Version classiqueVersion mobile

Human and Machine Consciousness

 | 
David Gamez

12. Conclusion

Texte intégral

  • 1 Popper (2002, p. 94).

Science does not rest on solid bedrock. The bold structure of its theories rises, as it were, above a swamp. It is like a building erected on piles. The piles are driven down from above into the swamp, but not down into any natural or ‘given’ base; and if we stop driving the piles deeper, it is not because we have reached firm ground. We simply stop when we are satisfied that the piles are firm enough to carry the structure, at least for the time being.
Karl Popper, The Logic of Scientific Discovery1

12.1 A Framework for the Science of Consciousness

  • 2 Metzinger describes the current state of consciousness research as follows: ‘The interdisciplinary (...)

1This book has set out a systematic framework for the scientific study of consciousness. It has tried to shift consciousness research into a paradigmatic state.2 The key points are as follows:

  • Consciousness is a bubble of experience. It consists of colours, sounds, smells, tastes, etc., which are arranged in a bubble of space centred on our bodies.
  • The physical world is invisible. It has none of the secondary qualities that are present in a bubble of experience. Primary qualities in a bubble of experience are unlikely to resemble primary qualities in the physical world.
  • There are three hard problems of consciousness. First, it is impossible to imagine the relationship between consciousness and the invisible physical world. Second, we find it difficult to imagine the connection between conscious experiences of brain activity and other conscious experiences. Third, there are brute regularities between consciousness and the physical world that cannot be broken down or further explained. None of these issues affect scientific research on consciousness.
  • To scientifically study consciousness we need to measure consciousness, measure the physical world and look for mathematical relationships between these measurements.
  • Consciousness is measured through the external behaviour (c-reports) of systems that we assume to be conscious (platinum standard systems).
  • Normally functioning adult human brains are platinum standard systems.
  • Measurements of consciousness need to be expressed in a formal way (a c-description), so that they can be incorporated into mathematical theories of consciousness.
  • The correlates of a conscious state are a set of spatiotemporal structures in the physical world (a CC set) that is only present when the conscious state is present. A CC set is functionally connected to a bubble of experience and it e-causes c-reports about the bubble of experience.
  • CC sets need to be described in a formal context-free way (a p-description), so that they can be incorporated into mathematical theories of consciousness.
  • Pilot studies attempt to identify the CC sets that are associated with individual conscious states.
  • C-theories are mathematical relationships between c-descriptions of conscious states and p-descriptions of CC sets.
  • Computational methods should be used to discover c-theories.
  • Information c-theories claim that information patterns could form CC sets by themselves. Computation c-theories claim that computations could form CC sets by themselves. These types of c-theory do not conform to the constraints on scientific theories of consciousness (C1-C4).
  • Physical c-theories link patterns in particular materials to conscious states. They conform to the constraints on CC sets and fit in well with other scientific theories.
  • C-theories can generate predictions about consciousness or the physical world. Predictions can only be tested on platinum standard systems.
  • C-theories can make conservative and liberal deductions about the consciousness of non-platinum standard systems, such as bats, infants and robots. Deductions are logical consequences of a c-theory that cannot be checked.
  • 3 For example, whether consciousness is a non-physical substance, the hard problem, solipsism, zombi (...)

2This framework handles or sets aside most of the philosophical problems with consciousness.3 We cannot solve these problems. We can only show that they are pseudo problems, suspend judgement about them or set them aside with assumptions.

  • 4 In the standard version of dualism there is a bidirectional e-causal relationship between non-phys (...)

3This framework is compatible with some of the metaphysical approaches to consciousness, such as physicalism and epiphenomenalism. It suspends judgement about which (if any) are correct. It is not compatible with panpsychism, dualism,4 or with information and computation c-theories.

4This framework prescribes the form that legitimate theories of consciousness should take. Our final c-theories will not be lengthy pieces of natural language. They will be mathematical relationships between c-descriptions and p-descriptions.

5This framework is neutral about which physical c-theory is correct. While I have often used neurons and electromagnetic fields as examples, I have no idea about which patterns and materials are actually linked to consciousness. This question should be addressed by scientific experiments.

  • 5 Assumptions A7-A9 are more pragmatic and might not be needed by the science of consciousness.

6A person who accepts this framework can focus on measuring consciousness, measuring the physical world and identifying the relationships between these measurements. Their results can be considered to be true given assumptions A1-A6—they cannot be obtained or justified without these assumptions.5

12.2 Technological Limitations

7The science of consciousness can only fully develop when we have accurate high resolution measurements of consciousness and the physical world.

8The problems with measuring consciousness are mostly conceptual and methodological (see Chapter 4). If enough effort is spent, we should be able to obtain detailed and reasonably complete measurements of conscious states.

9To accurately measure the physical world we need p-description methods that avoid the heavy reliance on context that is common in biology (see Section 5.1). We also need better access to the 100 billion neurons in the living human brain. The most commonly used technologies are as follows:

  • Functional Magnetic Resonance Imaging (fMRI). An indirect measure of brain activity with a spatial resolution of a few thousand voxels. Each voxel corresponds to the activity of approximately 50,000 neurons averaged over several seconds.
  • Diffusion Magnetic Resonance Imaging (dMRI). Identifies structural connections between brain areas, but does not show their direction. Can only help to identify CC sets when combined with other methods.
  • Electroencephalography (EEG). Approximately 300 electrodes are placed on the scalp to measure the brain’s electrical field with good temporal resolution and very poor spatial resolution.
  • Magnetoencephalography (MEG). Measures the magnetic fields generated by groups of 50,000 neurons at approximately 300 points on the head with good temporal resolution.
  • Implanted electrodes. Up to 300 electrodes can be implanted in the brain to measure electromagnetic fields and neuron activity with good spatial and temporal resolution.6 Electrodes are rarely implanted in human subjects.
  • Optogenetics. Neurons can be genetically engineered to emit light when they fire, which enables their individual activity to be recorded using light sheet microscopy. This technique can be used to record from 100,000 neurons in a zebrafish larva in close to real time.7 It is more challenging to use optogenetics in mammalian brains and for ethical reasons it has not been used on human subjects.
  • 8 For example, Shanahan and Wildie (2012) have proposed a ‘knotty centrality’ measure that might be (...)

10The data that is extracted using these techniques can be processed into higher level properties. For example, we can use Granger causality or dynamic causal modelling to identify effective connections between brain areas. These connectivity patterns can be further analyzed using graph theory.8

  • 9 An assumption that a mammalian brain is a platinum standard system will have much less impact on t (...)

11Optogenetics is the most promising technology for obtaining high resolution data from living brains. However, there are ethical and safety concerns about using it on humans. To get around these problems we can use animal brains to make inferences about CC sets in humans. Or we can assume that monkeys and mice are platinum standard systems.9

12C-theories can be based on patterns that have higher resolution than our current measurement technologies. For example, we could develop c-theories based on neuron activity patterns and predict how these neuron activity patterns would appear in EEG data or a fMRI scan.

13The scanning and uploading of a human brain could help to address our measurement problems. We could identify the structures in a simulated brain that cause its simulated c-reports. This might help us to develop c-theories that are not limited by our current technologies.

12.3 Other Limitations

  • 10 Wittgenstein (1969, remark 94).

I did not get my picture of the world by satisfying myself of its correctness; nor do I have it because I am satisfied of its correctness. No: it is the inherited background against which I distinguish between true and false.
Ludwig Wittgenstein, On Certainty10

  • 11 For example, consider the assumption that all conscious states associated with a platinum standard (...)

14The framework presented in this book cannot be shown to be correct. It is a condition of possibility of experimental work on consciousness that cannot be verified by experimental work on consciousness.11 Scientific research within this framework might be fruitful and yield reliable c-theories. Or a science of consciousness based on this framework might reach a point at which it no longer coherently hangs together. We might have to formulate a completely new set of framing principles. Or we might have to abandon the attempt to scientifically study consciousness.

15The inappropriate use of intuition, thought experiments and imagination has led to many problems in the philosophy of consciousness. I have tried to banish these as much as possible, but they cannot be completely eliminated. For example, I have assumed that normally functioning adult human brains are platinum standard systems. But which systems count as normally functioning adult human brains? What counts as a legitimate chemical modification? There are no natural boundaries—we have to use our intuition and imagination to decide which human brains are platinum standard systems.

16This framework leaves many questions unanswered. It does not explain what consciousness is, what consciousness does, what consciousness is for, how consciousness arose, why there is a functional connection between consciousness and the physical world or how this connection actually works.

17These questions are about consciousness in general. But it makes no sense to ask about the physical world in general. We can ask about the origin and function of particular physical structures—we cannot meaningfully ask about the origin and function of the entire universe. The case is similar with consciousness—it is meaningless to ask most of these questions about consciousness in general.

18Let’s rephrase these questions. Consider a state of consciousness, c8. c8 is a bubble of experience in which you are peeping through a hole at an old woman in a bath. We can ask what c8 is, what c8 does, what c8 is for, how c8 arose, why c8 is functionally connected to the physical world and how this connection actually works.

19These questions can be answered if we assume that c8 is identical to its associated CC set, cc 8. This explains the connection between c8 and the physical world. As the science of consciousness progresses we will get a better understanding of what cc8 is, what cc8 does and how it arose through e-causal processes, such as evolution. All of our questions about c8 can be answered by transferring them to cc8.

20Our questions about c8 can also be answered by assuming that it is a teapot. This tells us what c8 is (a teapot), how it arose (a factory in China) and what it is for (making tea). But assumptions about consciousness have to be valid, they have to make sense. An assumption’s validity has to be decided independently of our desire to obtain cheap and easy answers about consciousness. No answers are better than bad answers.

  • 12 This is the position of Berkeley (1957). Husserl (1960) developed his phenomenological program by (...)

21It makes little sense to say that colourful smelly noisy bubbles of experience are identical to something that is invisible, silent and without smell. This discards the properties of bubbles of experience and ignores the reality of our day-to-day world. We could equally well discard the physical world and declare that it is a fairytale told by simple folk to explain regularities in consciousness.12 Neither reduction is part of the framework that is set out in this book. Consciousness and the physical world are both taken as basic realities that can be measured and scientifically studied.

22The physical sciences’ assumption that the physical universe exists leaves many questions unanswered. Many questions will remain unanswered if conscious states are not reduced to physical states. We will simply have to accept that consciousness exists and study the brute regularities between consciousness and the physical world.

12.4 Future Research

  • 13 Hume (1993, p. 114).

If we take in our hand any volume; of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames: For it can contain nothing but sophistry and illusion.
David Hume, An Enquiry Concerning Human Understanding13

23The following types of consciousness research are likely to be productive:

  • Revision of the assumptions. We might be able to reduce the number of assumptions, improve their consistency and enhance the way in which they relate to general principles in the philosophy of science and the study of consciousness.
  • Creation of c-description format. We need a precise formal way of describing states of consciousness that can be incorporated into mathematical c-theories.
  • Improvement of methods for measuring consciousness. More work is required on how we can obtain detailed measurements of conscious states.
  • Creation of a context-free p-description format. We need a formal context-free way of describing biological structures, such as neurons.
  • Development of a precise definition of a physical context. Conservative and liberal deductions can only be distinguished when we have a precise definition of a physical context.
  • Increase the spatial and temporal resolution of our brain measurements. We can refine existing methods, develop new technologies and create better mathematical techniques for processing data into higher level properties.
  • Pilot studies on the correlates of consciousness. More pilot studies could help us to identify the patterns and materials that form CC sets.
  • Development of physical c-theories. In the medium to longer term we need to move beyond pilot studies and identify mathematical relationships between bubbles of experience and physical states.14
  • Experimental testing of physical c-theories. Physical c-theories will only be considered to be reliable when their predictions have been experimentally confirmed.
  • Construction of computer models of CC sets that simulate c-reports. This could help us to identify the patterns and materials that form CC sets. These models could also be used to develop methods for the computational discovery of c-theories.
  • Development of methods for the computational discovery of c-theories. This could apply existing work on the computational discovery of scientific theories to data from consciousness and the brain.
  • Deductions about the consciousness of non-platinum standard systems. When we have a reliable c-theory we will be able to answer questions about the consciousness of bats, infants and robots. Deductions also have important medical applications.
  • Experimental work on the modification and enhancement of consciousness. When we have reliable c-theories and better technology for modifying the brain we will be able to systematically modify and enhance human consciousness.
  • Construction of MC1-MC4 machines. This has many practical applications and could be a useful way of studying human consciousness.

24A book or paper on consciousness that describes none of these things should be committed to the flames. Or carefully checked for sophistry and illusion.

Notes

1 Popper (2002, p. 94).

2 Metzinger describes the current state of consciousness research as follows: ‘The interdisciplinary project of consciousness research, now experiencing such an impressive renaissance with the turn of the century, faces two fundamental problems. First, there is yet no single, unified and paradigmatic theory of consciousness in existence which could serve as an object for constructive criticism and as a backdrop against which new attempts could be formulated. Consciousness research is still in a preparadigmatic stage. Second, there is no systematic and comprehensive catalogue of explananda. Although philosophers have done considerable work on the analysanda, the interdisciplinary community has nothing remotely resembling an agenda for research. We do not yet have a precisely formulated list of explanatory targets which could be used in the construction of systematic research programs.’ (Metzinger 2003, pp. 116-7).

3 For example, whether consciousness is a non-physical substance, the hard problem, solipsism, zombies, colour inversion and the causal relationship between consciousness and the physical world.

4 In the standard version of dualism there is a bidirectional e-causal relationship between non-physical consciousness and the physical world. This is ruled out by assumption A5.

5 Assumptions A7-A9 are more pragmatic and might not be needed by the science of consciousness.

6 NeuroNexus sells electrodes that can record from 256 locations. It is working to expand this to 1,000 electrodes (Marx 2014).

7 See Ahrens and Keller (2013).

8 For example, Shanahan and Wildie (2012) have proposed a ‘knotty centrality’ measure that might be linked to consciousness.

9 An assumption that a mammalian brain is a platinum standard system will have much less impact on the science of consciousness than a similar assumption about an artificial system.

10 Wittgenstein (1969, remark 94).

11 For example, consider the assumption that all conscious states associated with a platinum standard system are linked to c-reports (A2). We could disprove this assumption if we could show that there are aspects of consciousness that are not accessible through c-reports. But since these aspects of consciousness cannot be accessed, we cannot prove that they do or do not exist.

12 This is the position of Berkeley (1957). Husserl (1960) developed his phenomenological program by suspending commitment to the reality of the physical world.

13 Hume (1993, p. 114).

14 At present the most mathematical theories of consciousness are information c-theories, which can be re-interpreted as physical c-theories. For example, Tononi’s (2008) information integration theory of consciousness can be re-interpreted as a theory about the relationship between neuron activity and consciousness. Causal density (Seth et al. 2006), liveliness (Gamez and Aleksander 2011) and Casali et al.’s (2013) perturbational complexity index can also be reinterpreted as physical c-theories.

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search