# GANalyze: Toward Visual Definitions of Cognitive Image Properties

## TL;DR

This paper presents GANalyze, a framework using GANs to generate and analyze images with varying cognitive properties like memorability, aesthetics, and emotional valence, revealing visual features underlying these attributes.

## Contribution

It introduces a novel GAN-based method to visualize and understand the visual properties associated with cognitive image attributes.

## Key findings

- GANalyze can generate images with controllable memorability, aesthetics, and emotional valence.
- Behavioral experiments confirm causal effects of manipulated image properties on human memory.
- The framework identifies visual features underlying cognitive image properties.

## Abstract

We introduce a framework that uses Generative Adversarial Networks (GANs) to study cognitive properties like memorability, aesthetics, and emotional valence. These attributes are of interest because we do not have a concrete visual definition of what they entail. What does it look like for a dog to be more or less memorable? GANs allow us to generate a manifold of natural-looking images with fine-grained differences in their visual attributes. By navigating this manifold in directions that increase memorability, we can visualize what it looks like for a particular generated image to become more or less memorable. The resulting ``visual definitions" surface image properties (like ``object size") that may underlie memorability. Through behavioral experiments, we verify that our method indeed discovers image manipulations that causally affect human memory performance. We further demonstrate that the same framework can be used to analyze image aesthetics and emotional valence. Visit the GANalyze website at http://ganalyze.csail.mit.edu/.