Table 2 Comparison of concepts communicated by the DEM and by many tree diagrams

Affordances of tree-based diagrams that suggest important aspects of evolutionary theory, and which are retained in the DEM

• Common descent differentiates pre-evolutionary and evolution-based descriptions of the Natural System. While tree-based diagrams suggest common descent by the overall structure of the tree that connects different groups, the DEM suggests common descent in the branching pattern of movement of the dots that split from a central origin (Fig. 5), and then move across the map space in a radial pattern. In tree diagrams, the connection between groups remains in the image the entire time the viewer is looking at it, while in the DEM the connection between groups is more ephemeral, and based on movement rather than a persistent visual connection.

• Cladogenesis and anagenesis are suggested in tree diagrams by the branching pattern of the tree and growth of the limb from the branch point, respectively. In the DEM, movement illustrates these evolutionary patterns. Cladogenesis is represented by dots splitting and moving apart, and anagenesis is represented by the movement of dots across the map space. As with shared descent, this movement-based mode of representation primarily provides visual affordances for cladogenesis and anagenesis as the viewer is interacting with the visualization.

• Continuity from ancestral to descendant species is a concept that is present in evolutionary trees, but that may not be interpreted correctly. Tree diagrams show viewers a set of taxa connected by branch points representing hypothetical common ancestors. In the DEM, continuity is to an extent depicted more strongly by the motion of continuous dots across the map space. One important point is that the dots might suggest continuity too strongly because they do not change color or shape over time, and might therefore suggest to viewers that avian orders have retained the same features over time.

• Time’s passage is an important aspect of the evolutionary process. Tree diagrams, however, can be interpreted as showing either the evolving pattern of life over time, or as showing a hierarchical arrangement of groups within a single time period. In the DEM, time is tied to the visualization as the viewer advances or goes backward through the animation.

Conceptual constraints found in many tree diagrams that limit evolutionary understanding or contribute to misunderstandings about evolution, and which the DEM was designed to avoid evoking

• Differentiation between trunk and branches in tree diagrams may obscure the large-scale branching pattern of evolution, simplify deep evolutionary history, suggest that evolution in the distant past occurred via different mechanisms than it does today, and suggest a direction or “main line” of evolution. The DEM avoids the distinction between trunk and branches by depicting all the groups on the map in the same way, as uniformly sized dots. There is no “main line” of evolution on the map; all groups radiate from the center so as to avoid a predominant direction of movement.

• Upward progress can be suggested by several elements in tree diagrams, thus fostering misconceptions about teleology and directed evolution. On the DEM, all groups are labeled with rollover text, and each group is highlighted with additional information at least twice. The map avoids differential resolution by including similar amounts of detail about groups across the span of the diagram. The directional and hierarchical aspect of trees is also avoided in the DEM by animating the visualization in a radial pattern.

Concepts that are not well visualized in either tree diagrams or in the DEM

• The unit of evolution is the clade, or ancestor–descendant group, rather than the individual. The DEM, as well as many tree visualizations, focuses on the large-scale pattern of evolution, so the changing genetic composition of groups of organisms is not visually apparent. Viewers may infer that each dot (on the map) or branch (on a tree) represents an individual, rather than a group, and this inference can contribute to the misconception that evolution is directed in a specific direction.

• The causal forces that shape the pattern of evolution are also not well represented in either evolutionary trees or the DEM. While text in the visualization does describe some evolutionary processes, the pictorial elements of the map do not exemplify these processes by themselves. As with the previous element, the focus on pattern can suggest that evolution occurs in a directed, rather than stochastic manner.

• The pattern of evolution in the DEM depicts the evolution of one group of organisms through splitting and diversification from a single origin to the present day. This pattern does not capture the complexity of the evolutionary pattern that includes extinctions that “trim” the tree over time. Extinction is largely addressed within the descriptions of individual present-day orders, rather than by adding dots for orders that are now extinct.

Concepts visible in the DEM but not in tree diagrams, which may be conceptual constraints suggesting evolutionary misconceptions

• Spatial movement through the real landscape is implied in the DEM by the dots that move across the map space. One of the events that can trigger cladogenesis is the physical separation of populations, which may then differentiate and form different species. However, spatial movement is not the only way in which new species form. Similarly, anagenesis is not necessarily tied to movement across a landscape. The spatial movement of dots across the screen may, therefore, overemphasize the importance of physical movement in the real world in shaping the pattern of evolution.

• Geographical location is primarily suggested in the DEM in the final disposition of orders across the map space. For example, viewers might infer that the map space is oriented similarly to common projections of world maps, and that perching birds (located in the upper left corner) are therefore found primarily in Alaska (which is located in the upper left of both North American and world maps used in the United States). While there are no geographical outlines on the map space, and viewers are told that it corresponds to a virtual genetic space, this might lead to misconceptions.