What kind of evidence is relevant to understanding consciousness?

The answer we give here is contrastive analysis ⎯ a set of paired comparisons between similar conscious and unconscious processes (Baars, 1983; 1988; 1997).

The classical experimental paradigm here is binocular rivalry, where two incompatible streams of visual information are presented to the two eyes in such a way that they cannot be fused into a single image. Numerous studies show conscious visual streams show more cortical activation, more oscillatory phase-linking, and wider propagation than closely matched unconscious stimuli. Conscious visual stimuli, for example, propagate not only within the visual cortex, but spread to prefrontal and plausibly hippocampal regions for episodic memory coding.

In contrast, experimentally matched unconscious events are not voluntarily reportable, they evoke less neuronal amplitude, show less oscillatory phase-linking, and spread much more locally in cortex.

Many other experimental methods have been used for comparing conscious and unconscious contents, such as visual backward masking, the attentional blink, inattentional blindness, and similar effects in the other senses, like selective listening. Brain measures of these experimental comparisons range from fMRI to PET, intracranial EEG, local field potentials, and even single-neuron recordings. This literature is now very large.

Figure 6 shows the fMRI response for conscious and unconscious rivalry in the face area versus the place area of the temporal lobe. (Called F and H.)

Conscious events can also evoke voluntary actions, notably accurate stimulus reporting.

The set of indirect effects of conscious events is very large, and may be an open set — including sensorimotor control, inner speech, mental problem solving, and emotional reactions. Some downstream events, like amygdala activation, do not present reportable contents, but they often follow from clearly reportable experiences, like a fearful image. Emotions in turn trigger widespread autonomic activity, neurohormonal responding, and a large suite of gene expressions.

Thus a careful scientist might talk about “observer reports” to index conscious experiences, rather than trying to join in the subject’s own subjectivity. The scientific aim is to include subjectivity, but also to compare subjective events to closely matched brain events that are not reportable or experienced. That strategy has been immensely fruitful.

Conscious events can be defined in practice as those brain activities that subjects can report with high accuracy under optimal conditions, such as minimal distraction and minimal time delay. Reportability is a major empirical index that corresponds well with our everyday understanding. Indeed, the words “accurately reportable” can often be used instead of “conscious.”

However, that would miss something essential, the fact that accurate reports refer to a rich surround of experiences that we all claim to have. Nevertheless, empirical indices evolve over time, better indices of conscious activity are indeed emerging.

An obvious example is the “waking EEG,” which is different from sleep and drowsiness. Waking EEG has been known since 1929, and provides a primary medical measure of the conscious state, when patients are unable to communicate. Waking EEGs are not a perfect measure, and better indices are constantly being developed. In contrast, unconscious events can be defined as those that are known to exist without the ability report them, such as subliminal activation of cortical regions for color, shape, or object identity.

Thus we have useful empirical indices for both conscious and unconscious brain events even today. As recording methods improve, the number of indices continues to grow.

Figure 5 & 6 Courtesy of Frank Tong: Ken Nakayama; J. Thomas Vaughan; Nancy Kanwisher, (1998), Binocular Rivalry and Visual Awareness in Human Extrastriate Cortex, in Neuron ISSN: 0896-6273, Vol: 21, Issue: 4, Page: 753-759.