Irreducible Mind

Although the widespread impression is that NDEs occur among patients who have been clinically dead and then resuscitated, they in fact occur in a wide variety of medical circumstances. An examination of medical records in cases in our collection (all of them experiences reported to us retrospectively) showed that slightly more than half of the patients, although ill enough to have been hospitalized, were at no time in danger of dying (Owens, Cook [Kelly], & Stevenson, 1990; Stevenson, Cook [Kelly], & McClean-Rice, 1989–1990); NDEs may therefore occur when patients fear they are dying even if in fact they are not. Moreover, NDEs can also occur when patients are suddenly confronted with death but escape unharmed, as in falls or near-accidents (e.g., Heim, 1892/1972).3
Nonetheless, there remain a substantial number of cases in which patients were clinically near death, such as during cardiac arrest or some other, usually sudden, loss of vital functions (Finkelmeier, Kenwood, & Summers, 1984; Greyson, 2003; Owens et al., 1990; Parnia et al., 2001; Sabom, 1982; Schoenbeck & Hocutt, 1991; Schwaninger et al., 2002; van Lommel et al., 2001). These include some in children who suffered cardiac arrest (Gabbard & Twemlow, 1984, pp. 154–156; M. Morse, 1983, 1994a, pp. 67–69, 1994b; Serdahely, 1990). In one recent prospective study of 344 cardiac arrest patients, 62 reported NDEs following resuscitation from cardiac arrest (van Lommel et al., 2001). In our own collection of retrospectively reported NDEs, out of 114 cases for which we have obtained and rated medical records, 35 were rated “4” on a 4-point scale of severity of condition, meaning that there was some documentation of loss of vital signs, often including cardiac arrest

Many experiences called “near-death” experiences have occurred when the person was not physiologically near death, and individual features associated with NDEs occur in a wide variety of conditions in which the person is also clearly not near death. We believe that the difficulties in explaining, and even in defining, the NDE stem at least in part from the failure to examine this phenomenon within the context of this larger family of related experiences

It was only the investigation of extraordinary circumstances, involving extremely small or large distances, speeds, or mass, that revealed the limits of the Newtonian model and the need for additional explanatory models. So too with the question of the mind-brain relationship: As Myers understood in urging his colleagues in psychology to study subliminal and other unusual psychological phenomena, exploration of extraordinary circumstances may reveal limitations of the current model of mind-brain identity and the need for a more comprehensive explanatory model.

If consciousness be a mere epiphenomenon… accompanying, but in no way guiding, certain molecular changes in the brain, we shall of course expect… that consciousness is exclusively linked with the functional disintegration of central nervous elements, and varies in its intensity with the rapidity or energy of that disintegration. And ordinary experience, at least within physiological limits, will support some view like this. Yet now and then we find a case where vivid consciousness has existed during a state of apparent coma… tranquilly and intelligently co-existing with an almost complete abeyance of ordinary vital function…. Until this new field has been more fully worked—until the traces of memory which may survive from comatose, ecstatic, syncopal conditions have been revived (by hypnotic suggestion or otherwise), and carefully compared, we have no right to make any absolute assertion as to the concomitant cerebral processes on which consciousness depends. (Myers, 1891c, pp. 116–117)

The last 20 years or so of computational modeling of the mind have been dominated by the attempts of these two warring paradigms to refine themselves and work out their relationship. Classical symbol-manipulation modelers, for example, are trying to find ways to make their models more adaptable and less brittle, while connectionists are looking for better ways of generating orderly behavior in the absence of explicit rules. Each paradigm clearly aspires to absorbing the other, but some workers are also exploring “hybrid” computational systems in which the connectionist part takes care of things like perceptual learning and category formation—the subsymbolic level or “microstructure” of cognition—while a more classical symbol-manipulation part deals with things like problem-solving and language. It is not at all clear how things will sort out, although the connectionist faction seems presently ascendant

It is perhaps not surprising, given these properties, that unbridled optimism soon reappeared within the field. One leading connectionist, Smolensky (1988), has proclaimed that “it is likely that connectionist models will offer the most significant progress of the past several millennia on the mind/body problem” (p. 3). Many contemporary psychologists agree, and even some philosophers of mind, including in particular Daniel Dennett and the Church-lands, are no less enthusiastic.
Significant problems have also come to light, however. Although network models are often said to be “neurally inspired,” the current level of neurophysiological realism is typically very low. Both the “neurons” themselves and their connectivity patterns are routinely idealized and distorted, and the most successful learning rule (“back propagation” or generalized delta) still has no generally recognized counterpart in the nervous system (Crick, 1994). Models often have large numbers of free parameters which must be adjusted to fit specific situations, raising doubts about their generality. Similarly, generation of a targeted behavior is sometimes strongly and unrealistically dependent on the exact content and order of previous network experience or training

The fundamental faith of connectionists is that intelligence emerges from the interactions of large numbers of simple processing units organized in a network of appropriate structure. By modifying features of network architecture such as the number of elementary units, the number of layers, their connectivity patterns (feed forward only vs. recurrent), the rules for activating units (simple thresholds, sigmoid functions), and the rules for modifying the connection strengths among units in light of network performance or experience (Hebb, delta, generalized delta), an enormous variety of interesting behaviors can be produced. Networks have proved especially good at some things at which classical models were conspicuously bad, such as pattern recognition, perceptual learning and generalization, and filling in of incomplete input. They also display psychologically interesting and desirable properties such as content-addressable memory and “graceful degradation”

Together, they undoubtedly constitute a major methodological advance for cognitive neuroscience. Indeed, scarcely an issue now goes by of any cognitive neuroscience journal that does not contain one or more papers featuring images outfitted with colored spots identifying regions of “significant” brain “activation” produced by some stimulus or task.13
The final thread is the one most directly germane to our primary subject, the computational theory of the mind (CTM). Specifically, discouragement with the progress of classical or symbolic cognitivism led in the 1980s to an enormous resurgence of interest in a fundamentally different style of computation that also seems more directly comparable to what actually goes on in brains

The general thrust of this work was to suggest that skilled cognitive performances of all kinds characteristically involve cooperation of a number of localized cortical and subcortical regions of the brain, each presumptively specialized for some particular role in the overall performance. Classical cognitivism quickly adapted to this emerging picture of things, assimilating its fundamental theme of the mind as a computational or information-processing mechanism to a “modular” view of its components and internal organization (J. Fodor, 1983; Pinker, 1997)

Anderson argued that since both kinds of representations (and potentially many others as well) could be made internally coherent, and would lead to identical behavioral predictions, a fundamental theoretical indeterminacy had emerged. Considerations of parsimony and efficiency might lead us to prefer one such theory to its competitors, but only physiological observations could potentially determine which was in fact correct. Thus it became evident, more generally, that neurophysiological data can sometimes provide important constraints on psychological theory