Chapter 5 Order and complexity

Status: general ideas relation aesthetics and Prägnanz presented as a conference poster on VSAC 2018 (Van Geert & Wagemans, 2018a); OCTA manuscript in preparation

As discussed in Chapter 2, Prägnanz has long been related to beauty and aesthetic appreciation. In this chapter, I will focus on how order and complexity relate to both Prägnanz and aesthetic appreciation. In what follows, I will define complexity as the quantity and variety of information and order as the structure and organization of information (Van Geert & Wagemans, 2020). This applies to stimuli as well as to percepts.

Order and complexity can be taken as partially opposite tendencies, as complexity reduces order and order reduces complexity, but they also complement each other in their relation to aesthetic appreciation (Arnheim, 1966; Van Geert & Wagemans, 2020). Whereas order needs complexity to show its structuring and clarifying potential, complexity needs order to be understood and appreciated (Van Geert & Wagemans, 2020). As order and complexity are interrelated, it is important to study them together when investigating their relation with aesthetic appreciation (Van Geert & Wagemans, 2020).

How do order and complexity relate to Prägnanz? A right ‘balance’ or ‘combination’ of order and complexity cannot only increase the aesthetic appreciation of a Gestalt, it can also increase the ‘goodness’ or Prägnanz of the Gestalt. In my opinion, aesthetic appreciation is not related to Prägnanz as such, but to an increase in Prägnanz. This view is similar to other perspectives on aesthetics that describe aesthetic appreciation as related to (the promise of) an increase in order (Muth & Carbon, 2013) or a reduction in prediction error (Van de Cruys & Wagemans, 2011).

For both Prägnanz and aesthetic appreciation, I expect a minimum level of order as a prerequisite. One could say that this requirement is similar to the requirement of some degree of regularity for a Gestalt to be called prägnant stated by Rausch (1966). In addition, I expect the optimal complexity level to depend on the prevailing conditions (i.e., individual, context, stimulus, and their interactions). Preliminary evidence in line with these hypotheses exists: the relation between order and aesthetic appreciation was more stable across both participants and cultures than the relation between complexity and aesthetic appreciation (Van Geert & Wagemans, 2020, 2019).

Earlier research mainly focused on the separate influence of order and complexity on aesthetic appreciation, or on rather specific types of order (i.e., balance or symmetry). Progress has also been limited by the lack of an easy way of creating reproducible and parametrically controlled stimulus sets, including both order and complexity manipulations. To be able to investigate the relations between order, complexity, goodness, and aesthetic appreciation in a more coherent and parametrically controlled way, I am currently developing the Order & Complexity Toolbox for Aesthetics (OCTA), a Python toolbox that allows manipulating order and complexity on several dimensions (e.g., position, shape, size, color, orientation) independently.

An additional benefit is that OCTA will provide an open and reproducible way of creating stimuli for aesthetics research on order and complexity. Furthermore, OCTA aims to be accessible for all: researchers with programming experience can use the available functions or specify additional functionality tailored to their own wishes, whereas researchers less familiar with programming can use the point-and-click interface that is provided (Figure 5.1).

Figure 5.1: The OCTA Shiny application. Try out the app in the interactive version of this document.

Together with a master’s student, I am currently preparing a first empirical study using stimuli, varying on multiple dimensions of order and complexity, which were created with this toolbox.

Future research ideas

5.1 OCTA, meaning, and Sinnprägnanz

OCTA can also be an ideal tool to investigate meaning as a facet of Prägnanz (cf. Rausch’ seventh Prägnanz aspect described in Chapter 2). For example, one could compare aesthetic appreciation for and goodness of a display with related objects and a display with unrelated objects. In the display with related objects one could distinguish objects with the same and with complementary functions. In this respect, Metzger (1941, p. 107) writes: “The visual grouping of collections of known objects is usually determined by essential properties like their purpose of use rather than color, size, shape or material, where the most preferred grouping is that into groups of things which complement each other rather than into groups of the same function”.

5.2 OCTA, working memory capacity, and Prägnanz

OCTA can also be used to clarify the role of an individual’s working memory (WM) capacity on Prägnanz, and indirectly on memorability and aesthetic appreciation. The relation between complexity and aesthetic appreciation will depend, amongst others, on the processing capacities of the individual. Hence, WM capacity of the particular individual at that moment in time will increase the Prägnanz of more complex percepts and moderate the relation between complexity and aesthetic appreciation. Preliminary evidence supports that a positive relation exists between Openness to Experience (i.e., a personality trait related to intelligence, and possibly WM capacity) and a preference for complexity (e.g., Van Geert & Wagemans, 2019).

Predictions are that individuals with a higher WM capacity would show higher accuracy in an incidental (i.e., unannounced) recognition memory task and would more often prefer more complex stimuli than individuals with a lower WM capacity.

5.3 Individual differences in preference for order and complexity

In future studies, I am also interested in whether individual differences in preferences for order and complexity are linked to other interindividual differences, e.g., individual differences in the size of temporal context effects such as hysteresis and adaptation.