Tests involving the previously discussed "Happy-face" anomaly, conducted in July 2005.
Fig. 1: "Happy-face" displaying its abilities in expected form. When text is written, it can add onto it in the next line. Three examples are listed: computer-generated text, text written with a mouse, and intentionally distorted text written with a mouse. Note that it copies the format of the text, even when the original is difficult to read.Fig. 2: A test involving text written using dots. "Happy-face" is able to copy this dot format.Fig. 3a: Another test involving text written using dots.Fig. 3b: Once the original text is written, the image is populated by identical random dots. Later tests found subjects unable to recognize this as text without seeing the isolated message first.Fig. 3c: The image after "Happy-face" added to it.Fig. 3d: Isolating only the original dots and the added dots, the "Happy-face" message seems to read ["huh?" -:-0]. In later conversation, it claimed not to viewed the original text before the addition of random dots. If this is true, this implies that it is able to add to text even when it is not clear that text exists.Fig. 4a: An image consisting solely of random dots. This image was shown to the "Happy-face" anomaly with the false explanation that there was another hidden message. However, it did not add to this image, and in later conversation expressed that it did not think text was present.Fig. 4b: An attempt to find text within the randomly generated image by isolating specific existing dots. A research attempts to write "Sup".Fig. 4c: The image after "Happy-face" added to it, this time by unisolating specific dots. It is unclear what this text says, although what may be a "hi" is visible.
