In Dominions, all non-Holy spells are organized into one of seven schools of magic: Conjuration, Alteration, Evocation, Construction, Enchantment, Thaumaturgy, and Blood Magic (not the path). Where a spell's path represents the power source, the school a spell belongs to ostensibly represents the methods used to manipulate that power. In order for their mages to cast more advanced spells and forge more powerful magic items, the player has to invest research into these schools. Each school has nine levels of research.

The order in which these schools are researched is decided through the Research Screen, which can be accessed by pressing F5 in-game. Simply left-click on the name of a school to queue up one level of research into that school. Any research points left over after the queue is finished will be assigned to an unfinished school, starting from the top of the Research Screen.

* Placeholder: needs prose etc.
* Test interaction with magic/drain and non-mage researchers? Which of these benefit/are penalised by them?

Only commanders with the research ability are capable of contributing to their nation's research. A unit with the research ability will have a icon on their ability bar, next to their magic paths, with a number indicating how many points they contribute to the pool. Troops cannot research.

Typically, units require some amount of magic paths in order to be able to research. All mages (with a few exceptions) have a base research ability of 5, and each level a mage has in one of the eight non-Holy magic paths increases their research ability by 2. Experience stars also increase a mage's research ability by 1 each. However, there are several other factors that might change how much a unit contributes:

• Units with the "Adept Researcher" or "Inept Researcher" abilities, receive an increase or decrease, respectively, to how many research points they generate. The independent, recruitable Wolf Tribe Shamans are an example of an inept researcher; the Great Sage pretender is an example of an adept researcher. Regardless of penalties, the minimum amount that a mage can contribute to Research is 1.

• The scales of the province a unit is researching in influence their output. Each point of magic scales will add a research point to each researcher in that province, if the province also has friendly dominion. Likewise, each point of drain on the scales will reduce research by one, regardless of dominion status. Units with the "Mundane researcher" ability, most notably the Smiths of MA Ulm, are immune to this research penalty, as are the rare non-mage units that have research ability (e.g. Iassacharite Sages).

• There also exist "Inspiring Researchers" - these commanders will increase the research points of all other mages in the province, making them especially powerful when stationed in research hubs such as your capital. For example, an Inspiring Researcher +1 in a province with 25 other mages will be adding +25 research points to your pool. A number of pretenders have this ability - for example, the Great Sage.

• There are magical items that increase research points when equipped by a mage. There are several of these scattered throughout the different paths, the non legendary are the 5 Owl Quill, the 5 Imp Familiar, the 10 Skull Mentor, and the 5 Lightless Lantern.

Each nation begins the game with an amount of RP based on its pretender's magic/drain scale. This free RP is not influenced by the age the game is set in and cannot be disabled.

If the game is using the "random start research" option (the default), the RP is allocated to random schools in 50-RP parcels. The default total is 150 RP; each level of magic adds 50 more RP and each level of drain subtracts 50 RP.

If the game is not using random start research, the RP is divided evenly between each school. Due to rounding, the total amount of RP will be slightly different from a random start research game:

The amount of research points that must be invested in a school in order to reach each level is dependent on the game's settings. Most multiplayer games use the standard research difficulty setting: