JP4619582B2 - Information presentation apparatus and information presentation method using eye movement - Google Patents

Description

第2行は光点列の横幅を0.0625度とした時の結果で、視力0.6程度の被験者は消灯時間が0.4msまで破線を知覚し、視力1.0程度の被験者は消灯時間が0.3msまで破線を知覚した。第3行は光点列の横幅を0.03125度とした時の結果で、視力0.6程度の被験者は消灯時間が0.3msまで破線を知覚し、視力1.0程度の被験者は消灯時間が0.2msまで破線を知覚した。
【００２８】
光点列の横幅が小さく、視力がよくなる程少ない消灯時間でも破線が知覚された。視野角10度のサッケードを起こした場合、およそ20ms程度が情報提示に利用できるので、横方向に表示可能なピクセル数Ｘは、光点列の横幅0.03125度・視力1.0程度の場合でＸ=20(ms)/0.2(ms/ピクセル)=100ピクセル、光点列の横幅0.0625度・視力0.6程度の場合でＸ=20(ms)/0.4(ms/ピクセル)=50ピクセル程度となる。縦方向に表示可能なピクセル数については、光点列を何行配置するかによって調節が可能であるが、本研究においては最大7行の光点列で情報提示を行ったが、十分提示可能であった。本情報提示手法をディスプレイと考えると、本実験を行った設定では、縦7ピクセル、横50〜100ピクセル程度の情報提示能力を持ったディスプレイと考えられる。
【００２９】
次に本実験において、どのような時に破線が知覚されなくなるか、その物理的条件について考察する。図１１左に示すように、光点列の横幅をＡ(度)、消灯時間をｔｍ(秒)、消灯している間の眼球運動の速度をＶ(度/秒)とすると、光点列は消灯している間に網膜上をＶ×ｔｍ移動する。その時、大きさｄ=V ×tm−Ａの隙間が物理的に存在するが、その隙間が視認されると破線が知覚される。この隙間は、破線の線が存在する部分に比べて小さく、この隙間が視認できるか調べるということは、ランドルト環によって計測されている最小可読閾の視力を求めているの同等と考えられ、その視力を具体的に求めると表２のようになる。
【表２】

表２の第1行は消灯時間ｔｍ、第2行は消灯時間のうちに光点列が網膜上を移動する視野角αであり、この値は、視野角10度のサッケードにおいて情報提示に使用できる視野角は約5度なので、サッケード中の情報提示可能な時間をＴとすると、α=5度×ｔｍ／Ｔによって求められる。第3行は物理的に存在する隙間ｄであり、αから光点列の横幅A(0.0625度)を引いた値である。第4行は隙間ｄを視認できる視力である。この視力以上ないと、隙間を視認することはできない。第5行、第6行は異なる光点列の横幅(0.03125度)で隙間、視力を求めたものである。また、第3行、第5行で値が負になっているのは、図１１右のようにｄ＝Ｖ×ｔｍ−Ａ＜０となり隙間が物理的に存在しなくなった状態であり、この時破線は知覚され得ない。
【００３０】
表２の消灯時間と視力の関係に着目すると、光点列の横幅が0.0625度の時(第3行、第4行)、視力が0.6程度だと消灯時間が0.4msまで隙間が物理的に視認可能であり、視力が1.0程度だと消灯時間が0.4msもしくは0.3msまで隙間が物理的に視認可能である。光点列の横幅が0.03125度の時(第5行、第6行)、視力が0.6だと消灯時間が0.3msまで隙間が物理的に視認可能であり、視力が1.0程度だと消灯時間が0.2msまで隙間が物理的に視認可能である。これは、表１の実験結果とよく合致しており、本手法の情報提示能力を考える場合には、物理的制約が支配的であることがわかった。また、視力はサッケード中、低下するという報告があるが、残像の視認については、あまり低下は見られないようである。
【００３１】
本実験においては、光点列の横幅を2通り用意して実験を行ったが、理論的には、光点列の横幅を2本の直線のずれを見分けることのできる最小のずれ(副尺視力)の大きさ2"(0.00056度)とし、視認できる最小の隙間の大きさを、人間の視力限界、視力2.0の場合まで、消灯時間を短くすることが可能である。視力2.0の人間が視認できる最小の隙間はｄ＝0.00833度であり、ｄ＝5度 ×ｔｍ／Ｔ−Ａの関係式(視野角10度のサッケード)に、情報提示に利用できる時間Ｔ＝20ms、光点列の横幅Ａ＝0.00056、を代入してｔｍを求めると、ｔｍ＝0.0356msとなり、その時、横方向に表示可能なピクセル数Ｘ＝Ｔ／ｔｍ=562ピクセルとなる。
【００３２】
つまり理論的には、10度のサッケード1回でVGA(横方向640ピクセル)の解像度を持ったディスプレイの9割程度の2次元情報を提示可能であるということがわかった。しかし、より多くの人に情報を提示するためには、被提示者の視力は0.6程度に設定すべきであるため、表示可能なピクセル数は上記の計算より少なくなる。次に、視力0.6で同様の計算を行うと、副尺視力も変化して、光点列の横幅Ａ＝0.00185度となり、ｔｍ＝0.119msと計算される。この時、横方向に表示可能なピクセル数Ｘ＝168ピクセルとなる。この、より汎用的な条件設定では、VGAの1/4程度の2次元情報を提示可能であり、英数字を提示した場合には、英数字1文字の横幅を5ピクセルとすると、30文字程度が提示可能である。ただし、英数字などの文字を提示する場合には、多くの情報を物理的に提示できたとしても、1度の残像で認識できる文字の数は限られておりそれに関する限界も考慮する必要がある。
【００３３】
具体的設計論について述べる。本手法を用いて実際に情報提示を行う場合、どのように各パラメータ(光点列の光るタイミング、サッケードの大きさ、注視点・視標間の距離等)を決定すればよいか、これまでの議論に基づいて設計の指針を示す。まず、サッケード中の眼球運動の速度と経過時間の関係を考えると図１２のグラフのようになる。このグラフは縦軸に眼球の運動速度、横軸に経過時間を取り、サッケードが起こり始めてから、サッケードが終わるまでを示したものである。サッケードが起こり始めた時間をｔ＝０とした。グラフにあるように、サッケードが起こり始めてから眼球の運動速度は上昇し、あるピークを迎え、減少し始め、サッケード終了とともに0になる。本手法によって情報提示を行うには、眼球運動によって残像が起こらなければいけないので、眼球の運動速度がある一定速度(図１２のグラフ縦軸Ｖ_０)を超えている間のみ、情報提示が可能となる。サッケードが起きてから初めて情報提示可能な眼球速度になる時間をｔ_１とし、眼球速度がピークを迎え、情報提示が不可能になる時間をｔ_２とする。情報提示に使用できる時間Ｔ＝ｔ_２−ｔ_１である。この時、空間における1ピクセルを提示するための最小単位時間をｔｍとすると、横方向に表示可能なピクセル数Ｘ＝Ｔ／ｔｍによって決定される。以上の議論を逆に辿ると、本情報提示手法を利用して情報提示を行う場合、提示したい情報のピクセル数Ｘが決定すれば、その情報を提示するために必要な時間Ｔが決定し、その時のサッケードの大きさθも決定する。また、図１３にあるように、注視点・視標間の距離Ｄは注視点・被情報提示者間距離Ｌとサッケードの大きさθを使って、Ｄ＝Ｌ×ｔａｎθと表される。
【００３４】
具体的な設計例として、1m手前から視力0.6程度の人に対して、15文字の英数字を提示することを考える。1文字の横方向を5ピクセルと考えると、表示に必要な横方向のピクセル数Ｘ＝15(文字)× 5(ピクセル/文字)＝75ピクセルなので、1ピクセル提示するための最小単位時間ｔｍを0.4msとすると、情報提示のために必要な時間Ｔ＝75(ピクセル)×0.4(ms/ピクセル)＝30msとなる。ただし、tm＝0.4msという値は、前章の実験において、被験者の視力が0.6程度、光点列の横幅が0.0625度の時の値である。次に、情報提示に利用できる時間が30ms以上となる。
【００３５】
サッケードの視野角θを決定するが、予備実験においてサッケードの大きさと情報提示に利用可能な時間の関係を調べた対応表を表３に示す。
【表３】

表によると、情報提示に利用できる時間が30ms以上となるサッケードの視野角θは20度の場合である。サッケードの視野角θが決定されれば、注視点・視標の位置関係も決定される。眼球と注視点の距離Ｌが1mの時、注視点と視標の距離Ｄ＝Ｌ×ｔａｎθ＝36cmとなる。具体的構成イメージ図を図１４左に示す。また、その時の注視点、視標、光点列の光るタイミングチャートを図１４右に示す。はじめに、注視点が点灯し、視線を注視点に向けさせる。次に、注視点が消灯してから、200ms経過後に視標を点灯させ、潜時が100ms程度で安定しているエクスプレスサッケードを誘発させる。眼球は100ms程度の潜時の後、視標に向かってサッケードを開始する。視野角20度のサッケードは約80msであり、サッケード中に情報提示が可能な時間は約30msである。その情報提示可能な30msの間に、0.4msずつ光点列を端から文字をスキャンするパターンで点灯させ、文字を提示する(図１４右の矢印の間)。ただしこの図１４の場合、眼球が右に動いているので、文字を右からスキャンする形で光点列を点灯させる。
【００３６】
本明細書ではエクスプレスサッケードという高速で時間的に安定した眼球運動に着目することにより、簡便で汎用的な情報提示手法の提案を行った。この情報提示手法を使用した場合、1度のサッケードで提示可能な情報量は、VGAディスプレイの1/4程度、英数字にして30文字程度であることがわかった。また、本手法を使用して安定した情報提示を行うための条件として、注視点・視標の光るタイミング、注視点・視標の光る明るさの2点に着目し、それぞれ安定した情報提示のための条件を特定した。次いで、本手法を使って情報提示を行う場合、どのように提示系を設計すればよいのか設計の指針を示した。
【００３７】
本手法の特徴を生かした応用例として、いくつかの方向性が考えられる。
まず第1として、本手法は光点2つと光点列が1列あれば情報提示が可能であるので、本手法を使用したデバイスは小型化が期待でき、図１５に示すようにウェアラブル情報提示デバイスに応用可能である。ウェアラブルデバイスに実装する場合、眼球運動を簡単に測定可能であれば、それにあわせて確実に情報提示が可能となる。
【００３８】
第2として、本手法はデバイスの無い空間にも情報提示することが可能なので、普段生活している街中でも光点を両端の建物に、光点列を片方の光点の近くに配置することによって空中に情報を提示することも可能である。ただし、前章でも議論したように、屋外で本手法を使用する場合は、夕方から夜にかけてなど、やや暗い環境において使用したほうが、より効果的な情報提示が可能となる。
【００３９】
第3としては、光点や光点列は、コンピュータの電源等すでに何らかの装置に組み込まれているので、本手法を利用すると、それらすでにあるリソースを組み合わせることによって情報提示することも可能となる。
【００４０】
そして第4に、本手法は人間の眼球運動を利用しているので、眼球運動を起こした人のみ情報を見ることが可能であり、情報提示の選択性がある。この特徴を考えると、多くの情報を選択的に提示することが必要となるユビキタスコンピューティング等への応用も考えられる。本手法を使って伝えられる情報量は1度に英単語数語程度であるが、ユビキタスコンピューティングにおいて個々のユニットの状態を提示することは十分可能である。図１６はユビキタスコンピューティング、具体的には美術館等の展示における個々の情報（作者、年代、値段等）を提示する場合の応用例である。今後、ますます多くの情報が発信される中で、情報の全てを垂れ流すのではなく、情報を見たい人にのみ、情報を提示するということが重要となり、本手法の特徴が活かされると考えられる。
【図面の簡単な説明】
【図１】残像によって線が見えるのを示す図であり、（ａ）は光点の移動による残像、（ｂ）は眼球運動による残像である。
【図２】 光点列による情報提示を示す図であり、（ａ）は光点列の移動と発光パターンの変化による記号の提示、（ｂ）は眼球運動と光点列の発光パターンの変化による記号の提示である。
【図３】本発明の基本的構成を示す図である。
【図４】注視点・視標の光るタイミングと眼球の運動を表したタイムチャートである。
【図５】サッケードの種類と代表的な誘発刺激を示す図である。
【図６】注視点・視表の明るさとエクスプレスサッケードの誘発頻度の関係を調べる実験の実験条件の概要を示す図である。
【図７】明るさ実験のタイムチャートである。
【図８】明るさと誘発頻度の関係の実験結果を示す図である。
【図９】横方向に表示可能なピクセル数の算出を示す図である。
【図１０】解像度実験のタイムチャートである。
【図１１】破線が知覚される場合を示す図である。
【図１２】眼球速度と時間の関係を示す図である。
【図１３】位置関係の決定を示す図である。
【図１４】具体的な設計例を示す図である。
【図１５】本発明のウェアブルへの応用を示す図である。
【図１６】本発明のユピキタスへの応用を示す図である。
【図１７】提示される情報の一例を示す図である。
【図１８】本発明の実験装置の概略図である。[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a display and information presentation method using eye movement, and more particularly to a display and information presentation method using saccade (impulsive eye movement). For example, the information presentation method according to the present invention is expected to be used by a parasite human, which is a wearing system that works as a symbiotic information subject.
[0002]
[Prior art]
Conventionally, information is presented by moving a light spot group. Specifically, there is known a method of presenting information such as “inquiry” or “stop” by manually shaking a signal lamp having a light spot group.
[0003]
The present invention focuses on eye movement and presents information, which is different from the conventional one. Eye movement (saccade) is used in the medical field such as examination of brain disease, for example, but there is no technique for presenting information by paying attention to eye movement.
[0004]
[Problems to be solved by the invention]
The present invention was devised based on a completely new idea that has not existed in the past, and it is simple and versatile by focusing on human eye movements, particularly saccades, which are fast and time-stable eye movements. The purpose is to provide an information presentation technique.
[0005]
[Means for Solving the Problems]
The technical means employed by the present invention includes a gazing point, a target, and a light spot disposed between the gazing point and the target, and performs eye movement from the gazing point to the target. In addition to inducing, the light spot is emitted in accordance with the eye movement, and information is presented. Preferably, the light spot is a light spot group composed of a plurality of light spots, and more preferably, the light spot group is a line of light spots. Information can be presented by changing the light emission pattern of the light spot group over time according to the eye movement.
[0006]
When the light spot moves at a certain speed as shown in FIG. 1A, the locus appears as a continuous line due to the afterimage, but the light spot does not move as shown in FIG. 1B. Even if the eyeball moves at a certain speed or more, a line can be seen by the same afterimage effect. The present invention utilizes this fact. As shown in FIG. 2A, information such as characters and symbols can be presented by moving the light spots in a line and changing the light emission pattern of the light spots in the middle of the movement. . Therefore, by using the principle of FIG. 1B, it is possible to present information such as characters and symbols by changing the light emission pattern of the light spot sequence with time during the movement of the eyeball. FIG. 2 (b)).
[0007]
The gazing point and the target are preferably light spots. The gaze point, the index, and the light spot array are configured by LEDs in one preferred embodiment, but are not limited thereto, and may be appropriately adopted from other known light sources.
[0008]
The arrangement mode of the light spot group is not limited and does not necessarily have to be arranged in a linear form. However, in one most preferable aspect, the light spot group is a single light spot line, and a plurality of rows that are continuous in the vertical direction are arranged. I have. If there is only one light spot array, the control of the light spots constituting the light spot array is easier than the control of the light spots of a plurality of light spot arrays. Don't get in the way. Further, in the present invention, the information having one light spot is not excluded, but the perceived information is a broken line or a straight line, so that the information amount is small.
[0009]
In one preferred embodiment, eye movement (saccade) is triggered by first illuminating the gazing point, and then stopping illuminating the gazing point and causing the optotype to illuminate. With regard to the timing of stopping the light emission of the gazing point and the light emission of the visual target, in one preferred embodiment, a certain amount of time (200 ms) is taken until the light emission of the visual target stops when the light emission of the gazing point stops. Therefore, the expression “stop the light emission of the gaze point and light the target” includes the light emission of the target after the light emission of the gaze point is stopped (the stop of the gaze point and the light emission of the target simultaneously). Not what you do).
[0010]
In addition, one great advantage of the present invention is that information can be presented with fewer light spots than the spatially required display points. That is, in order to present information as shown in FIG. 17, usually 25 light spots are required. According to the present invention, for example, similar information can be presented by moving a light spot column (5 light spots) in 5 rows and 1 column horizontally. Therefore, according to the present invention, the apparatus can be reduced in size, and can be realized if there are two light spots and one light spot array, and resources that already exist can be used. It is possible to selectively present information that information can be presented only to a person who has caused eye movement.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
[Outline of Information Presentation Device and Method]
The basic configuration of the present invention will be described. As shown in FIG. 3A, the apparatus according to the present invention includes a gazing point, a visual target, and a light spot array disposed between the gazing point and the visual target. In this case, a single column of light spots composed of three (three rows) of light spots continuous in the vertical direction is employed. FIG. 18 is a schematic diagram of an experimental apparatus, in which a panel 4 is mounted at a central portion in the left-right direction of the substrate 5, and a plurality of light spots 3 are arranged on the panel 4 in the posterior direction. On the substrate 5, a gazing point 1 and a target 2 using LEDs are arranged so as to sandwich the panel 4. The gazing point 1, the target 2 and the light spot 3 are respectively provided with wirings not shown, and the timing of light emission at each point is controlled by an IC not shown. In the experiment, only a single vertical light spot array in the center of the panel was used. In the embodiment, the gaze point and the target are arranged in the left-right direction, but the arrangement mode of the gaze point and the target is not limited to this, and the gaze point and the target are arranged in the vertical direction or the oblique direction. It may be provided. In order to present information using eye movement, eye movement is induced on the information presentation side, and a light spot sequence is emitted in accordance with the timing to present information. In addition, in order to cause an afterimage by eye movement, high-speed eye movement is required. Therefore, a saccade (movement speed of 300 to 500 deg / sec) that is a high-speed movement component of tracking eye movement is induced.
[0012]
In practice, as shown in FIG. 3 (a), first, the point of interest is illuminated and attention is directed to the point of interest. Next, the gaze point is turned off and the target is lit. By doing so, the line of sight is shifted from the point of sight to the target and a saccade is induced. During the saccade, the light spot sequence is caused to emit light, and information is presented as shown in FIG.
[0013]
FIG. 4 is a time chart showing the timing at which the gazing point / target shines and the movement of the eyeball. The thick solid line portion represents the time zone when the gazing point / target is shining, and the dotted line represents the movement of the eyeball. First, the gazing point is shining and the eyeball is facing the direction of the gazing point. Next, the gaze point disappears and the target glows. At this time, the eyeball does not start to move with respect to the target at the moment when the target shines, but starts to move after a certain period of time (latency). While the eyeball starts to move and exceeds a certain speed, the light spot sequence is changed over time during the time indicated by the arrow in FIG. 4 to present information. Normally, when a saccade with a viewing angle of 10 degrees is generated, about 20 ms to 30 ms can be used for information presentation.
[0014]
In the method according to the present invention, while the eyeball is rotating, the light spot is lit, changed, and information is presented. Therefore, the timing at which the light spot row is lit according to the eyeball movement is important. . If the time required for the saccade is assumed to be constant, it is necessary to induce a saccade in which the time until the eyeball starts moving after the target shines and the latency is stable in time. In other words, in order to present information stably and reproducibly, the more stable the time from when a stimulus is presented to when a saccade occurs, the clearer is when the light spot should be emitted. The fact that the timing of shining the dots is stable means that stable information presentation is possible.
[0015]
[Saccade type]
In the method according to the present invention, a saccade is induced by sequentially illuminating a gazing point and a target, but there are several types of saccades having different characteristics such as latency, maximum speed, and gain. The types of these saccades are different depending on the timing of gazing at the point of interest and the target. FIG. 5 shows the main saccade types and the light emission timings of typical gazing points and targets that induce them.
[0016]
As a major saccade, a predictive saccade that was caused by predicting the position of the target in advance (Fig. 5 (a)); Occurrence saccade (Fig. 5 (b)), gaze point disappears and indicator is light, saccade generated outside visual guidance induced by it (Fig. 5 (c)), indicator after a certain time after gaze point disappears There is a glowing express saccade (FIG. 5 (d)).
[0017]
Since the predictive saccade predicts in advance the position where the target appears, the latency (in this case, the time after the point of gaze disappears) varies as short as 20 to 80 ms. The saccade generated within the visual guidance has a latency of about 150 ms to 200 ms, and the saccade generated outside the visual guidance has a latency of about 150 ms to 200 ms. The maximum speed is faster and the gain is larger than the saccade generated within the visual guidance. The express saccade has a latency of around 100 ms and is shorter than the two saccades.
[0018]
Among them, it is known that the latency is most stable in terms of express saccade. In particular, since the latency is most stable under the condition that the interval from the point of gazing point disappears until the target shines is 200 ms, this method uses an express saccade with the interval from the point of sight point disappearing until the target shines as 200 ms. Induced. Furthermore, it is known that the latency of the express saccade becomes more stable by repeating.
[0019]
[Gaze point / target brightness]
The light emission timing of the gazing point / target is 200 ms from when the gazing point disappears until the target shines, but this does not always induce an express saccade, and it depends on the brightness, size, etc. of the gazing point. The frequency is different. Therefore, an experiment was conducted to investigate the relationship between the brightness of the gazing point and the target and the induction frequency of express saccades. We examined how often express saccade occurs by changing the brightness of the point of interest and the target.
[0020]
An outline of the experimental conditions is shown in FIG. Experiment is 0.02 cd / m brightness ² In the dark room, the subject sat down so that the eyeball of one eye was located at a position 45 cm from the gazing point, and caused a saccade with one eye. The distance between the gazing point and the target was 8 cm, and the viewing angle was 10 degrees. This is because there are about 10 saccades seen every day. LEDs were used for presentation of the gazing point, target, and light spot sequence. The size of the gazing point and the target is a viewing angle of 1.3 degrees and the brightness is bright (0.8 cd / m ² ), Middle (0.3 cd / m ² ), Dark (0.1 cd / m ² ), The experiment was performed with a total of nine combinations of brightness. The size of each light spot in the light spot array is a viewing angle of 0.25 degrees, and the brightness is 14 cd / m. ² ).
[0021]
A time chart of the experiment is shown in FIG. The gazing point is light, and the subject looks at the gazing point. Next, the gazing point disappears, and the target on either the left or right side shines randomly after 200 ms elapses. The reason for the randomness is to prevent an expected saccade from occurring due to repeated experiments.
[0022]
A light spot column of 3 rows and 1 column is arranged between the gazing point and the right target, and after the right target shines, one light spot is turned on for 40 ms to 100 ms, and two lights are emitted for 100 ms to 160 ms. Illuminate the dots and illuminate 3 light spots for 160 ms to 220 ms. This means that an expected saccade with a latency of 40 ms to 100 ms occurred when the afterimage of the line was seen at position 1; an express saccade with a latency of 100 ms to 160 ms when seen at position 2; It is a visually-guided regular saccade at 160 to 220 ms. 50 trials were performed for each combination of brightness.
[0023]
The experimental results of one of the subjects are shown in FIG. The dark gazing point shows that the express saccade occurs more frequently. This is probably because if the gazing point is too bright, the disappearance of the gazing point itself may induce a predictive saccade. In addition, no difference appears in the frequency of occurrence of express saccades due to variations in the brightness of the target. Other subjects show similar trends. From this experiment, it was found that, in order to induce express saccades with a high frequency, the point of sight is not too bright and induces more express saccades.
[0024]
[Information presentation ability]
An experiment was conducted to investigate the amount of information that can be presented in a single saccade when this method is used for information presentation. Here, the information presentation ability means how many pixels of information can be presented at one time in the space when this method is used. Specifically, as shown in FIG. The number of displayable pixels X can be calculated by X = T / tm, where T is the time available for information presentation in the saccade and tm is the minimum presentation time for presenting one pixel in space. Since the time T that can be used for presenting information during a saccade is known, the minimum presentation time tm was determined by experiment.
[0025]
Experimental conditions will be described. The experiment was conducted in the same dark room as the brightness experiment, and the same LED was used. The positional relationship between the subject, the gazing point, and the target is the same as in FIG. 6, but the distance between the subject and the gazing point is 92 cm, and the distance between the gazing point and the target is 16 cm. The viewing angle from the gazing point to the target is 10 degrees. Gaze point and target brightness are 0.1 cd / m ² , The target is 0.8cd / m ² It was. For the light spot array, only the light spot at the center (2) in FIG. 6 was used. The width of the light spot array was 1mm and 0.5mm, and two viewing angles of 0.0625 degrees and 0.03125 degrees were prepared.
[0026]
The experimental method will be described. FIG. 10 shows a time chart of the timing at which the gazing point, the target, and the light spot sequence shine in this experiment. First, the gazing point shines and the subject turns his gaze to the gazing point. Next, the point of interest disappears, and after 200 ms, either the left or right target lights up randomly. Then, while the saccade is occurring from the gazing point toward the target, the light spot in the center of the light spot row (2) in FIG. 10 blinks at a certain time interval. Then, a broken line is perceived as indicated by an arrow at the bottom of FIG. Therefore, as the interval at which the light spot sequence blinks is gradually reduced, the interval between the broken lines gradually decreases, and finally a straight line is perceived. Since the time interval that is the boundary between the broken line and the straight line can be considered as the minimum unit time for information presentation, it was determined how much the light line blinking interval should be perceived as a straight line. Specifically, we started with a pattern of 2ms on and 2ms off, and reduced the extinguishing time in 0.1ms increments, and identified the boundary by the up-and-down method when perceived as a straight line. However, the width of the light spot array is 0.0625 degrees and 0.03125 degrees as the viewing angle, and the subjects are also experimented with two groups with a visual acuity of about 0.6 and a visual acuity of about 1.0. The experiment was conducted.
[0027]
The experimental results will be described. The boundary between the straight line and the broken line specified by the vertical method varied depending on the lateral width and visual acuity of the light spot array. Table 1 shows the minimum turn-off time during which a broken line can be perceived.
[Table 1]

The second row is the result when the width of the light spot column is 0.0625 degrees, and subjects with a visual acuity of about 0.6 perceive a broken line until the extinguishing time is 0.4 ms, and subjects with a visual acuity of about 1.0 show a broken line until the extinguishing time is 0.3 ms. Perceived. The third row is the result when the width of the light spot sequence is 0.03125 degrees. Perceived.
[0028]
The broken line was perceived even when the light spot sequence was small in width and the extinction time was short enough to improve visual acuity. When a saccade with a viewing angle of 10 degrees occurs, about 20 ms can be used for information presentation, so the number of pixels X that can be displayed in the horizontal direction is X = 20 when the horizontal width of the light spot row is 0.03125 degrees and the visual acuity is about 1.0. In the case where (ms) /0.2 (ms / pixel) = 100 pixels, the horizontal width of the light spot array is 0.0625 degrees, and the visual acuity is about 0.6, X = 20 (ms) /0.4 (ms / pixel) = about 50 pixels. The number of pixels that can be displayed in the vertical direction can be adjusted according to how many rows of light spots are arranged. In this study, information was presented with a maximum of 7 rows of light spots, but it could be presented sufficiently. Met. If this information presentation method is considered as a display, in the setting in which this experiment was performed, it is considered that the display has an information presentation capability of about 7 pixels vertically and 50-100 pixels horizontally.
[0029]
Next, in this experiment, we will consider the physical condition of when the broken line is not perceived. As shown in the left of FIG. 11, when the horizontal width of the light spot sequence is A (degrees), the extinction time is tm (seconds), and the speed of eye movement while the light is extinguished is V (degrees / second), Moves on the retina by V × tm while the light is off. At that time, a gap of size d = V × tm−A physically exists, but a broken line is perceived when the gap is visually recognized. This gap is small compared to the part where the broken line exists, and examining whether this gap is visible is considered equivalent to seeking the minimum readable threshold visual acuity measured by the Landolt ring. Table 2 shows specific visual acuity.
[Table 2]

The first row in Table 2 is the turn-off time tm, and the second row is the viewing angle α in which the light spot sequence moves on the retina during the turn-off time. This value is used for information presentation in a saccade with a viewing angle of 10 degrees. Since the viewing angle that can be obtained is about 5 degrees, when the time during which information can be presented during the saccade is T, it is obtained by α = 5 degrees × tm / T. The third row is a physically existing gap d, which is a value obtained by subtracting the lateral width A (0.0625 degrees) of the light spot array from α. The fourth row is visual acuity that allows the gap d to be visually recognized. Without this visual acuity, the gap cannot be visually recognized. The fifth and sixth rows are obtained by calculating the gap and the visual acuity with the lateral width (0.03125 degrees) of different light spot sequences. Further, the values in the third and fifth rows are negative when d = V × tm−A <0 as shown in the right of FIG. 11 and the gap no longer physically exists. Sometimes broken lines cannot be perceived.
[0030]
Paying attention to the relationship between extinction time and visual acuity in Table 2, when the horizontal width of the light spot row is 0.0625 degrees (3rd and 4th rows), if the visual acuity is around 0.6, the gap is physically up to 0.4ms. When the eyesight is about 1.0, the gap can be visually recognized until the extinguishing time is 0.4 ms or 0.3 ms. When the width of the light spot row is 0.03125 degrees (5th and 6th rows), if the eyesight is 0.6, the gap is physically visible up to 0.3 ms, and if the eyesight is about 1.0, the light-off time The gap is physically visible up to 0.2 ms. This agrees well with the experimental results in Table 1, and it was found that physical constraints are dominant when considering the information presentation ability of this method. Moreover, although it has been reported that the visual acuity decreases during the saccade, it seems that there is not much reduction in the visual recognition of the afterimage.
[0031]
In this experiment, the experiment was carried out with two horizontal widths of the light spot array.Theoretically, the horizontal width of the light spot array is the minimum deviation (sub-scale) that can distinguish the deviation of the two straight lines. It is possible to shorten the turn-off time until the size of the smallest gap that can be visually recognized is the limit of human eyesight, that is, the eyesight of 2.0. The smallest visible gap is d = 0.00833 degrees, and d = 5 degrees × tm / TA relational expression (saccade with a viewing angle of 10 degrees), time T = 20 ms available for information presentation, When tm is calculated by substituting the horizontal width A = 0.00056, tm = 0.0356 ms, and at that time, the number of pixels that can be displayed in the horizontal direction X = T / tm = 562 pixels.
[0032]
In other words, theoretically, it was found that 90% of 2D information of a display with VGA resolution (640 pixels in the horizontal direction) can be presented with a single 10 degree saccade. However, in order to present information to more people, the visual acuity of the person to be presented should be set to about 0.6, and therefore the number of pixels that can be displayed is smaller than the above calculation. Next, if the same calculation is performed with a visual acuity of 0.6, the vernier visual acuity also changes, and the light spot row width A = 0.00185 degrees, and tm = 0.119 ms is calculated. At this time, the number of pixels that can be displayed in the horizontal direction is X = 168 pixels. With this more general-purpose condition setting, it is possible to present two-dimensional information about 1/4 of VGA, and when alphanumeric characters are presented, if the width of one alphanumeric character is 5 pixels, about 30 characters Can be presented. However, when presenting characters such as alphanumeric characters, even if a lot of information can be physically presented, the number of characters that can be recognized at one afterimage is limited, and it is necessary to consider its limitations. is there.
[0033]
The concrete design theory is described. When actually presenting information using this method, how to determine each parameter (timing of light spot sequence, saccade size, distance between gazing point and target), etc. Design guidelines are presented based on the above discussion. First, considering the relationship between the speed of eye movement during the saccade and the elapsed time, the graph in FIG. 12 is obtained. This graph shows the movement speed of the eyeball on the vertical axis and the elapsed time on the horizontal axis, and shows the time from the start of saccade to the end of saccade. The time when the saccade began to occur was set to t = 0. As shown in the graph, the movement speed of the eyeball increases after the saccade begins to occur, reaches a certain peak, begins to decrease, and becomes 0 when the saccade ends. In order to present information by this method, afterimages must occur due to eye movements, so that the movement speed of the eyeballs is a certain constant speed (the graph vertical axis V in FIG. 12). ₀ It is possible to present information only while it exceeds (). The time to reach the eye speed at which information can be presented for the first time after a saccade occurs ₁ T is the time when the eyeball speed reaches its peak and information cannot be presented. ₂ And Time T = t that can be used for information presentation ₂ -T ₁ It is. At this time, if the minimum unit time for presenting one pixel in the space is tm, the number of pixels that can be displayed in the horizontal direction is determined by X = T / tm. When the above discussion is traced back, when information presentation is performed using this information presentation method, if the number of pixels X of information to be presented is determined, the time T required to present the information is determined, The saccade size θ at that time is also determined. As shown in FIG. 13, the distance D between the gazing point and the target is expressed as D = L × tan θ using the gazing point / information-presenter distance L and the saccade size θ.
[0034]
As a specific design example, consider presenting 15 alphanumeric characters to a person with a visual acuity of about 0.6 from 1 meter before. Assuming that the horizontal direction of one character is 5 pixels, the number of horizontal pixels necessary for display X = 15 (characters) × 5 (pixels / character) = 75 pixels, so the minimum unit time tm for displaying one pixel is Assuming 0.4 ms, the time required for information presentation T = 75 (pixels) × 0.4 (ms / pixel) = 30 ms. However, the value tm = 0.4 ms is the value when the subject's visual acuity is about 0.6 and the width of the light spot array is 0.0625 degrees in the experiment of the previous chapter. Next, the time available for information presentation is 30 ms or more.
[0035]
The saccade viewing angle θ is determined. Table 3 shows a correspondence table in which the relationship between the saccade size and the time available for information presentation was examined in a preliminary experiment.
[Table 3]

According to the table, the viewing angle θ of the saccade where the time available for information presentation is 30 ms or more is 20 degrees. If the viewing angle θ of the saccade is determined, the positional relationship between the gazing point and the target is also determined. When the distance L between the eyeball and the gazing point is 1 m, the distance between the gazing point and the target is D = L × tan θ = 36 cm. A specific configuration image is shown on the left of FIG. In addition, a timing chart at which the gazing point, the target, and the light spot sequence are illuminated is shown on the right side of FIG. First, the gazing point is turned on and the line of sight is directed toward the gazing point. Next, after the gazing point is turned off, the target is turned on after 200 ms has elapsed, and an express saccade that is stable with a latency of about 100 ms is induced. The eyeball begins to saccade toward the target after a latency of about 100 ms. The saccade with a viewing angle of 20 degrees is about 80 ms, and the time during which information can be presented during the saccade is about 30 ms. During the 30 ms during which the information can be presented, the light spot string is lit in a pattern of scanning characters from the end by 0.4 ms, and the characters are presented (between the arrows on the right in FIG. 14). However, in the case of FIG. 14, since the eyeball is moving to the right, the light spot sequence is turned on in such a manner that characters are scanned from the right.
[0036]
In this specification, a simple and general-purpose information presentation method has been proposed by paying attention to the fast and temporally stable eye movement called express saccade. When this information presentation method was used, it was found that the amount of information that can be presented in one saccade is about 1/4 of the VGA display, and about 30 alphanumeric characters. In addition, as conditions for stable information presentation using this method, pay attention to two points: the timing at which the gazing point / target shines, and the brightness at which the gazing point / target shines. Identified the conditions for. Then, when information is presented using this method, a design guideline is shown how to design the presentation system.
[0037]
There are several possible directions as application examples that take advantage of the features of this method.
First of all, since this method can present information if there are two light spots and one light spot sequence, a device using this method can be expected to be miniaturized, and wearable information presentation as shown in FIG. Applicable to devices. When mounted on a wearable device, if eye movements can be easily measured, information can be reliably presented accordingly.
[0038]
Second, since this method can present information even in a space where there is no device, the light spots should be placed in the buildings at both ends and the light spot array near one of the light spots in the city where you normally live. It is also possible to present information in the air. However, as discussed in the previous chapter, when using this method outdoors, it is possible to present information more effectively when used in a slightly dark environment, such as from evening to night.
[0039]
Thirdly, since the light spot and the light spot sequence are already incorporated in some device such as a computer power supply, using this method makes it possible to present information by combining these existing resources.
[0040]
Fourthly, since this method uses human eye movements, only the person who has caused eye movements can view information, and there is selectivity of information presentation. Considering this feature, it can be applied to ubiquitous computing or the like that requires selective presentation of a large amount of information. The amount of information transmitted using this method is only a few English words at a time, but it is possible to present the status of individual units in ubiquitous computing. FIG. 16 shows an application example in the case of presenting individual information (author, age, price, etc.) in an exhibition such as a ubiquitous computing, such as a museum. In the future, as more and more information is transmitted, it will be important to present information only to those who want to see the information, rather than dripping down all of the information. Conceivable.
[Brief description of the drawings]
FIGS. 1A and 1B are diagrams illustrating a line visible by an afterimage, where FIG. 1A is an afterimage due to movement of a light spot, and FIG. 1B is an afterimage due to eye movement;
FIGS. 2A and 2B are diagrams showing information presentation by a light spot sequence, where FIG. 2A is a symbol presentation based on movement of the light spot train and a change in the light emission pattern, and FIG. This is the presentation of the symbol.
FIG. 3 is a diagram showing a basic configuration of the present invention.
FIG. 4 is a time chart showing the timing at which the gazing point / target shines and the movement of the eyeball.
FIG. 5 is a diagram showing saccade types and typical evoked stimuli.
FIG. 6 is a diagram showing an outline of experimental conditions of an experiment for examining the relationship between the brightness of a gazing point / visual table and the induction frequency of express saccades.
FIG. 7 is a time chart of a brightness experiment.
FIG. 8 is a diagram illustrating an experimental result of a relationship between brightness and induction frequency.
FIG. 9 is a diagram illustrating calculation of the number of pixels that can be displayed in the horizontal direction.
FIG. 10 is a time chart of a resolution experiment.
FIG. 11 is a diagram illustrating a case where a broken line is perceived.
FIG. 12 is a diagram showing the relationship between eyeball speed and time.
FIG. 13 is a diagram illustrating determination of a positional relationship.
FIG. 14 is a diagram showing a specific design example.
FIG. 15 is a diagram showing an application of the present invention to a wearable.
FIG. 16 is a diagram showing an application of the present invention to ubiquitous.
FIG. 17 is a diagram illustrating an example of presented information.
FIG. 18 is a schematic view of an experimental apparatus of the present invention.

Claims (8)

A gazing point consisting of light spots,
A target consisting of a light spot;
A light spot group disposed between the gazing point and the target;
A control IC that controls the timing of light emission of each light spot of the light spot group, the gazing point, the target, and
The control IC first causes the gazing point to emit light, and then stops the emission of the gazing point and causes the target to emit light, thereby causing a saccade to be induced from the gazing point to the target. An information presentation apparatus using eye movement, wherein information is presented by changing the light emission pattern of the light spot group over time according to the saccade. 2. The information presentation apparatus according to claim 1, wherein the light spot group is one or a plurality of light spot arrays. The information presentation apparatus according to claim 1, wherein the saccade is an express saccade. 4. The information is presented according to claim 1, wherein the light emission pattern of the light spot group is changed with time, whereby the information is presented by a light spot having a smaller number of display points spatially necessary to present certain information. An information presentation device characterized by that. 5. The information presentation apparatus according to claim 1, wherein brightness of the target is made brighter than brightness of the gazing point. A light spot group is arranged between a gazing point composed of a light spot and a target composed of a light spot, and the timing of light emission of each light spot of the gazing point, the target, and the light point group is controlled. The control IC first causes the gazing point to emit light, and then stops the emission of the gazing point and causes the target to emit light to induce a saccade from the gazing point to the target. An information presentation method using eye movement, characterized in that information is presented by temporally changing the light emission pattern of the light spot group according to the saccade. 7. The information presentation method according to claim 6, wherein brightness of the target is made brighter than brightness of the gazing point. 8. The information presenting method according to claim 6, wherein the saccade is an express saccade.

Images