Abstract

This thesis examines whether it is possible to generate fear-eliciting media that custom fits to the user. The system described uses a genetic algorithm to produce images that get more scary through the generations in reaction to either physiological signals obtained from the user or a user-provided fear rating. The system was able to detect differing levels of fear using a regression trained on EEG and heart rate data gathered while users view clips from horror movies. It was also found to produce images with significantly higher fear ratings at the fifth generation as compared to the first generation. These higher scoring images were found to be unique between subjects.

Library of Congress Subject Headings

Fear--Interactive multimedia--Design; User interfaces (Computer systems)--Design; Genetic algorithms

Publication Date

8-2015

Document Type

Thesis

Student Type

Graduate

Degree Name

Computer Science (MS)

Department, Program, or Center

Computer Science (GCCIS)

Advisor

Joe Geigel

Advisor/Committee Member

Arthur Nunes Harwitt

Advisor/Committee Member

Phil White

Comments

Physical copy available from RIT's Wallace Library at BF575.F2 D46 2015

Campus

RIT – Main Campus

Plan Codes

COMPSCI-MS

Download

Share

COinS