The process of performing initial investigations on data to spot outliers, discover patterns, and test assumptions with statistical insight and graphical visualization is called:

A t-test is a specific inferential statistical hypothesis test used to compare the means of two groups, not a broad process for initial data investigation.

Performance analysis typically refers to evaluating the efficiency or speed of systems or processes, not the initial investigation of data patterns and outliers.

Exploratory Data Analysis (EDA) involves using statistical graphics and other data visualization methods to summarize the main characteristics of a dataset, identify patterns, detect anomalies, and formulate hypotheses. It is a crucial first step in any data analysis project to understand the data's structure and properties.

Link analysis is a technique used to evaluate relationships or connections between nodes in a network, which is a specialized form of analysis and not the overarching process described.