His main fields of work are nonlinear dynamics, complex systems, and psychophysical dualism. He is, with Hans Primas, an editor of Recasting Reality: Wolfgang Pauli's Philosophical Ideas and Contemporary Science and he wrote the Stanford Encyclopedia of Philosophy page on Quantum Approaches to Consciousness.

With Robert Bishop, Atmanspacher developed the theory of "contextual emergence," in which the description of properties of a lower level of emergence offers necessary but not sufficient conditions to rigorously describe the properties of the next higher level. The sufficient conditions are provided by "contextual" and contingent conditions.

Atmanspacher distinguishes between levels of description and levels of reality, i.e. between epistemological and ontological frameworks for reduction and emergence. He says:

Broadly speaking, descriptive terms are subjects of epistemological discourse while elements of reality are subjects of ontological discourse. Both types of discourse are used in reductionist and emergentist approaches. The concept of reference establishes a connection between descriptive terms and described elements of reality (leaving aside difficult questions about reference itself).

If one wants to have the option of ontic elements at each level of description rather than only at one or a few fundamental levels, a straightforward and strictly reductive scheme for interlevel relations becomes impossible and must be relaxed. The way in which ontic and epistemic descriptions are related to each other motivates contextual emergence as a viable alternative.

In order to clearly distinguish between different concepts of reduction and emergence, it is desirable to have a transparent classification scheme, so that the basic characteristics of these concepts can be discussed coherently. A useful approach toward such a classification is based on the role which contingent contexts play in reduction and emergence. More precisely, the way in which necessary and sufficient conditions are assumed in the relation between different levels of description can be used to distinguish four classes of relations:

("Contextual Emergence from Physics to Cognitive Neuroscience," Journal of Consciousness Studies 14(1/2), 18–36 (2007))

Atmanspacher situates contextual emergence in a matrix of related pictures of reductionism and emergence. He says the way in which necessary and sufficient conditions are assumed in the relation between different levels of description can be used to distinguish four classes of relations:

(1) The description of features of a system at a particular level of description offers both necessary and sufficient conditions to rigorously derive the description of features at a higher level. This is the strictest possible form of reduction. It was most popular under the influence of positivist thinking in the mid-20th century.

(2) The description of features of a system at a particular level of description offers necessary but not sufficient conditions to derive the description of features at a higher level. This version is called contextual emergence, because contingent contextual conditions are required in addition to the lower-level description for a rigorous derivation of higher-level features.

(3) The description of features of a system at a particular level of description offers sufficient but not necessary conditions to derive the description of features at a higher level. This version includes the idea that a lower-level description offers multiple realizations of a particular feature at a higher level, which is characteristic of supervenience.

(4) The description of features of a system at a particular level of description offers neither necessary nor sufficient conditions to derive the description of features at a higher level. This represents a form of radical emergence insofar as there are no relevant conditions connecting the two levels whatsoever.

("Contextual Emergence from Physics to Cognitive Neuroscience," Journal of Consciousness Studies 14(1/2), 18–36 (2007))

In a recent article, Atmanspacher makes the argument that genuine chance may be involved in sub-cellular brain processes, although neural networks may be adequately determined.

Any distribution characterizing a state can in principle be interpreted as due to genuine chance (ontically) or due to ignorance (epistemically). This ignorance, or missing information, can in turn be deliberate, e.g. in order to disregard details that are inessential in a particular context, or it can be caused by uncontrollable perturbations. At the ion channel level, where quantum effects must be expected to occur, an ontic interpretation in terms of indeterminate states is possible or likely.

However, the fact that the stochastic dynamics of ion channels typically yields highly reliable and precise neuronal behaviour suggests that any potentially genuine micro-stochasticity is inefficacious at the neuronal level (and even more so at the network level). Therefore, statistical neuronal states are assumed to be of epistemic nature and genuine indeterministic contributions to the dynamics of neurons seem to be of low relevance. After all, the representation of the neurodynamics in a neuronal state space amounts to a fairly well-defined trajectory of quasi-ontic states...

Since stochastic descriptions can generally be transformed into deterministic descriptions, it is an important issue whether or not there is a genuinely stochastic neuronal dynamics for which a (hidden) deterministic interpretation can be excluded. To all our present knowledge this is the case for stochastic quantum processes, for which the Schrödinger dynamics breaks down so that they are to be conceived as ontologically indeterministic.

Although the neurobiological literature does not refer to quantum phenomena a lot, it would be premature to assume that indeterministic quantum processes do not occur on relevant scales in the brain.

("On determinacy or its absence in the brain," Free Will and Modern Science, ed. R. Swinburne, 2011, p.94-96)

"Two-stage models of free will" combine ontologically indeterministic quantum processes as a low-level generator of noise in the brain with more macroscopic but still stochastic processes. They use quantum-level indeterminism in the first stage, which generates alternative possibilities for action. These are evaluated in the second stage, which is a "willful" and "adequately" determined process that considers motives, reasons, and desires to choose between the "freely" generated alternatives of the first stage.

The two-stage model can be described simply as first chance, then choice (cf. the book Chance and Choice, ed. by Atmanspacher and Bishop), or first "free" then "will." As it was
first described by William James, it is a model of "mental evolution" of creative new ideas analogous to the two-step process of Darwinian evolution of new species.

An "Undetermined liberty" is one chosen indeterministically from multiple equally good alternatives
since they all are equally well-motivated (see Robert Kane's "torn decision").

The classic starting point for most contemporary discussions of mind-matter relations is Descartes' ontologically conceived dualism of the mental (res cogitans, thought) and the material (res extensa, extended matter). In the history of philosophy, Descartes' position was immediately upgraded, criticized, or replaced by essentially three forms of thought: (i) alternative dualistic approaches (occasionalism, parallelism), (ii) essentially monistic approaches (idealism, materialism), and (iii) approaches combining (i) and (ii) by assuming a monistic domain underlying the mind-matter distinction. An early protagonist of this latter view is Baruch de Spinoza.

Spinoza's monism provides an elegant and robust sense in which mind and matter are related to a "unity of essence". It does so by concatenating an ontological monism with an epistemological dualism, yielding an overall worldview in which both philosophy and the sciences can find appropriate places and mutual relations. This framework began to be explicitly exploited in the mid 19th century, by both philosophers and scientists, and today we can recognize two main reactions to Spinozism, called dual-aspect monism and neutral monism.

Unfortunately, there is no authoritative delineation of the two - the many versions of dual-aspect monism and neutral monism that are around today have a tendency to blend into each other in ways that make clear assignments to one or the other problematic. This can be seen in Stubenberg's (2010-SEP) excellent overview, see also Silberstein's (2009) taxonomy and Seager's (2009) discussion, among others. There are important commonalities but also key differences between dual-aspect monism and neutral monism:

1. For neutral monists both mind and matter reduce to an underlying, neutral domain, while for dual-aspect monists mind and matter are two basic and irreducible aspects of that underlying domain. In both frameworks, the underlying, ontic domain (and its elements, respectively), is mostly characterized as neither mental nor material.

2. For neutral monists mind and matter are separate insofar as particular configurations of elements of the neutral domain underlie the mental while other, distinct configurations of elements of the neutral domain underlie the material. For dual-aspect monists the underlying, ontic domain manifests itself generally and equally under both aspects, tied together inseparably.

An interesting implication of these two criteria is that the aspects in dual-aspect monism are not a priori given, but depend on epistemic issues and contexts. Distinctions of aspects are generated by "epistemic splits" of the distinction-free, unseparated underlying domain, and in principle there can be as many aspects as there are contexts. This is at variance with neutral monism where the mind-matter distinction is assumed to be preformed in the neutral domain: particular configurations of neutral elements underlie the mental while other, distinct configurations of neutral elements underlie the material.

Moreover, the two criteria entail an interesting further dividing line between prevalent philosophical inclinations toward the underlying domain. For dual-aspect monists this domain is apprehensible only indirectly, through the aspects, while neutral monists deny this restriction. Therefore, it is natural for dual-aspect monists to nurture metaphysical conceptions of the underlying domain, while neutral monists typically refer to direct, basic, pure, raw modes of apprehending it, for instance experientially or phenomenologically.

Stubenberg's (2010) review clearly supports this observation. Ernst Mach, William James, and Bertrand Russell, the forefront of neutral monism, refer to "sensation", "pure experience", and again "sensation", respectively, concerning the neutral domain, and these notions are redigested by other neutral monists such as Avenarius, Petzoldt, Perry, Holt, Sayre etc. Their empirical, hence anti-metaphysical, inclination explains why their notions of the neutral domain all bear the risk of confusing the neutral with some mental capacity. Assuming that the neutral can be apprehended directly, how could it be apprehended if not mentally?

By contrast, dual-aspect monists do not hesitate to embrace ontology and metaphysics. Since their underlying domain (substantial or processual) is conceived of metaphysically, it cannot be apprehensible in any direct way but manifests itself in its aspects. Beyond the well-known historical representatives of dual-aspect monism such as Spinoza, Fechner, Schopenhauer, and others, a number of scientists, notably physicists and psychologists, have explored the dual-aspect route since the mid 20th century. Needless to say, none of their attempts has resolved all aspects of the mind-matter relation. In the following some of these approaches will be presented. The subsequent sections will elaborate on one of them, due to Pauli and Jung, and outline how it may be potentially viable.

("Dual-Aspect Monism à la Pauli and Jung," Journal of Consciousness Studies 19(9-10) (2012), section 1.1)

Atmanspacher connects dual-aspect monism with the important concept of complementarity, one of many great dualisms.

The notion of complementarity was originally coined by William James (1890, p. 206) and adopted by some psychologists, for instance referring to the bistable perception of ambiguous stimuli. Bohr imported it into physics, originally with the purpose of replacing the term wave-particle duality, in his "Como Lecture" in 1927 (Bohr 1928, p. 566). But his extensive later writings about complementarity make it clear that Bohr's preeminent concern was to extend the idea of complementarity beyond physics. In the same spirit, Pauli (1950, p. 79) advanced the opinion that the "issue of complementarity within physics naturally leads beyond the narrow field of physics to analogous conditions of human knowledge".

According to Bernays (1948), there are two different types of complementarity already in quantum physics. One of them refers to incompatible observables, formally grounded in the non-commutativity of corresponding operators acting on quantum states. This non-commutativity entails all kinds of typical quantum features such as superpositions, quantum probabilities, indeterminism, uncertainty relations, and the violation of Bell inequalities...

The second kind of complementarity according to Bernays opens up a wider scope. It refers to conceptual issues such as the quantum physical complementarity of wave and particle pictures: Two descriptions are complementary if they mutually exclude each other, yet are both necessary to describe a situation exhaustively. Complementarity in this sense refers to incompatible aspects which cannot be combined in a single description based on a purely Boolean, two-valued logic... The feature of complementarity in a non-Boolean logic means that a proposition and its complement pertain to two aspects of a situation that are incompatible. They are both together necessary to describe the situation exhaustively. Neither one of them alone is sufficient, yet observing one of them in a given empirical context excludes observing the other one in the same context. For a dual-aspect monism, where the underlying domain is neither physical nor mental (cf. Sec. 1.1), complementarity thus implies that either the physical or the mental aspect is accessible in a given empirical context, although both of them are necessary for a complete picture.

When Pauli (1952, p. 164) says that "it would be most satisfactory if physis and psyche could be conceived as complementary aspects of the same reality", he proposes a dual-aspect monism precisely in this sense...

("Dual-Aspect Monism à la Pauli and Jung," section 1.3)

(1) the relation between local realism and holism in (quantum) physics,
(2) the relation between consciousness and the unconscious in Jung's psychology,
(3) the common, psychophysically neutral ground of both the mental, conscious realm and the physical, local realm, and
(4) the relation between these realms as a consequence of or as mediated by their common ground.

One of the central problems, if not the problem, of quantum mechanics is the process of measurement. Although much progress has been achieved with respect to its understanding since the early days of quantum mechanics, the problem is still not completely solved. However, empirical results and modern formulations of quantum theory allow us to state it in a way that is more precise than ever before. From a conceptual point of view, measurement can be viewed as an intervention decomposing a system constituting an inseparable whole into locally separate parts.

The empirical cornerstone of our understanding of this decomposition involves so-called nonlocal correlations (Einstein et al. 1935, Bell 1964, Aspect et al. 1982). They are generic in any system requiring a description in terms of non-commuting observables. These correlations can be measured in suitable experiments and indicate post festum that the measured system was in a holistic state before measurement. Conceptually, this means that one can indirectly infer knowledge about an unmeasured state by the result of a controlled intervention into that state due to measurement. At the same time, this controlled intervention entails that the observed system changes its state in a basically uncontrollable way (Bohr 1935).

It is tempting to say that such nonlocal correlations correlate everything with everything else, thus suggesting a holistic concept of reality through and through. But this would be misleading without precise qualifications. Quantum holism is only one among two reality concepts that modern quantum theory requires. Equally important is the ('common sense") concept of a local reality which was considered to be the reality for centuries of physicists from Newton to Einstein. As Bohr has emphasized over and over, local realism is unavoidable for a proper description of experiments and their results by Boolean (yes-no) propositions.

Today we know that both concepts together are necessary for a comprehensive description of reality, neither of them is sufficient on its own. In the framework of algebraic quantum theory, the dierence between them can be mathematically formalized and clearly understood by two different state concepts: those of ontic and epistemic states...

While epistemic states are those states to which epistemic, i.e. empirical access is possible by measurement (and observation in general), ontic states are supposed to characterize the system independent of its observation and our resulting knowledge. One may wonder why it is useful to have an ontic level of description for which empirical (or operational) access is no option at all. However, a most appealing feature at this ontic level is the existence of rst principles and universal laws that are unavailable in an epistemic description. From such an ontic level, it is possible to deduce proper epistemic descriptions given enough details - contexts as it were - about empirically given situations.

The distinction of ontic and epistemic states provides an important clue to understand the distinction between a holistic and a local concept of reality. Ontic states and associated intrinsic properties refer to the holistic concept of reality and are operationally inaccessible, whereas epistemic states and associated contextual properties refer to a local concept of an operationally accessible reality. The process of measurement represents the link between the two. Measurement suppresses the connectedness constituting a holistic reality and generates approximately separate local objects constituting a local reality.

Although this is a fairly modern picture, it also has a conservative aspect:

Many of the unothodox "interpretations" of quantum mechanics depend on observers

Quantum theory as of today does not at any place refer to the mental world of human observers, to their cognitive capabilities or psychological condition in general. The standard view in quantum theory is that measurement should be treated in terms of an interaction between an observed system and its environment, including the observing device(s). For instance, Heisenberg (1936) was very explicit about this, talking about a "cut between the system to be observed and the measuring devices". And Pauli (1957) says: "As Heisenberg has emphasized, quantum mechanics rests on a sharp cut between observer or instrument of observation on one hand and the system observed on the other."

Hugh Everett's "relative-state interpretation" of quantum mechanics employs non-animate machines to make "measurements"

In general, the idea is that any inanimate environment can be understood as a "measuring device", though in a non-intentional manner. No consciousness is necessary for measuring a quantum state. On the other hand, as soon as controlled experiments are considered, it is clear that issues like the design of an experiment, the choice of observables of interest, or the interpretation of the results of a measurement play crucial roles. They depend on decisions based on the intentions of human observers and are not part of the formalism of quantum theory.

In this context, Pauli speculated in a letter to Fierz of August 10, 1954

It might be that matter, for instance considered from the perspective of life, is not treated "properly" if it is observed as in quantum mechanics, namely totally neglecting the inner state of the "observer". (...) The well-known "incompleteness" of quantum mechanics (Einstein) is certainly an existing fact somehow-somewhere, but of course it cannot be removed by reverting to classical field physics (that is only a "neurotic misunderstanding" of Einstein), it has much more to do with holistic relationships between "inside" and "outside" which contemporary science does not contain.

("Dual-Aspect Monism à la Pauli and Jung," section 2.1)

Conceiving the mind-matter distinction in terms of a splitting of a psychophysically neutral domain implies correlations between mind and matter as a direct and generic consequence. It is important, though, to stress right at the outset that these correlations are not due to causal interactions (in the sense of efficient causation as usually looked for in science) between the mental and the material. In a dual-aspect framework of thinking it would be wrong to interpret mind (or mental states) as directly caused by matter (or material states) or vice versa.

Pauli and Jung discussed such correlations extensively in their correspondence between June 1949 and February 1951 when Jung drafted his article on "synchronicity" for the book that he published jointly with Pauli (Jung and Pauli 1952). In a condensed form, two (or more) seemingly accidental, but not necessarily simultaneous events are called synchronistic if the following three conditions are satisfied.

1. Each pair of synchronistic events includes an internally conceived and an externally perceived component.

2. Any presumption of a causal relationship between the events is absurd or even inconceivable.

3. The events correspond with one another by a common meaning, often expressed symbolically.

The first criterion makes clear that synchronistic phenomena are psychophysical phenomena, intractable when dealing with mind or matter alone. The second criterion repeats the inapplicability of causation in the narrow sense of a conventional cause-and-effect-relation. And the third criterion suggests the concept of meaning as a constructive way to characterize mind-matter correlations. Since synchronistic phenomena are not necessarily temporally "synchronous" (in the sense of "simultaneous"), synchronicity is a somewhat misleading term. For this reason Pauli preferred to speak of "meaningful correspondences" ("Sinnkorrespondenzen") under the influence of an archetypal "acausal ordering". He considered both Jung's synchronicity and the old teleological idea of finality (in the general sense of a process oriented toward a goal) as particular instances of such an acausal ordering which cannot be set up intentionally. In contrast, the mathematical notion of "blind" chance (referring to stochastically accidental events) might be considered as the limiting case of a meaningless correspondence.

Similar to their idea of complementary notions of effcient causation and meaningful correspondence, Pauli and Jung discussed a possible complementarity of statistical limit theorems and singular synchronistic events. The upshot of this proposal is that synchronistic phenomena cannot be corroborated by statistical methods as they are usually applied. In a letter to Fierz of June 3, 1952, Pauli wrote (von Meyenn 1996, pp. 634{635):

... synchronistic phenomena ... elude being captured in natural "laws" since they are not reproducible, i.e., unique, and are blurred by the statistics of large numbers. By contrast, "acausalities" in physics are precisely described by statistical laws (of large numbers).

And in his "Lecture to the Foreign People" (Atmanspacher et al. 1995, p. 326), where he sketches some of his ideas about biological evolution, he states his impression that

external physical circumstances on the one hand and corresponding adaptive hereditary alterations of genes (mutations) on the other are not connected causally-reproducibly, but occur - correcting the "blind" chance fluctuations of the mutations - meaningfully and purposefully as inseparable wholes together with the external circumstances.

According to this hypothesis, which differs from both Darwin's and Lamarck's conception, we encounter the requested third type of natural laws, consisting of corrections of the fluctuations of chance due to meaningful or purposeful coincidences of non-causally connected events.

What Pauli here postulates is a kind of lawful regularity beyond both deterministic and statistical laws, based on the notion of meaning and, thus, entirely outside the natural sciences of his time and also, more or less, of today...

For the mindset of a psychologist like Jung, the issue of meaning is of primary signicance anyway. For a long time, Jung insisted that the concept of synchronicity should be reserved for cases of distinctly numinous character, when the experience of meaning takes on existential dimensions. With this understanding synchronistic correlations would be extremely rare, thus contradicting their supposedly generic nature. Only in later years, Jung opened up toward the possibility that synchronicity might be a notion that should be conceived as ubiquitous as indicated above (Jung 1969, par. 440):

As soon as a psychic content transgresses the threshold to consciousness, its synchronistic byproducts disappear. Space and time resume their accustomed sway, and consciousness is again isolated in its subjectivity. This is one of those cases which can best be captured by the term "complementarity", known from physics. When an unconscious content trespasses into consciousness, its synchronistic manifestation ceases and, conversely, synchronistic phenomena can be elicited by putting a subject into an unconscious state (trance). The same relation of complementarity can be observed in those frequent medical cases in which particular clinical symptoms disappear when their corresponding unconscious contents become conscious. We also know that a number of psychosomatic phenomena, otherwise outside the control of volition, can be induced by hypnosis, i.e. by an attenuation of consciousness.

In this regard, the development of Pauli's and Jung's views about archetypes and their role in manifesting synchronicities suggests a distinction between two different kinds of synchronistic events which I suggest to denote as "structural" versus "induced". Structural synchronicity refers to the role of archetypes as ordering factors with a strictly unidirectional influence on the material and the mental (Pauli's letter to Fierz of 1948, see Sec. 2.3). Induced synchronicity refers to the uncontrollable backreaction that changes of consciousness induce in the unconscious and, consequently, in the physical world as well. This way, the picture is extended to a bidirectional relation (Pauli's letter to Jung of 1954).

It is important to keep in mind that even in induced synchronicity, not only in structural synchronicity, there is no direct causal relation from the mental to the physical (i.e. no direct "mental causation") or vice versa. The problem of an "interaction" between categorically distinct regimes is thus avoided. Moreover, I should stress that the "meaning" connecting synchronistic events, although being subjectively ascribed (by the experiencing subject), is not completely arbitrary. It depends on the situation as a whole, including conditions that are not consciously available to the subject.

("Dual-Aspect Monism à la Pauli and Jung," section 2.4)

• Mind and matter as asymptotically disjoint, inequivalent representations with broken time-reversal symmetry
• The Hidden Side of Wolfgang Pauli
• Contextual Emergence from Physics to Cognitive Neuroscience
• Contextual Emergence in the Description of Properties (with Robert Bishop)
• Video seminar: Contextual emergence of mental states from neurodynamics (2008)
• Video: Does quantum physics go beyond causality?
• Dual-Aspect Monism à la Pauli and Jung