Binary slice expressions and capacity

Is it guaranteed that, for a slice s, s[a:b] is equivalent to s[a:b:cap(s)]?

It’s plausible, but I can’t find any definitive explanation about this in The Go Programming Language Specification - The Go Programming Language. or am I misinterpreting the spec?

EDIT: Though not a spec, Go Slices: usage and internals - The Go Programming Language implies so.