How does retention compare when a speaker presents information both visually and orally, compared to presenting information visually or orally?