Added poster and updated README

README.md

@@ -1,22 +1,7 @@
-## Harry's Habits: Investigating Habit Rewards
+## Harry's Habits: A Chatbot Investigating the Impact of Positive Reinforcement Rewards on Habit Formation.
 
-Written at the University of Bristol.
+Written at the University of Bristol for my MSc Advanced Computer Science [thesis](thesis/main.pdf). I built a [chatbot](https://github.com/harrymt/harryshabits) that helps people form new positive habits by deliving different types of positive reinforcement rewards.
 
+## Overview
 
-My MSc [thesis](thesis/main.pdf) measures how rewards from different modalities effects peoples habit strength, using a chatbot to track habits by sending notifications to reminder users and delivering rewards from three modalities.
-
-An evaluation trial tests this chatbot using real-world participants that will use the bot to track any simple habit over 30 days.
-
-
-### Chatbot
-
-A [chatbot](https://github.com/harrymt/habit-reward-chatbot) tracks users habits and rewards them from different modalities.
-
-
-### Research Overview
-
-A [research review](https://github.com/harrymt/habit-reward-thesis/releases/tag/0.1) the for a review on the relevant material.
-
-
-
-[Thesis](thesis/main.pdf) for my MSc Advanced Computing, Creative Technology.
+Habit formation systems use rewards to motivate people to form habits. This thesis looks at the effect of three types of positive reinforcement rewards on habit formation. The rewards are delivered by a chatbot from three modes: visual, auditory and visual-auditory combined. The findings are evaluated against two hypotheses: i) positive reinforcement is effective at supporting habit formation by increasing automaticity and regular habit performance, ii) multiple modalities rewards are more effective then singular mode rewards. 60 people were recruited to participate in a 4-week study followed by voluntary semi-structured interviews. The findings showed that participants receiving the bot-delivered rewards had higher habit performance than the control group without rewards. Participants with visual-auditory rewards had the highest habit automaticity score, however, these were not statistically significant. Finally, all participants interviewed (N = 7) reported a drop in habit performance after one week without the prototype. Further research for using different rewards with behaviour change technology is needed to validate how each modality affected habit automaticity and habit performance.