This paper describes a visual effect dictionary that can visually express images of words used as modifiers, such as adjectives, onomatopoeia by adding 3D objects with a visual effect. This visual effect dictionary links words and visual effects for the target objects. The same modifier can have different meanings depending on the target object thus, an intelligent algorithm to select the visual effects is required. This dictionary can be used as communication support for images that are difficult to express only by modifier words such as adjectives, onomatopoeia. When multiple regression analysis was conducted in the questionnaire results, it was found that there are appropriate visual effects and inadequate visual effects depending on nouns.