Several covariates in ANCOVA?

I am running a study where I compare changes in brand attitudes across time for four brands, all within subjects. Between Time 1 and time 2, participants see some information and i want to see how it impacts their rating of each brand. I want to use ANCOVA to control for Time 1 brand attitudes and then look for changes in brand attitudes at Time 2. Should I just put all four Time 1 brand attitudes in the ANCOVA as covariates? This seems odd since i feel like they should be matched to their respective Time 2 attitudes...

Any help would be appreciated. Thanks. Alex.