My answer is no.
There's two major reasons for this:
1) The key word in your question is 'intends'. Witness A has already stated his / her state of mental confusion. If the context is that this witness has been through a thoroughly mentally and emotionally disruptive few days, then such confusion would be within normal psychological parameters. If the context is also that said witness has been informed that a second witness (B), whom she trusts, has placed her at a crime scene, contrary to her own memories, and the police, whom she also trusts, have informed her they have 'hard evidence' that places her there, then that mental confusion would be compounded. A well respected expert in the psychological community has written many peer-reviewed papers about exactly what the consequences of such a state are for many people. Anyway, in a general sense, this kind of state can very well limit the mental capacity required for true intentionality.
Also, you seem to think that there's binary states; conscious and unconscious. That the only alternative to a fully conscious and intentional act is one that the subject is unaware of making completely. This is clearly not the case. The law clearly recognises that there are in between states, hence the necessity for proving mens rea as well as actus rea. And although these literally translate into guilty mind and guilty act, it refers to the wider sphere of human action, and states that in order for there to be an action, there must be a physical component and a mental component. And that mental component has to be executed by a mind which has the capacity to understand the nature of the act itself, and to be able to understand some of the relevant context of the act, for example the potential consequences. Actus rea, if witness A said those words, is proved, fine. But to show mens rea, you'd have some more work to do, other than stating that the witness was aware of making these sounds with her mouth.
So once you can see this, then you have to admit that your estimate of probability is flawed. However unlikely it was that witness A was completely unaware of her words (a state only likely with neurological problems or severe / unusual psychiatric disorder)the added possibilities that the witness did not understand the nature of the act, or the context acted in, raises the probability that witness A 'did not mean to place such evidence'.
2) Witness A's reasonable expectation of the truth-seeking nature of police investigation, meant that she reasonably expected that these words said within the context of previously stated mental confusion, and not directly connecting Mr X to the crime, would not be taken as 'evidence against' anyone, but that instead would be thoroughly investigated and tested for the purposes of corroboration.
The answer to your question is no. And these are the exact reasons why witness A should not be held morally or legally accountable for statements made against Mr Y either.
I don't know why you can't understand the notion of false memories, false confession syndrome, and until you publish (or reference) some peer-reviewed work explaining why these pieces of mainstream psychology are wrong, I guess they will stand as scientific fact.
The correct interpretation of witness A's statement could very well be that she / he was under a great deal of stress, getting less sleep than usual, probably having nightmares about a traumatic event, and was convinced (falsely) that his / her memories were not correct. In an effort to help police, the witness did a desperate mental inventory in this mentally clouded state, and found images, fragments of mental content (imaginings, dreams, memories) and tried to sort through them by theme (blood, violence, discomfort) instead of by category, and allowed herself to be (at least partially) convinced that these were memories, or if they were memories (like Mr X preparing raw fish) allowing them to be recontextualised within a new narrative.
