T-spherical fuzzy sets, the direct extension of fuzzy sets, intuitionistic fuzzy sets and picture fuzzy sets are examined in this composition, and a mathematical examination among them is set up. A T-spherical fuzzy set can demonstrate phenomenon like choice utilizing four trademark capacities indicating the level of choice of inclusion, restraint, resistance, and exclusion, another example of such situation is that human opinion cannot be restricted to yes or no but it can be yes, abstain, no and refusal. T-spherical fuzzy set can deal the said situation with a boundless space. With the assistance of some mathematical outcomes, it is talked about that current similarity measures have a few drawbacks and could not be implemented where the data is in T-spherical fuzzy mode. Thus, some new similarity measures in T-spherical fuzzy environment are proposed, with the assistance of certain outcomes, it is demonstrated that the suggested similarity measures are generalization of current ones. Further the proposed similarity measures are applied in pattern recognition with numerical supportive examples. The maximum spanning tree clustering algorithm has been extended into T-spherical fuzzy context and supports our theory with numerical examples. A parallel investigation of fresh and existing similarity measures have been made and some of the benefits of designated work have been discussed.