If matter can’t explain consciousness, it is unclear why its information derivative should do better, but integrated information theory argues that “consciousness is integrated information” (Tononi, 2008), generalizing an earlier theory that brain senses like vision, hearing, and smell store data in a global workspace (Baars, 1988).
Nerves from the eye go to the back of the brain, from the ears to its sides, and smell to its center, so how do sizzling sausages whose smell, sight, and sound are analyzed in distant brain areas appear as one object? The global workspace answer is that sense data is analyzed then put into a common area that higher functions like memory and language can act upon because:
“… the information has entered into a specific storage area that makes it available to the rest of the brain.” (Dehaene, 2014), p163.
Integrated information theory equates this unification to consciousness, but clinical studies deny that a specific brain data-center causes consciousness (Block, 1995), as does information theory. To exchange information, sender and receiver must agree on what physical signals mean. If they don’t, the result is gibberish, like a foreign language, so the auditory area of the brain can no more read visual data than I can read Chinese. Unless a global workspace had a sight, sound, and smell translator that the brain doesn’t seem to have, brain functions couldn’t use it.
The same logic applies to apps on a computer. For example, Windows Notepad reads text data and Paint reads picture data, but loading text into Paint, or a picture into Notepad, gives nonsense, so even if both put their data into a common area, neither could read what the other posted. Programs like Word that read text and pictures need a lot of code to do that, so they become very big, and still can’t read zip files for example. In general, for one application to read all data denies the benefits of specialization, and brain functions are specialized not centralized. One can’t plug the optic nerve into the auditory cortex and expect information to flow like water.
Nonetheless, workspace theory goes on to suggest that neural systems chat like little people:
“… neural systems do not merely report to their superiors; they also chat among themselves.” (Dehaene, 2014), p176.
The brain then uses neuron sociology (Nunez, 2016) p18 as the Internet uses crowd control:
“… it is helpful to think metaphorically of a theater of mind. In the conscious spotlight on stage – the global workspace – an actor speaks, and his words and gestures are distributed to many unconscious audience members, sitting in the darkened hall. Different listeners understand the performance in different ways. But as the audience claps or boos in response, the actor can change his words, or walk off to yield to the next performer.” (Baars & Laureys, 2004), p672.
The analogy is seductive but that neural areas chat like little people over nerve phone lines, or clap and boo each other as we do online, contradicts information theory. Electrical signals going down a wire don’t exchange information unless designed to do so. For example, the Internet needs data, network, transport, and application protocols to work, namely ethernet, internet protocol (IP), transmission control program (TCP), and hypertext transfer (http) respectively. The Internet’s TCP/IP/Http protocols took decades to develop from the original Arpanet, and it was done by a central group the brain doesn’t have. And if the Internet integrates data, shouldn’t it also be conscious? Information integration theorists expect it be so soon (Koch, 2014), despite no evidence.
Theories of brain consciousness must respect information science, but information integration theory doesn’t, and what doesn’t work for computer networks won’t work for brains either. The cartoonish concept that neural areas chat like little people, by a common language the brain doesn’t have, merging claps or boos they can’t make, on a central stage that doesn’t exist, is a fantasy. If integrated information caused consciousness, the Internet would have had it long ago, but it hasn’t.