Ned Block (2005), Trends in Cognitive Sciences 9(2), 46–52. An Opinion article. A longer version is in Block’s Consciousness, Function, and Representation (MIT Press, 2007), which was not read.
Key takeaways
- Phenomenal NCC. The minimal neural basis of an experience’s content, what differs between the experience as of red and as of green. For seen motion, the evidence points to recurrent activation of MT/V5, which TMS shows depends on a feedback loop to V1 (Pascual-Leone & Walsh: a V1 pulse 5–45 ms after the MT pulse makes moving phosphenes stationary). There are different phenomenal NCCs for different contents.
- Access NCC. The neural basis of a content being broadcast in the global workspace: activation of frontal and parietal areas by occipital and inferior temporal ones (Dehaene and Changeux’s workspace neurons).
- Why the two can come apart. Access is a winner-take-all competition. In the attentional blink, subjects rate the blinked target as fully seen or fully unseen, almost never in between. A recurrent sensory loop can lose that competition and so stay unaccessed.
- Signal detection as a method. Whether a subject reports seeing depends on a criterion as well as on what is seen. A criterion shift can explain the “exclusion” result without a difference in consciousness. In Supèr et al.’s monkeys, V1 figure–ground modulation persists, at the same level whether or not the monkey saccades, when salience is low or catch trials frequent. Block takes it as an indication, “if not directly part of”, a phenomenal NCC.
- Against reports as the gold standard. No evidence is privileged; a theory built only on introspective reports would find only the basis of higher-order thought; and monkeys’ “nothing” responses in blindsight are more plausibly reports about the world than about their experience.
Standing in the record
Filed on 2026-10-03 at the owner’s request, with six other papers by Block on consciousness. No anthology topic holds a theory of the neural basis of experience, and the paper carries no instruction for machine-learning practice.
It extends Block (1995), LIT-643. It opens by restating that paper’s phenomenal/access distinction and moves it from concepts to the brain: two neural correlates, and an empirical case that they come apart. The longer argument for the same picture is Block (2007), LIT-646.
Bearing on the new theories.
- THEORY-102 (Vishne et al., LIT-644) finds content carried for as long as an image stays on by a stable code in ventral temporal cortex, while prefrontal content is a transient at onset. That pattern is what this paper’s division predicts if the sustained code is a phenomenal NCC and the onset burst is access. Vishne et al. report that the result met both IIT’s and the workspace’s predictions. On this paper’s terms it would not discriminate them as theories of access, but it bears on where a phenomenal NCC is. That reading is mine.
- THEORY-100 (crow NCL) is a correlate of the reported percept. In this paper’s terms that is an access NCC candidate. It shows nothing about the phenomenal NCC, which is the caveat that theory already states.
No relation is declared to those works. The papers the argument rests on (Supèr et al. 2001; Pascual-Leone & Walsh 2001; Dehaene et al. 2003; Visser & Merikle 1999) are not held.