Pinning down the origin of collective phenomena at the LHC
The origin of Quark Gluon Plasma-like (QGP-like) phenomena within small collision systems has been a puzzle for researchers for over a decade. Throughout these years of investigation, the presence of collective effects has been confirmed. Moreover, numerous theoretical approaches have been attempted to describe these phenomena. Nonetheless, it has been determined that the phenomena cannot be adequately explained by simple mechanisms including either the initial or final stages alone. Considering the complexity of the QGP-like phenomena in a small collision system, the best way to investigate the origin of the collectivity, or maybe the only possible way to pin down the origin of the collectivity, is to do the analysis with multi-observable analysis. In this thesis, the full Run 2 data sample with the ALICE detector at the LHC is used, providing the most precise measurements. Multiple observables are measured in pp, p--Pb, and Pb--Pb collisions. The measured observables include multiple particle correlations, cumulants, flow coefficients, symmetric cumulants, and correlations between flow and mean transverse momentum. Various methods for suppressing short-range correlations have been applied. The measurements probe various mechanisms within the collisions. The initial stage of the collision is examined using flow coefficients $v_n$ and nonlinear flow coefficients $v_{4,22}$ and $\chi_{4,22}$. Moreover, the event-by-event properties of the initial stage are assessed through measurements involving NSC(3,2), which address the initial eccentricity correlations. The initial size and shape correlation is investigated through the $\rho(v_2^2,[\pt])$. The flow coefficients, the nonlinear flow $v_{4,22}$, and the NSC(4,2) are also influenced by the system's evolution. The $\rho_{4,22}$ addresses the event plane correlation. Observables such as $v_n\{2\}$ and $\chi_{4,22}$ have shown sensitivity to variations in the $|\Delta\eta|$ separation. Conversely, other observables like $c_2\{4\}$, $v_{4,22}$, and $\rho(v_2^2,[p_\mathrm{T}])$ are less sensitivity to non-flow contamination. The measurements impose constraints on the model and shed light on the system's origins, aiding in the building of a comprehensive model that describes the collective phenomenon in both large and small collision systems.