Direct Manipulation


Direct manipulation has been beneficial in the realm of interface design for the last two decades. The benefits have prompted designers to reduce indirection and change domain objects to first-class interfaces. Before the invention of direct manipulation, Human Computer Interaction (HCI) largely depended on a conversation metaphor such as command-line interfaces. This interface was abstract and indirect hence the new paradigm improved human interaction with computers. This paper examines the principles of direct manipulation, benefits and weaknesses of direct manipulation as it relates to command-line interfaces, and its usage in video game control.

Principles of Direct Manipulation

Direct manipulation is linked with three major principles. First, there is constant demonstration of the object of interest demonstrated in a permanent visual feedback. Second, it is characterized by physical action rather than intricate syntax. Third, there is spiral or layered method to learning, which allows usage with minimum knowledge (Shneiderman, 1982; Shneiderman, & Maes, 1997). In video games, users are able to spontaneously initiate control using knobs and buttons, which basically map out the character’s movement on-screen. Joysticks and buttons are the common de facto methods of input in direct manipulation interface in video games. The objects on-screen are visually indistinguishable from real-world objects, which they represent (Dix, 2009).

For example, an avatar represents the player in the virtual world. Another example of the application of the direct manipulation principles is that users’ actions are continuously displayed on-screen. For example, a gun used by an avatar serves similar functions to that in the real world. Lastly, there is immediacy of user actions. For example, when a player presses shoot button in a one-person shooter game, the avatar initiates a shooting action accordingly. For this reason, actions executed by a player are instantaneous.

Video game-type interfaces

First, video games interfaces are limited in their ability to represent reality. For example, such interfaces cannot create sensations of warmth, weight, taste, and smell. Presently, video game interfaces can only activate the senses of sight and hearing. Therefore, video games are not effective in emulating reality. Second, every command is routed through a user’s thumb to the physical character. Humans operate by different muscles which control activities as opposed to one thumb used in video game interfaces. For this reason, the interfaces limit real-life application due to the limited nature of their controls. Third, video games are unable to inhabit a narrative or designed environment. The player and the avatar can only interact through haptic, visual and audible means. Such interaction immerses the player a fictional world that cannot truly be represented in a real world. An imaginary divide, which separates the player from the virtual world, known as the fourth wall is a barrier to complete immersion. This means that video game interfaces are not effective in breaking this fourth wall, hence a player cannot fully immerse in a game to engage in real-world experiences (Fagerholt & Lorentzon, 2009).

In a video game interface, the user, usually a person interacts with a system, often an electronic device such as computer or game console, via a controlling mechanism such as keyboard or gamepad. In general, the controlling mechanisms have largely been buttons and joysticks. Joysticks correspond to directionality and/or spatial movement of objects while buttons usually correspond to user’s actions (Fagerholt & Lorentzon, 2009).

Benefits of Direct Manipulation

When assessing the benefits of direct manipulation against those of command-line interfaces, the example of travelling in a car sums up advantages. Using direct interface, a user drives the car by manipulating the pedals and steering wheel. The effect is that the car responds according to the user’s actions through a continuous visual feedback. If the user makes a mistake while driving the car such as turning suddenly on a sharp corner, the user can rapidly realize the mistake and initiate corrective measures. On the other hand, with a command-driven interface, a user is a passenger in the car issuing directions to a stranger (Dennehy, nd).  In this case, a user relies on the actions of a stranger who, if they possess inadequate skills may not fulfill the purpose of the journey.

This example shows that direct manipulation yields immediate, incremental and reversible results that are continuously visible by the user. This approach improves the impression that the user is executing the task. In addition, the user is in control as opposed to the computer responding to user requests while the user powerlessly waits to know if the commands are correct akin to the command-line interfaces (Dennehy, nd ).

Since direct manipulation is interactive in nature, the paradigm has benefits of visual representations related to learning retention and speed. It exploits these benefits leading to a system whose process is easy to understand and hard to forget. Another benefit is that because users do not need to memorize complex syntax, they can use analogical reasoning instead thus eliminating the likelihood of errors. However, when errors are made, users can easily correct them due to the ability of the paradigm to reverse actions through a constant visual feedback (Javed, Elmqvist, & Yi, 2011).

Another advantage of direct manipulation is that it is flexible to user specifications. The system allows users to customize functions compared to command-line interfaces. Therefore, users can eliminate complex functions or customize them to suit their preferences. In gaming, for example, users can change the functionality of the gamepad so that certain buttons perform different functions compared to the default settings. In addition, users can adjust some functionality such as speed, color, volume, and size accordingly. Direct manipulation enables users to make changes to the system without the need for system alerts and notifications common in command-line interfaces.  Such alerts often interrupt user experiences. For this reason, direct manipulation enables users to have a smooth experience without interruptions (Heer & Shneiderman, 2012).

Weaknesses of Direct Manipulation

There are various challenges associated with direct manipulation, especially for video content. Direct manipulation is criticized for deficiencies such as access, manipulation of multiple objects and intangible properties. The problem of access is based on the difficulty of direct manipulation to direct distant, small, or attribute-rich aspects under high precision, limited space, and high density. When an object is too small, users may find it difficult to grab or select it for resizing. The problem of access is compounded if there are numerous objects located in one area or in a distant area or outside of the screen in ubiquitous applications. This problem may be solved by zooming but it is subject to extra cognitive load and effort. Furthermore, objects that are inherently rich in attributes may incur the problem of access. Attributes may not be manipulated because of limited visual space to show controls. Visual representation may also be affected when interface controls are displayed (Heer & Shneiderman, 2012).

Interaction with multiple objects is cumbersome using direct manipulation, especially when they have different values such as font size. This means that the cumulative attributes of the chosen objects cannot be correctly presented, hence ambiguity in representation. As such, the predictability associated with the direct manipulation paradigm is lost because of such ambiguity in visual representation. Another challenge of this interface is manipulating properties of intangible objects. An example is abstract attributes with no natural visual aspect, such as spacing between many objects. Other examples include alignment, layout, and sorting order of objects. Manipulating such properties may induce visual clutter in applications (Tudoreanu & Hart, 2004).