---
title: "A Context-Dependent Attention System for a Social Robot"
authors:
  - "Cynthia Breazeal"
  - "Brian Scassellati"
doi: ""
url: "https://ijcai.org/proceedings/1999/163"
pdf: "https://ijcai.org/proceedings/1999/0163.pdf"
venue: "Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence"
year: 1999
pages: "1146-1152"
track: "Robotics and Perception"
keywords:
---

# A Context-Dependent Attention System for a Social Robot

Cynthia Breazeal, Brian Scassellati

*Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence*, pages 1146-1152.

## Abstract

Thi s pap er presents part of an on-going project to integrat e perception, attention, drives , emotions, behavior a rbitration , a nd expressive acts for a robot designed to interact socially with humans . We prese nt the design of a visual attention system based on a model of human visual search behavior from Wolfe (1994) . The a ttention sys tem inte grat es perceptions (motion det ection, color saliency, and face popo ut s) with habituation effec ts a nd influences from th e robot 's motivational and behavioral st a te to cre ate a conte xt -de pendent attention activation map . This activation map is used to direct eye movements and to sati a te the drives of th e motivational system.