1. INTRODUCTION

The intrusion of the observer into physics appeared at the inception of quantum theory eight decades ago. With this "quantum measurement problem," the physics discipline encountered something apparently beyond "physics" (Greenstein & Zajonc 1997).

Photons and electrons manifested "wave-particle duality": They exhibited wave properties or particle properties depending on the experimental technique used to observe them. Wave properties imply a spread-out entity, while particle properties imply a not spread-out entity. The contradiction was perhaps acceptable for these not-quite-real objects seen only as effects on macroscopic measuring apparatus (Heisenberg 1971).

Today, quantum weirdness is demonstrated in increasingly large systems (Haroche & Raimond 2006), and interpretations of "what it all means" proliferate (Elitzur et. al. 2006). Essentially every interpretation ultimately requires the intrusion of the conscious observer to account for the classical-like world of our experience (Squires 1994, Penrose, R. 2005).

The quantum measurement problem is often considered a problem of the quantum theory: How to explain the collapse of the multiple possibilities of the wavefunction to a single observed actuality. This is indeed unresolved (Squires, 1993). However, the measurement problem also arises directly from the quantum-theory-neutral experiment, and depends crucially on the assumption of free will of the experimenter. We present a version of the archetypal quantum experiment illustrating the intrusion of the conscious observer into the experiment.

2. THE ARCHETYPAL QUANTUM EXPERIMENT

According to Richard Feynman, "[The two-slit experiment] contains the only mystery. We cannot make the mystery go away by "explaining" how it works. . . In telling you how it works we will have told you about the basic peculiarities of all quantum mechanics" (Feynman, et. al. 2006).

In the two-slit experiment one can choose to demonstrate either of two contradictory things: that each object was a compact entity coming through a single slit or that each was a spread out entity coming through both. Similar experiments have been done with photons, electrons, atoms, large molecules, and are being attempted with yet larger objects such as live viruses (Clauser 2010). We will just refer to "objects."

We present an equivalent version of the two-slit experiment in which one can choose to show that an object was wholly in a single box (Rosenblum & Kuttner 2002). But one could have chosen to show that that same object was not wholly in a single box. By telling the story with objects captured in boxes, one can decide at leisure which of the two contradictory situations to demonstrate for each isolated object. This displays the quantum challenge to our intuition that an observer-independent physical reality exists "out there." We describe quantum-theory-neutral experiments, by telling only what could be directly observed.

The experimenter is presented with a set of box pairs, say twenty pairs of boxes. Each pair of boxes contains a single object. How the box pairs were prepared is irrelevant for these quantum-theory-neutral experiments. Since it’s easiest to describe the preparation in quantum language, we do that in the next section.

The "Which Box" Experiment The experimenter is instructed to determine, for this set of box pairs, which box of each pair contains the object. He does this by placing each box pair in turn in the same position in front of a screen that an object would mark on impact. He then cuts a narrow slit first in one box of the pair, and then the other. For some box pairs, an impact occurs only on opening the first box, and then not on opening the second box. For others, the impact occurs only on the second opening. In this "which box" experiment, the experimenter thus determines which box of each pair contained the object, and which box was empty. The experimenter establishes that for this set of box pairs, each object had been wholly in a single box of its pair.

Repeating this with box pairs placed in the same position in front of the screen, the experimenter notes a more or less random spread of marks on the screen.

The "Interference" Experiment The experimenter is presented with second set of box pairs. This time he is instructed to cut slits in both boxes of each pair at about the same time. He does so, positioning each box pair in the same position in front of the screen. (It’s an interference experiment, and we so name it. But we make no reference to waves.)

This time the objects do not impact randomly. There are regions where many objects land, and regions where none land. Each object followed a rule specifying the regions in which it was allowed to land.

To investigate the nature of this rule, the experimenter repeats the experiment with different spacings of the boxes of a pair from each other. He finds that the rule each object follows depends on the spacing of its box pair. Each object "knows" the spacing of its box pair. Something of each object therefore had to have been in each box of its pair. The experimenter establishes that–unlike the previous set, for which objects were each wholly in a single box–for this set of box pairs, objects were not wholly in a single box.

The Free Choice Of Experiment The experimenter is reminded that he established that the objects in the first set of box pairs were wholly in a single box, while the objects in the second set of box pairs were not wholly in a single box. Now presented with a third set of box pairs, he is asked to establish whether the objects in this set of box pairs are, or are not, wholly in a single box.

The experimenter arbitrarily chooses to do the "which box" experiment, and thus establishes that this box-pair set had contained objects wholly in a single box. Given another set of box pairs and similar instructions, he chooses the "interference" experiment, and establishes that the objects in this set were not wholly in a single box.

Offered further sets of box pairs, the experimenter finds that each time he chooses to do a "which box" experiment, objects were wholly in a single box. Each time he chooses an "interference" experiment, he establishes a contradictory physical situation, that objects were not wholly in a single box. His free choice of experiment seemed to create the prior history of what had been in the boxes. He’s baffled.

If the experimenter’s choice of experiment were predetermined to match what was actually in each box pair set, he would see no problem. He recognizes this, but he is certain his choices were freely made. His conscious certainty of his free will causes him to experience a measurement problem with the archetypal quantum experiment. In fact, the free choice of the observation set is generally recognized as an essential aspect of any inductive science.

No experiment in classical physics raised the issue of free will. In classical physics, questions of free will arose only out of an ignorable aspect of the deterministic theory.

We emphasize that in describing the intrusion of the observer into the archetypal quantum experiment we never referred to quantum theory, wavefunctions, or waves of any kind. Even were quantum theory never invented, one could do these experiments, and the results would present an inexplicable enigma (Greenstein & Zajonc 1997).

3. THE BOX-PAIR EXPERIMENT IN QUANTUM THEORY

The Preparation Of The Box Pairs Objects are sent one at a time, at a known speed, toward a "mirror" that equally transmits and reflects their wavefunction. In Figure 1, a wavefunction is shown at three successive times. The reflected part is subsequently reflected so that each part is directed into one of a pair of boxes. The doors of the boxes are closed at a time when the wavefunction is within the boxes.

Dividing a wavefunction into well-separated regions is part of every interference experiment. Holding an object in a box pair without disturbing its wavefunction would be tricky, but doable in principle. Capturing an object in physical boxes is not actually required for our demonstration. A sufficiently extended path length would be enough. The box pair is a conceptual device to emphasize that a conscious choice can be made while an object exists as an isolated entity.

Observation In the box-pairs version of the two-slit experiment, the wavefunction spreads widely from the small slit in a box. In the "which box" experiment, it emerges from each single box and impinges rather uniformly on the detection screen. In the "interference" experiment, parts of the wavefunction emerge simultaneously from both boxes and combine to form regions of maxima and minima on the detection screen.

The Born postulate has the absolute square of the wavefunction in a region giving the probability of an object being "observed" there. In the Copenhagen interpretation of quantum mechanics, observation takes place, for all practical purposes, as soon as the microscopic quantum object encounters the macroscopic screen (Griffiths 1995, Stapp 1972). Macroscopic objects are then assumed to be classical-physics objects viewed by the experimenter. Other interpretations of quantum mechanics, attempting to go beyond practical purposes, consider observation to be more involved with the actual conscious experience of the experimental result.

History Creation Finding an object in a single box means the whole object came to that box on a particular single path after its earlier encounter with the semi-transparent mirror. Choosing an interference experiment would establish a different history: that aspects of the object came on two paths to both boxes after its earlier encounter with the semi-transparent mirror. (As noted above, the question of history creation also arose in the quantum-theory-neutral experiment.) Quantum cosmologist John Wheeler (1980) suggested that quantum theory’s history creation be tested. He would have the choice of which experiment to do delayed until after the object made its "decision" whether to come on a single path or whether to come on both at the semi-transparent mirror. The experiment was done with photons and a mirror arrangement much like our Figure 1. Getting the same results as in the usual quantum experiment would imply that the relevant history was indeed created by the later choice of experiment.

For a human to make a conscious choice of which experiment to do takes perhaps a second, in which a photon travels 186,000 miles. Therefore the actual "choice" of experiment was made by a fast electronic switch making random choices. The most rigorous version of the experiment was done in 2007 (Jacques et al., 2007), when reliable single-photon pulses could be generated, and fast enough electronics were available. The result (of course?) confirmed quantum theory’s predictions. Observation created the relevant history.

Non-locality and Connectedness When an object is observed to be in a particular location, its probability of being elsewhere becomes zero. Its wavefunction elsewhere "collapses" to zero, and to unity (a certainty) in the location in which the object was found. If an object is found to be in one box of its pair, its wavefunction in the other box instantaneously becomes zero–no matter how far apart the boxes are.

In its usual interpretation, quantum theory does not include an object in addition to the wavefunction of the object. The wavefunction is, in this sense, the physical entity itself. Thus the wavefunction being affected by observation everyplace at once is problematic in the light of special relativity, which prohibits any matter, or any message, to travel faster than the speed of light. The non-local, instantaneous collapse of the wavefunction on observation poses the quantum measurement problem as viewed from quantum theory.

The instantaneous, non-local collapse of the wavefunction provoked Einstein to challenge the completeness of quantum theory with the famous EPR paper (Einstein, et al. 1935). To avoid what Einstein later derided as "spooky action at a distance," EPR held that there must be properties at the microscopic level that quantum theory did not include. However, since EPR provided no experimental challenge to quantum theory, it was largely ignored by physicists for three decades as merely arguing a philosophical issue.

John Bell (1964) proved a theorem allowing experimental tests establishing the existence of an instantaneous connectedness, Einstein’s "spooky action." The experiments (Freedman & Clauser, 1972; Aspect et al. 1984) showed that if objects had interacted, what an observer chose to observe about one of them would instantaneously influence the result that an arbitrarily remote observer chose to observe for the other object.

The Bell results included the assumption of the free will of the observers, that their choices of what to observe were independent of each other and independent of all prior physical events. Denying that assumption would be "more mind boggling" than the connectedness the denial attempts to avoid. Such denial would imply, Bell wrote: "Apparently separate parts of the world would be conspiratorially entangled, and our apparent free will would be entangled with them" (Bell 1981).

4. THE ROBOT FALLACY

The most common argument that consciousness is not involved in the quantum experiment is that a not-conscious robot could do the experiment. However, for any experiment to be meaningful, a human must eventually evaluate it. A programmed robot sees no enigma. Consider the human evaluation of the robot’s experiment:

5. CONCLUSION

Extending the implications of quantum mechanics beyond the microscopic realm admittedly leads to ridiculous-seeming conclusions. Nevertheless, the experimental results are undisputed, and quantum theory is the most basic and most battle-tested theory in all of science.

The embarrassing intrusion of the conscious observer into physics can be mitigated by focusing on observation in quantum theory, the collapse (or decoherence) of the wavefunction. However, the inescapable assumption of free choice by the experimenter displays the intrusion of the observer in the quantum-theory-neutral quantum experiment, logically prior to the quantum theory.

The intrusion was less disturbing when confined to never-directly-observed microscopic objects. However, the vast no-man’s-land that once separated the microscopic and the macroscopic realms, allowing a tacit acceptance of this view, has been invaded by technology.

Bell’s theorem, and the experiments it stimulated, seems to rule out a resolution of the quantum measurement problem by the existence of an underlying structure, somehow involving only properties localized in quantum objects. An overarching structure, somehow involving conscious free will, seems required (Squires 1991).

Acknowledgments: Parts of this article have been taken from the second edition of our book, Quantum Enigma: Physics Encounters Consciousness, with the permission of Oxford University Press.