This is a past event
Perceptual Similarity, Visual Metaphor and Creativity
Researchers who study creativity in real-world situations have foundthat a primary hurdle facing human creativity is in stepping outsideof our habitual conceptual association. A few techniques have beensuggested to aid this process: making-the-familiar-strange andde-conceptualization, for example. The main objective of thesetechniques is to help the cognitive agent break the bond of conceptualassociation it has acquired culturally and through a lifetime ofexperiences. In this process, imagery, perceptual similarities, andvisual metaphors play a key role, as we will demonstrate using severalexamples.We are following a number of research veins to explore the role ofimagery and visual metaphor in creativity and cognition, and in thistalk I will present some of our results. In particular, we willpresent some experiments to compare and contrast mono-modal (text-textor image-image) vs. cross-modal (text-image) metaphors. We have alsoinvestigated how explicit imagery, presented before or after themetaphor, influences the comprehension of metaphor.We have argued, and demonstrated, in our past research, thatcomputer-based systems can be very helpful in stimulating creativity.For instance, we found that incorporating one familiar but unrelatedobject in a picture setting, or presenting unrelated pairs of objectsto a cognitive agent, and asking them to make a story out of it, or tomake sense of it in some other way, stimulates their creativity andimagination.In another line of research, we are exploring the hypothesis thatlow-level perceptual similarity — that is, similarity based onlow-level perceptual features like color, shape and texture — plays akey role in creation of novel conceptual associations. In otherwords, we are claiming that if the unrelated object that is introducedinto a picture or paired with another object bears some low-levelperceptual similarity with other objects in the picture, or with thepaired object, it is likely to be more effective in stimulatingcreativity than a random unrelated object. We will present here theresults of some of our preliminary experiments to explore thishypothesis.As the similarity with respect to low-level perceptual features isdetermined algorithmically using image-based pattern matching, ourapproach can be used to design more effective computer-basedcreativity-support systems. We will present an outline of sucharchitecture and mention a number of possible application domains forsuch systems.
- Speaker
- Dr Bipin Indurkhya
- Hosted by
- Frank Guerin
- Venue
- MT2