An Attention-Based Control Model of Consciousness (CODAM)

Despite the recent ‘Race for consciousness’, many neuro-scientific approaches have failed to use the concept of ‘attention’ as a guide to consciousness. In this review, John G. Taylor claims that consciousness can be more fruitfully regarded as created by processes arising from the movement of attention.

Introduction

Consciousness is a subtle phenomenon, which has so far resisted all attempts to understand it, in spite of the present ‘Race for Consciousness’ (1). Without attention to an input there is no awareness of it (2, 3). Yet several recent neuro-scientifically based papers on consciousness (4, 5, 6) have surprisingly neglected attention as guide to consciousness. To remedy this lacuna, the thesis of this note is that consciousness can be more fruitfully regarded as created by suitably specific processes arising from the movement of attention, leading to a tentative neural mechanism for the creation of mind from matter: the Corollary Discharge of Attention Movement (CODAM) model.

Recent progress in modelling attention based on these results has been summarized here and here. These models can all usefully be viewed from an engineering control framework, allowing a general picture of attention control to be developed consistent with the experimental data (7). Contained Consciousness can be thought of as created by suitably specific processes arising from the movement of attention within such a control model are the components of primary and secondary sensory and motor cortices with input activations being controlled by an inverse controller in parietal/frontal sites (indicating where or to what attention should next be directed), and a rules module containing the desired state into which attended input should be transformed (there being prefrontal top-down and superior colliculus bottom-up components). A ‘forward model’, or observer, can also be suggested, which estimates the present attention state used for rapid error correction before sensory feedback becomes available.

Philosophical and Meditation Aspects of Self

Consciousness is not in sight from an attention viewpoint, even though without attention there is no consciousness of a given input. However the implicit capture of attention in blindsighted subjects (8) and brought about by degraded inputs in normal subjects (9) show that attention is not sufficient for consciousness. We must search inside the control model of attention to single out what components of neural activity involved with attention would also be sufficient for conscious experience. Recent relevant developments have occurred in the philosophy of consciousness, as it relates to the perception of ‘self’ from a phenomenological viewpoint, especially in understanding the nature of altered states of consciousness. In phenomenology there are two components of consciousness: ‘consciousness of’ (= the intentional component) and the pre-reflective self.

The pre-reflective self is experienced as the ownership of one’s conscious experience and as the basis of all awareness; without it there would be content but no owner of that content (10). There should therefore be room for such ownership information to be present in a model of attention strong enough to contain conscious experience. From an independent avenue, meditation has been claimed by many to lead to the remarkable ‘pure consciousness experience’ or PCE, claimed to be content-free. Recent brain imaging has supported the existence of this distinct state of consciousness. Both the pre-reflective self and the PCE are critical components of consciousness.

I proposed (11) that PCE is a temporal extension of the pre-reflective The pre-reflective self is experienced as the ownership of one’s conscious experience and as the basis of all awareness self (12). This resolves the contradiction between the Western analytic approach to mind, where consciousness is identified only with content (in contrast to the Western phenomenological approach) and that of the East, for which the PCE state is arguably the supreme target of meditation (13): the Eastern mystic attempts to attain a state already experienced, albeit briefly (and possibly unknowingly), by all humans.

An Attention-Based Control Model of Consciousness (the CODAM model)

I now fuse the control model of attention with the philosophical and Eastern meditation-based understanding of experience. In the control model, the observer is assumed to contain a buffered copy of the controller signal, which is used to achieve more rapid updating of the movement control signal (14, 15). Such a copy will not be bound in any attention-based manner to the content of consciousness, since that can only be present on feedback from the sensory cortex. The corollary discharge signal will therefore have no content. It can however be identified with the experience of ‘ownership’, that of the about-to-appear amplified input being attended to. The input activity is amplified by the new attention signal, so accessing its buffer and thereby leading to consciousness of the input. The copy of the attention movement signal contains the information that this is about to happen, so carries ownership of the consciousness of the input. Such a signal can also grant immunity to error through misidentification (11, 14), if the corollary discharge buffer can only let onto the working memory buffer what it has been told to by the inverse attention controller. As such it inhibits all other possible entrants to contentful consciousness on the working memory buffer. This inhibition lasts for a brief period (100-200 msecs) before the attention-amplified input from sensory cortex arrives. The corollary discharge is then inhibited in its turn. Such complex processing is supported by the siting of much of the attention control structures nearby in the parietal lobe, singled out recently as crucial for consciousness to arise (16). A possible control model for this, involving an additional observer component, was given in (17, 18).

This results in the CODAM (corollary discharge of attention movement) model of consciousness (11, 14):

The pre-reflective self is identified as experienced as the corollary discharge of the attention movement control signal residing briefly in its buffer until the arrival of the associated attended input activation arriving at its buffer.

The crucial feature is that the buffered corollary discharge signal is identified with pre-reflective consciousness. This is a stronger claim than a functionalist one – that pre-reflective consciousness is thereby ‘generated’ – since there would still be unknown steps in such generation. It is supported by the immunity to error and PCE experiences, but still needs to be justified by many more features identifiable with other aspects of experience (16).

Attention and the PCE

The PCE state would be created in CODAM by development through meditation of the ability to direct one’s attention solely to one’s own movement of attention: all content arriving from earlier sensory cortices is inhibited The buffered corollary discharge signal is identified with pre-reflective consciousness from gaining access to its working memory buffer, and one attends solely to one’s own attention movement. This ‘inner eye’ of the mind looking at itself has been suggested (although unsuccessfully (10)) in the past as a mechanism for the creation of consciousness; it was not based on attention. Such a process is here given, for the first time, a specific attention-based neural circuitry to achieve it in the brain without the difficulties of an infinite regress (19). The key is to use corollary discharge, with ‘splitting of the attention beam’, so that one component can attend to its own copy. Such ‘inner attention’ also needs to have an associated goal state – that of not possessing any content in experience – set up in the frontal rules module. Frontal activity has indeed been observed during such meditation, so supports the CODAM model (20). The process thus provides specific and testable temporal predictions of neural activity in the appropriate sites.

The CODAM model also explains how ‘a good proportion of perception occurs without awareness.’ (2, p 149), such as is seen in the phenomenon of capture of attention by a cue below the threshold for awareness (9). In this paradigm, a processed but degraded cue accesses the inverse controller module and causes attention to be directed to possible inputs. However the cue would not access the buffer site of conscious content due to its degradation (thereby losing the competition against the associated corollary discharge), so would not be in awareness. Other cases of implicit attention capture (2) can be explained similarly. A comparable explanation can be developed for the attention-catching but unaware degraded signals in blindsight (9).

Conclusions

An overview has been given of recent developments in modelling attention and understanding the physical correlates of meditation states. A control model of the movement of attention, supported by experimental data, was extended to a mechanism for the creation of consciousness through the CODAM model. This led to a tentative computational understanding of the minimal or pre-reflective self. A neuro-scientific underpinning was then arrived at of certain meditation states, claimed by some to be at the basis of consciousness itself. The broad range of the review requires filling in of considerable detail. Some of the many resulting outstanding problems are listed in the set of questions. If the approach is correct, as guided by the material reviewed, it provides a mechanism to help resolve the hard problem concerning human consciousness (21).

By John G. Taylor
Department of Mathematics, King’s College, Strand, London WC2R2LS, UK
Email

© 2002 J.G. Taylor

Conclusions

1. Taylor J. G. (1999, 2001) The Race for Consciousness, MIT Press
2. Simons D. J. (2000) Attentional capture and inattentional blindness. Trends Cognit. Sci. 4, 147-155
3. Mack A. and Rock I. (1998) Inattentional Blindness, MIT Press
4. Gallagher S. (2000) Philosophical conceptions of the self: implications for cognitive science. Trends Cognit. Sci. 4, 14-21
5. Frith C. (1992) The Cognitive Neuropsychology of Schizophrenia, Erlbaum
6. Tassi P. and Muzei A. (2001) Defining the state of consciousness. Neurosci. Biobehav. Rev. 25, 175-191
7. Taylor J. G. (2001) Attention as a Neural Control System. Invited paper, Proc. Int. Jt..Conf. Neur. Nets. 2001, IEEE Press, 2001; Taylor JG & Rogers M (2002) A control model of the movement of attention. Neural Networks 15:309-326.
8. Kentridge B. W. et al (1999) Attention without awareness in blindsight. Proc. Roy. Soc. Lond. B 266, 1805-1811
9. McCormick P. A. (1997) Orienting Attention Without Awareness. J. Exp. Psychol. 23, 168-180
10. Zahavi D. and Parnas J. (1998) Phenomenal Consciousness and Self-Awareness: A Phenomenological Critique of Representational Theory. J. Conscious. Stud. 5:, 687-705
11. Taylor J. G. (2002) Consciousness: Neural theories of. In Handbook of Brain Theory and Neural Computation a, ed. M. Arbib, MIT Press (to appear)
12. Strawson G. (1999) The Self. Ch 1, pp 1-24 in Models of the Self. S. Gallagher & J. Shear (eds), Imprint Academic.
13. Forman R. K.C. (1990) The Problem of Pure Consciousness. Oxford University Press
14. Taylor J. G. (2000) The Central Representation: The Where, What, and How of Consciousness. pp 149-170 in The Emergence of Mind. Fondazione Carlo Erba
15. Desmurget M. and Grafton S. (2000) Forward modeling allows feedback control for fast reaching movements. Trends Cognit. Sci. 4, 423-431
16. Taylor J. G. (2001) The Importance of the Parietal Lobes for Consciousness. Conscious. Cognit. 10, 379-417, 404-406
17. Taylor JG (2002) Paying Attention to Consciousness. Trends in Cognitive Sciences 6:206-210
18. Taylor JG (2002) From Matter to Mind. Journal of Consciousness Studies 9:3-22
19. Shoemaker S. (1868) Self-reference and self awareness. J. Phil. 65, 556-579
20. Lou H. C. et al (1998) A 15O-H2O PET Study of Meditation and the Resting State of Normal Consciousness. Hum. Br. Map. 7, 98-105
21. Chalmers D. (1996) The Conscious Mind: Towards a Fundamental Theory. Oxford University Press