Mammals (from rodents to primates) can detect and therefore become aware of visual stimuli exhibiting luminance-contrast levels as low as 1 to 2% (Schiller and Tehovnik 2015; Histed et al. 2012). Nevertheless, in the absence of the visual cortex the threshold to detect stimulus luminance-contrast surpasses 95%, which has been defined as blindsight in mammals such that visual awareness is lost under these conditions (Tehovnik, Patel, Tolias et al. 2021). Therefore, an intact neocortex is paramount for normal vision and for experiencing visual awareness with high sensory acuity; this also seems to be true of the other senses (Börnstein 1940; Exner 1881; Grüsser and Landis 1991; Guldin and Grüsser 1998; Heffner and Heffner 1986; Juenger et al. 2011; Paillard et al.1983; Zatorre and Jones-Gotman 1991).
Masking has been used as an effective tool to study conscious awareness. A target stimulus such as a spot or a face is presented first which is then followed by a mask that can be in the form of an annulus (to interfere with the spot) or a scrambled image (to interfere with the face) (Graziano 2019; Web, Graziano et al. 2016). By shortening the delay between the target stimulus and the mask (e.g., to 50 ms), the target stimulus becomes silent to consciousness, but at longer delays (e.g., 100 ms) the target is restored to consciousness. Although large regions of the neocortex (including the parietal, temporal, and frontal cortices) are activated under both conditions, it has been reported that when the mask fails to obstruct consciousness the neurons are activated more robustly, as measured with fMRI (see Fig. 4A of Web, Graziano et al. 2016).
Even though Graziano and colleagues (Graziano 2019; Web, Graziano et al. 2016) have interpreted the neocortical enhancement as being due to consciousness, we would suggest that along with consciousness the brain has been double stimulated since both the target and the mask activate the brain during consciousness. Consciousness is no different from detecting a stimulus, it too has a threshold related to the strength of synaptic connectivity but between the association areas (Tononi et al. 2008ab) which allows for the binding of perceptual elements (Kohler 1929; Singer 2001).
Furthermore, Kelly, Graziano et al. (2014) found that task difficulty (for binding perception) is central to how robust the neocortex is activated during task performance. Human subjects were required to associate expressions on a face with an image representing danger or safety. For example, a burning building should be associated with a face exhibiting horror rather than happiness. The emotive elements within a face were manipulated to coincide or disagree with an image and a subject was required to judge the degree of correspondence. It was found that the more difficult the correspondence (i.e., the more ambiguous) the larger the region of neocortex activated, as assessed with fMRI. This coincides with the finding that when one performs an under-learned behavioral routine (i.e., sequence learning with the fingers) more neocortical tissue is activated than when one performs an overlearned routine (Lehericy et al. 2005). This also applies to language: executing a secondary language utilizes more neocortical neurons than executing a primary language (Ojemann 1991).
Therefore, the purpose of declarative learning is to automate a behavior, so as to minimize the amount of consciousness/neocortical tissue/neural energy required for task execution; once a task is overlearned the neocortex is released (in a relative sense) to concentrate on new issues that require new learning (Tehovnik, Hasanbegović, Chen 2024). In short, consciousness/more neocortical tissue directed toward a task is an acknowledgement that the task needs further learning—further binding (Kohler 1929; Singer 2001).