In process theories the crucial thing is what the brain does, not how it is doing it, i.e. the specific architecture of the brain is not crucial. Causal structure theories propose just the opposite. For example, the recurrent processing theory (RPT) and the integrated information theory (IIT) propose both that systems with feedforward processing are not conscious but also that all sorts of systems with recurrent processing (even very simple ones) are conscious. We will also introduce the ‘unfolding argument’ which holds that causal structure theories may be incompatible with the scientific method.

What are the characteristics of consciousness? How long does a conscious percept last? Are animals conscious? We review the research related to these and many other questions and discuss how the answers to these questions can help us to judge the merit of different theories of consciousness. For example, a theory that predicts that consciousness occurs only after several hundred milliseconds after stimulus onset, would be falsified if consciousness were to occur earlier than that. We will see that most theories of consciousness make very few precise predictions about such matters.

We portray processes theories which claim that consciousness is related to or identical with a computational process in the brain. The global workspace theory (GWT) proposes that consciousness occurs when information is made widely available via a global neuronal workspace. Higher order theories (HOTs) propose that consciousness occurs when there is a higher-order process that describes a first-order sensory-cognitive process. We will also introduce the ‘small network argument’, which holds that extremely simple systems may fulfil the criteria of process theories and hence would be considered conscious.

In Chapters Two and Three, we learned about the experimental tools for investigating consciousness and about its characteristics. We are now ready to delve into the realm of theories proposed to explain consciousness.

We introduce the two major branches of research on consciousness: state consciousness and content consciousness. We review and critically discuss the many empirical paradigms that are used in consciousness research, including visual masking, binocular rivalry, and anesthesia.

All theories up to now relate consciousness to neural processes, circuits, etc. Quantum theories, such as orchestrated objective reduction (OrchOR), propose that proto-conscious events occur in the entire universe and full consciousness occurs only at the subcellular level in the microtubules of neurons. We outline challenges with this approach.

Biological theories claim that consciousness occurs in specific neural circuits. Based on a wealth of neural data, the dendritic integration theory (DIT) claims that consciousness occurs when information in the dendrites of neurons is gated to higher-level neurons under the control of the thalamus. Based on patient studies, the felt uncertainty theory (FUT) claims that consciousness is not related to sensation and cognition but to emotions. We will also introduce the ‘other systems argument’, which holds that biological theories may not generalize to other kinds of systems.

What is consciousness? Can we study consciousness scientifically or is consciousness beyond empirical science, as many philosophers have claimed throughout the ages? This chapter provides a short history of the topic, including philosophical and scientific milestones, and gives an overview of what is to come in subsequent chapters.

Chapters Four through Nine portray the most well-known theories of consciousness. Each portrayal was written by an eminent proponent of that particular theory, and thus the portrayals provide accurate and unbiased information. Chapter Four introduces the topic, provides the tools for analyzing the theories logically, and shows that a theory needs to jointly address conscious and non-conscious brain activity which, as it turns out, many theories do not do. The chapter features the learning and consciousness theory (SOMA).

We provide an in-depth analysis of the theories portrayed in the previous chapters. As it turns out, all theories except illusionist theories are monofactorial theories, proposing that one thing is crucial for consciousness, such as a workspace. The monofactorial nature of such theories is the main reason why the theories face the challenges that we outlined. A further consequence is that consciousness is epiphenomenal according to these theories, i.e. consciousness has no impact on actions, feelings, or anything physical. Epiphenomenalism is hard to reconcile with empirical science. We argue that these problems are avoided by a multifactorial framework, in which consciousness is treated as a supra-ordinate term picking out a specific constellation of interacting mechanisms. On this view, progress hinges on identifying the computational or biological problem(s) that this constellation solves.