5. Describing the Behaviour of a 4 + 1 Dimensional Wave: The Emergence of the Hilbert Space Structure of Quantum Mechanics in Our Framework
In this Section, continuing the study of the behaviour of a 4 +1 dimensional wave in our framework, we will direct our efforts toward describing how a 3 + 1 dimensional observer can model the behaviour of a quantity that changes in reference to the 5th dimension.
In order to do that and in accordance with
Section 3, we turn our attention on a specific point in 4D space
=
(
). If the 4th dimension is treated as space, then at any such point the amplitude A of the 4th dimensional wave (not to be confused with the wave’s angular frequency in the 5th dimension which we also expressed as
above) will oscillate in the 5th dimension with frequency f
= ω
/2π as shown in
Figure 2.
= ().
This means that for a specific point in space and for a specific moment in time, as a 3D observer perceives it, a 4 + 1 dimensional wave would seem to possess many values of A which cannot be known in advance since the observer doesn’t have access to the 5th dimension.
From the very first moment we try to mathematically model a dynamic spacetime, in the sense that we explained above, problems start to arise. More specifically, is it possible for a 3D observer to describe changes that happen in the same moment in time (4th dimension), like the Amplitude oscillation mentioned above?
In order to answer this question and start giving our dynamic spacetime a mathematical foundation, we once again turn to the 3 - dimensional wave of the form given in (1) and we ask a different question which may give us some insight into our problem. Can we model some aspects of the interactions and interferences of 3 dimensional waves without the need of time, only by using space?
Not surprisingly the answer is yes. If these waves all travel with the same speed (
) and all obey the equation:
= f
λ , then we can make predictions about the Amplitude of the wave on a specific point in space in correlation with its Amplitude on another point in space and also make predictions about interference patterns if we know the geometry of the sources and the relative phases of the waves [
2]. This is where complex numbers come into play. For example, in a single wave if we measure the Amplitude (A
1) of the wave in one point in space we can know the amplitude (A
2) of another point at distance dx from the first point by multiplying it with a phase factor in the form:
A2 = Re[] , i = (imaginary unit)
Also, in the case of the double slit experiment for light (
Figure 3), we know that the Amplitude of the interference pattern for any point on the screen is analogous to
+
, where
,
the distances of the slits from the point measured on the screen and
[
2,
9].
If there was an observer oblivious to the concept of our time (4th dimension) in any point of the screen of the double slit experiment, it would seem to him that the Amplitude of the wave can take many possible values. The only way any conclusion or correlation about the wave and its behavior can arise is with the use of complex numbers. Still some information is lost to the observer (like the exact value of the Amplitude because it oscillates with time, which the observer can’t measure or understand) but at least a great portion of the total information of the system would be accessible (for example if there is a constructive or destructive interference like in the double slit experiment for light).
Taking that into account, the observer who can’t understand and measure time would have to make use of complex functions and associate them with observables which the observer can measure and understand such as wavelength λ or energy (if the energy of a wave is proportional to its frequency which is the case for electromagnetic radiation – photons and free fundamental particles). Also, such waves can not be entirely described only by spatial functions (for example ). Using the complex plane gives us a necessary extra degree of freedom, essential for our correlations.
Complex numbers are also essential for quantum mechanics. Experiments have shown that it is impossible to predict experimental results with real-number quantum theory. Also, the use of complex numbers is apparent in the fact that we can’t derive both Planck-Einstein and deBroglie relations (E=hf and p=h/λ) in quantum mechanics without their use.
A 3D wave is oscillating both in space and in time. For two different points (x
1, y
1, z
1, t
1) and (x
2, y
2, z
2, t
2), making precise correlations about the Amplitude in different times is impossible without any information about the time separation t
2-t
1. Analogous to this, if we (the 3D observer) wanted to describe a 4+1 dimensional wave and model its behaviour, the only way we could achieve this would be through the use of complex numbers, using them for correlations together with quantities measurable in the 3D plane (observables) such as distance, time separation or energy. This is where the connection with quantum mechanics in our framework starts to arise, since in quantum mechanics there is also a need for operators (which are measurable quantities) in order to determine the evolution of the quantum state and its expectation values, in reference with the values this quantum state possesses in a different point in space or in time [
4,
5].
More specifically, if we consider a 4 + 1 dimensional wave with a constant angular frequency in the 5th dimension, meaning a constant mass in our framework (which is logical since we try to draw conclusions about the similarity of these waves and quantum mechanical particles with constant mass), the wave’s amplitude would take the form:
Φ(x΄,) = Φ0(x΄) where x΄ = is the 4 dimensional space coordinates (3 + 1 dimensional spacetime for the observer) and is the 5th dimensional time for this wave, not to be confused with proper time in relativity (as we already mentioned in the previous Section).
Analogous to monochromatic classical waves expressed as:
ψ(r,t) = ψ(r)where one can calculate interference and interactions without involving time, only by knowing the spatial part, which is also complex.
Additionally, translations (which refer to shifting a system in 3 dimensional space and time) would only be possible by correlating the 4 dimensional spatial function Φ0(x΄) at a 4D point x΄ with the 4 dimensional spatial function at another 4D point, with the spatial difference between these two points and a measurable quantity that encodes the dynamics of the system in that direction. This is analogous what happens in classical waves where we multiply with phase factors in order to correlate the Amplitude of a wave in one point with the Amplitude of that wave in another point.
Mathematically, such translations would have to be implemented through unitary operators generated by observable quantities. For a small displacement in 4D space () the function Φ0(x΄) transforms as:
Φ0(x΄+) = Φ0(x΄)
where would be an operator that encodes the dynamics of the system in the direction of the displacement.
Since Φ0(x΄) is not the full description of our wave solution, which evolves also in the 5th dimension, its evolution must be state-dependent and must be generated by operators not simple numbers. These operators contain the dynamical rules (e.g., frequency, mass, momentum) that determine how the wave transforms when shifted, just as the momentum operator generates phase shifts in ordinary wave mechanics.
Also, in the classical case some observed physical quantities are dependent on the square of the complex spatial part ψ(r) like intensity (the energy per unit area per unit time transported by the wave).
I(r) ∝ ψ*(r)ψ(r)
This is also the case with quantum mechanics.
All the above show that in our framework since we intend to model the behaviour of 4 + 1 dimensional waves with only spatial components (4 dimensional space for the wave), the most effective thing to do would be to use complex 4 dimensional functions. For this purpose, the natural formalism would be a Hilbert space of complex-valued functions, where:
Observable quantities, which help us make correlations, would be treated as Operators
Expectation values would encode measurable quantities we are interested in measuring and would be calculated by:
= To summarize, the oscillatory wave solution of a harmonic 4 + 1 dimensional wave (which implies a particle with constant mass in our framework):
Φ(x΄,) = Φ0(x΄) naturally suggests that the space of all such solutions forms a complex vector space. We can define a complex vector space , where Φ0(x΄) ∈ and endow the space with an inner product (the superscript “2” comes from the type of integrability condition imposed on the functions in that space – real valued square-integrable functions). Observables are then modeled as linear operators acting on this space, with measurable quantities obtained via:
= This formalism aligns with the Hilbert space structure of quantum mechanics, allowing us to define observables as self-adjoint operators and extract physical quantities through expectation values.
Furthermore, any interaction of such a wave which results in an irreversible exchange of energy or an irreversible change in one or more of the wave’s characteristics would have to be interpreted as a discontinuous update of the 4-dimensional complex-valued function Φ0(x΄) ∈ This update can be modeled via projection operators associated with the eigenstates of a self-adjoint observable . Upon obtaining the outcome, the complex-valued function Φ0(x΄) collapses to the new state’s corresponding eigenfunction and all future 5th dimensional evolution proceeds from this new state. This process parallels the standard collapse postulate of quantum mechanics referring to the quantum measurement problem.
All of the above demonstrate that the use of quantum formalism in our framework is not merely an analogy, but a mathematically necessary structure. Moreover, quantum behavior emerges naturally from the underlying dynamics, rather than being introduced through external postulates. This is also a key difference between our framework and other classical higher dimensional theories where quantum correlations are not natural outcomes and have to be imposed by turning classical fields into quantum fields, promoting classical observables into operators and then defining probabilistic behavior through a Hilbert space, that does not emerge naturally from the theory.
Finally, we will try to give the most basic form of an equation in our framework by attempting to model a pure 4 +1 dimensional wave of a Scalar Field (Φ), which does not interact with lower dimensional disturbances of itself or any other field and propagates in a harmonic way. The magnitude of its rate of transmition is taken to be equal to the magnitude of the speed of light. For this wave the 5th dimension is acting as time and the 4th dimension (our time dimension) is acting as another spatial dimension. We are interested in modelling this wave in a way that makes sense to us, the 3 + 1 dimensional observer, following the same rules we imposed on the previous Sections.
This equation would take the form:
(Second time derivative term) = (rate of transmition)2 x (second spatial derivative)
The following apply:
- The wave is 4 + 1 dimensional which means that time for this wave is the 5th dimension
- For this wave our time (the 4th dimension) is behaving as a spatial dimension. For this reason, our time dimension will be included in the spatial derivative terms
- Since we are modelling the wave’s behaviour in the reference frame of a 3 + 1 dimensional observer, all quantities and relations must apply to that reference frame.
Taking all these into account our equation should have the form:
Second 5th dimensional derivative term = (second spatial derivative terms) (10)
Considering that the wave propagates in a harmonic way in the 5th dimension and taking a sinusoidal solution Φ, the second 5th dimensional derivative term will be in the form: Φ (11)
(where is the wave’s angular frequency in the 5th dimension)
The second spatial derivative terms will now include 3-dimensional time (4th dimension) and we will again make use of the Minkowski metric (metric tensor for flat spacetime) because we want the results to have a physical meaning to us the 3 + 1 dimensional observer. This means that the spatial derivative terms will take the form [
6]:
Φ (12)
Combining (10), (11) and (12) we get:
Φ = Φ →
Φ Φ = 0 →
which is equal to the Klein – Gordon equation if we consider that:
→ The same result for as the one we derived earlier!
This is very promising since the mass term we derived by alternative means in
Section 4 is identical to the mass related component in the Klein – Gordon and Dirac equations [
7,
8]:
(Klein–Gordon equation)
+ + + (Dirac equation)
The exact same result could alternatively be obtained by taking the solution proposed above:
Φ(
,
) = Φ
0(
)
differentiating and then factoring out the term
Also, in QFT the mass term is recognized as a term in the Lagrangian that is quadratic in the field and has the form
for some
(
∝ m the mass of the particle) [
7,
8]. Since we can not exactly model the behaviour of 4+1 dimensional waves as we showed in this Section and need to make correlations with observables, such a term would make sense to appear in any attempt of the 3D observer to model the dynamics and possible interactions of those waves with themselves and other lower dimensional waves.
What makes the present work fundamentally depart from other Kaluza – Klein theories, where mass is interpreted as momentum in the extra dimension is that in our approach the fifth dimension is not an extra spatial coordinate but represents a dynamical temporal dimension. This means that we do not need to rely on compactification or gauge unification and then impose quantization in order to make the equation quantum, but instead the formalism derived earlier both provide quantum behavior and compatibility with special relativity which is emergent from the framework itself.
Additionally, in other higher dimensional theories the values of the field in the fifth dimension correspond to different events in 5D spacetime and do not collapse into one 4D event unless some mechanism (e.g. compactification, brane confinement, integral over the 5th dimension) forces that [
16]. This means that in those theories both the deduction of the four-dimensional Klein-Gordon equation from a five-dimensional wave equation of zero (hyper-)mass and the superposition of different values of that field in one 4D event do not work unless we use a methodology or mathematical trick external to the theory. Our framework has no use for that.
Regarding interactions between fields, in a framework such as the one we propose here, the only meaningful way a 3D observer (who cannot measure changes in the 5th dimension) could model interactions, transitions, and conserved quantities regarding different fields of different dimensionality would be via continuous symmetries. Those are captured mathematically by Lie groups and their associated Lie algebras. These symmetries allow the reduction of redundant degrees of freedom, projecting the behavior of the 4+1 field into an effective 3+1 dimensional description recognizable by conventional physics.
Moreover, since the framework utilizes complex-valued field solutions to express correlations between field disturbances, the symmetries must act in a way that preserves the inner product and phase information inherent to these complex structures. This naturally leads to the use of unitary groups (U(n)) and, more specifically, their special unitary subgroups (SU(n)), whose Lie algebras provide a minimal and elegant structure for capturing interaction patterns while preserving probabilistic and norm-based interpretations. These unitary symmetries are essential for describing interactions that conserve physical quantities in a complex-valued setting.
In general, the number of generators in a symmetry group reflects the number of continuous transformations the system admits. However, this correspondence does not always align with the physical dimensions of the system. Especially in the context of our framework, since we want to describe the internal structure and the richness of field correlations rather than geometric rotations alone, the groups that describe theses should have more generators than the number of spatial dimensions. Also, it is well known that when projecting higher dimensional symmetries on a lower dimensional plane, these symmetries can either be broken, hidden, or reinterpreted as internal symmetries in the lower-dimensional effective theory. All the above considerations closely parallel the mechanisms encountered in QFT.
Finally, our framework provides a natural avenue for integrating different types of physical equations (including those that treat time and space asymmetrically) due to the dynamic role of the time dimension in our framework. Since time is not a fixed dimension for all phenomena (fields and their disturbances) and interactions but the +1 dimension relative to a field disturbance propagation, the mathematical form of energy dispersion may vary with dimensionality.
Take for example the case of the heat equation or diffusion in general.
= aIt is first-order in time but second-order in space, because it models irreversible energy dispersion that depends on spatial imbalance (curvature) but has no memory or oscillatory behavior. By contrast, the wave equation is second-order in both time and space because waves involve oscillations and the acceleration (2nd time derivative) is tied to spatial curvature, reflecting the symmetric, oscillatory nature of wave propagation.
In our framework, distinctions between how space and time are treated for the same interaction may emerge naturally from the dimensional context of the interaction. For example, a higher-dimensional field could experience diffusion therefore treating lower dimensional time in a second-order, while a lower-dimensional field might also evolve in a diffusion-like, first-order way in its own time coordinate. This opens the door to a geometric reinterpretation of the Schrödinger equation (first-order in time), the Dirac equation (first-order in time and space), and even non-Hermitian dissipative systems.
Therefore, the dynamic time concept allows us to consider hybrid evolution equations (first-order in some dimensions, second-order in others) all governed by how energy propagates and disperses across interacting fields of different dimensionalities. The consideration of different fields of different dimensionalities and their interactions may also eliminate constraints between the gradients along the different coordinates (since time in one field can behave as space in another), potentially accommodating the integration of distinct symmetries and symmetry violations across different interactions. This line of thinking may allow further generalization and unification of quantum, classical, and dissipative dynamics under a common higher-dimensional geometric structure, though this is much beyond the scope of this paper.
It should be emphasized that the preceding two arguments concerning the role of symmetries and the integration of differential equations of differing order across dimensions do not constitute a proof of the validity of the proposed framework. Rather, they demonstrate its conceptual compatibility and structural consistency with established principles in contemporary theoretical physics, particularly those underlying modern field theories.
Until now we have been focusing on wave solutions in our framework and not on system dynamics. This has been intentional because we want to focus on how both relativistic and quantum mechanical behaviour emerges from these solutions. However, the inherently dynamic structure of our model also provides a natural avenue for examining the emergence of physical laws from first principles. Let’s consider an example:
Suppose that in the context of our framework, every 4 + 1 dimensional field experiences a harmonic restoring force, aiming to restore the field to a minimum value. This would produce a net force per unit 4D volume (spacetime volume) acting on 4 + 1 dimensional fields that would take the form:
Assuming the restoring force acts uniformly in all 4D spatial directions, this would produce an isotropic pressure component p(r) in the energy-momentum tensor. This would result in a 4D spatial symmetry, or more specifically in the 3 + 1 dimensional observer’s reference frame in:
a spherical 3 dimensional space symmetry, where depends only on the radius r = a time symmetry where does not depend on the 4th dimension.
Taking the above into account the covariant energy-momentum tensor would have components:
ρ(r) ,
ρ(r)
Now according to the approach we took in
Section 4 where we expressed a tensorial refractive index:
and a normalized propagation vector:
by combining it with the wave propagation relation in
Section 3:
we get:
→
Assuming the form of the energy-momentum tensor we expressed above, this equation takes the form:
We now take the wave solution we proposed earlier in this Section:
Φ(,τ) = Φ0() Assuming harmonic propagation in the 5th dimension, we could solve the Klein - Gordon equation with the mass term derived earlier:
Φ0() Φ0() = 0 →
Φ0() Φ0() = 0
Taking into account the 3D spherical symmetry and the time symmetry the system has, the complex solution Φ0() would be variable separable and would take the form:
Φ0() = then
Φ0() = Φ0()
Therefore, the Klein Gordon equation above produces the radial equation:
This is a nonlinear eigenvalue equation for , because the term acts as a potential that depends on itself and is characteristic of the solution itself.
Exploring the case where ρ(r) decays exponentially with r, which is logical if we take into account the symmetries we imposed earlier
Where α is an arbitrary parameter which can be modified according to our needs
This creates a confining exponential potential in the wave equation, as r → ∞. The potential grows exponentially and the equation becomes:
This is similar in form to an inverted potential barrier. It forces the solution u(r) to decay rapidly as r→∞ , ensuring normalizability.
Only certain values of will satisfy this, therefore only certain values of mass survive. This quantization is not arbitrary, it is mathematically grounded in Sturm–Liouville theory generalized to nonlinear eigenvalue problems, a class of equations well known to yield discrete solutions under appropriate symmetry and boundary constraints. This ultimately leads to quantized mass levels, action units and frequency modes. This means that quantization may not need to be postulated externally in our framework but could emerge from the deeper structure of the dynamics of 4 + 1 dimensional fields. Furthermore, discrete values of arising from a Sturm–Liouville-type equation directly imply spatial confinement. The solutions are localized (contained) within regions where the effective potential allows oscillatory behavior. This localization is not optional, it is mathematically required by the nature of the eigenvalue problem.
Key issue that requires clarification: The adoption of the metric tensor with the above components in our approach may raise questions for readers well-versed in the subject. More specifically, since the 4+1 dimensional wave may move in any direction in the 4 dimensional space (or more specifically the unit 4 vector of its trajectory can take any value permitted) why did we take the metric’s elements to be proportional to the 3D radius r, which corresponds to a stationary in 3D 4-vector and then took its product with The answer is that we did that for simplicity and in order to make the result easier to visualize for the average reader. In fact, we could as well take the metric’s elements be:
ρ(ds) , ρ(ds)and a unit vector that is parallel to the displacement 4-vector Then, the product becomes equal to ρ(ds) and corresponds to the energy density observed in the frame moving along ds. The key result then would still be the same.
Conceptually the key issue that causes all this misconception is that the restoring force is produced by the disturbance of the field itself, which is always in motion and propagating parallel to the unit vector
For this reason we either have to adopt a symmetry or a projection parallel to the displacement 4-vector
Subtle note from the author: At this point it would be tempting to formulate a specific type of
with certain values of
and α so that the solution of the equation above would produce a
ratio and therefore a mass ratio that matches or approximates the mass ratio of either lepton or quark patterns or even propose a certain ratio that matches neutrino masses and then claim that this is an indicator of the testability of the framework. However, we deliberately refrain from doing so for two reasons. First of all, because as we mentioned in
Section 4 the equation we used to connect the normalized propagation vector with the stress energy momentum tensor
is only a first approach and we do not take into account 3 + 1 dimensional fields or interactions or potential coupling constants. Secondly, because it would not really be a true indicator of the testability of the framework, but rather a numerical fit of a potential term to observations. The true value or what we did above does not lie in the specific values, which can potentially be fitted to observation results, but in the fact that quantization of mass can be produced through the dynamics of 4 + 1 dimensional fields. This is where the compatibility of our framework with physical reality arises and where we get another indicator that quantum mechanical behavior can arise as a consequence of higher-dimensional structure in our framework.