Is learning in SRT tasks robust across procedural variations