The results suggest that the loci of the illusion are both before and after the binocular integration.
The author made video footage that summarized this illusion as a one-minute demonstration. The video was selected as Top 10 finalists in the Best Illusion of the Year Contest 2017.