I have been with IBM for past 5 years and worked on cluster interconnects. During this time I have developed significant research oriented interest in this area as well as in the area of multi-core programming.
I applied for MS to a few places this year. And I got an admit from OSU. In fact the professor working in these areas, who I had mentioned in my SOP, mailed me and advised me to convert the application to PhD (direct PhD). So I am assuming my work-ex played a huge role in getting me the admit. And although it hasn't been formally mentioned anywhere but it will be safe to assume that the same professor is sponsoring my GTA and in the normal course would be the best choice to be my guide.
But I also have another area of interest, cognitive science. This I have developed over the years on my own motivation (no formal courses or work-ex). And whenever in the past I have thought about a Phd, I have only wanted to pursue this area. My reason for applying to OSU was that this is one for the few universities with a dedicated dept for cog sci (although it is not a degree awarding dept). and this dept has quite a few cross affiliations with the CS dept.
Now, I am pretty sure I want to go ahead with the PhD (leaving comfy job and signing on for academic hardships for next 5-6 years was another turmoil but i am over it now).
But I am torn between these two areas.
Cluster interconnect offers great funding, intern and job opportunities. and its not like I hate it.
cog sci, I am really passionate about and phd is something i will only do once.
I suppose usually when someone starts a Phd, they are very certain what they want to do.
But I would like to experiment initially with both areas and find out which is best for me. Will this be possible ? Is it advisable ?