Congruity and dissonance
As explained above, attitudes, stereotypes, and values contribute to intentions and actions throughout life. Although individuals, in principle, can draw attention to their prejudices and sometimes change their opinions, attitudes and values are based on automatic information processing to a large degree. In the following, I use the principles of consistency and congruence (e.g. Festinger, 1957; Gawronski & Strack, 2012; Osgood & Tannenbaum, 1955) to show that changes in attitudes are based on classification rules (e.g. similarity judgements) and that comparisons between attitude objects (e.g. preferred coping strategies) are based on similarity perceptions (Dohmen et al., 1989; Sjbberg & Thorslund, 1979). I devote attention to consistence theories because they elaborate mental phenomena that contribute to our understanding of how we pursue goals and cope with discrepancies (e.g. stress, critical life events).
How are evaluations and similarity perceptions interconnected when individuals intentionally pursue goals? As a rule, individuals have unfavourable, neutral, or favourable attitudes towards objects (e.g. assertions, prejudices, preferences). Simple assertions regarding attitude objects are, for example, ‘Asian painting is good’, ‘I don’t like alcohol or cigarettes’, ‘I don’t mind horoscopes’. Given the heterogeneous views individuals have towards attitude objects, it is not important for the next step whether the specific content is appropriate or desirable. Rather, I would like to highlight here the dynamics of expressed objections, the incongruity that we feel when, for example, self-defining attitudes are insulted. Research on cognitive consistence has a long tradition in psychology. I focus on a few aspects that contribute to the understanding of how regulation of aspiration levels works.
According to Festinger’s (1957) theory of cognitive dissonance, affective states are strongly tied to cognitive representations. Inconsistent cognitions (e.g. somebody contradicts important subjective beliefs) produce aversive feelings of dissonance. Cognitive consistency is a fundamental principle of information processing linked to our motivation to change cognitions and behaviour. Festinger (1957, p. 7) stated illustratively: 'If a person were standing in the rain and yet could see no evidence of getting wet, these two cognitions would be dissonant with one another because the person knows from experience that getting wet follows from being out in the rain.’ What would you say after that? This example is possibly convincing because the experience of getting wet in rain is so common or well-accepted that we feel that everybody would approve. According to cognitive dissonance theory, tension arises when one is simultaneously aware of two inconsistent cognitions: we feel pressure for change. Dissonance theory pertains mostly to discrepancies between attitudes and behaviour.
Osgood and Tannenbaum (1955) investigated basic assumptions about human thinking. The principle of congruity denotes a thinking tendency according to which changes in evaluation are in the direction of increased congruity with the existing frame of reference. As the authors explain, it is possible to have logically incompatible attitudes towards objects without any stress, as long as the incompatibles are not brought into association. The issue of congruity arises whenever individuals receive a message which relates objects via an assertion. To the extent that the evaluative location of a particular qualifier differs from that of the thing qualified, some pressure towards congruity exists. The nature of an assertion and the existing attitude towards an object prior to reception of the message should be taken into account. In a ‘simplest of states in which human thinking operates, sources we like should always sponsor ideas we like and denounce ideas we are against, and vice versa’ (p. 44).
A prerequisite for (mis)trust, persuasion, and social interaction is that cognitive objects can be brought into association with one another, and that the cognitive arguments and the source can be derogated (Tannenbaum et al., 1966). In everyday life, we do not usually notice the degree to which we act on the basis of similarity judgements and tend to first recognize discrepancies when they are expressed in pointed or polarized form. Extreme judgements are simpler to identify than finely discriminated judgements of degree (see polarization and pressure towards all-or-nothing judgements along the evaluative dimension; Osgood & Tannenbaum, 1955). Simplicity and polarity are components of attitude formation that can be better observed in evaluations of others than in evaluations of ourselves. According to the authors, judgemental frames of reference tend towards simplicity, and less polarized attitudes change more than more polarized attitudes.
Preference and similarity judgements
What do we do when we entertain opinions and are convinced of our attitudes? We compare and classify attributes which are associated with our attitudes within a frame of reference. Perception and, as we will argue, preferences and intentions are interconnected with similarity judgements (Estes et al., 2012; Simmons & Estes, 2008; Tversky, 1977). At a closer look, it is astonishing how often we compare and evaluate ordinary things using categories of similarity or difference. We classify figures, cars, countries, friends, musical pieces, etc. according to categories of similarity and evaluate them (see the Gestalt principle of proximity, but also elections, or situations such as ‘likes’ on Facebook that ask for or provoke our approval). In the following, I attempt to illustrate these tendencies as well as, hopefully, some of the underlying mechanisms.
The component theory of attitude objects (Dohmen et al., 1989; Feger, 1979) attempts to account for global preferences for attitude objects and describes how individuals relate attitude objects (e.g. preferences for political parties, professional orientations, eating habits) to their components (associated properties, qualities, categories evaluated by the individual). In accordance with the consistency principle, the component theory assumes that global preferences for attitude objects can be predicted by their components, which in turn can be attributed to global attitude objects to a greater or lesser degree and evaluated more or less positively by the individual.
Does the principle of consistency hold true for individuals’ preferences? In the following study, I was interested in whether global preferences for attitude objects can be predicted by the evaluated components attributed to them. A total of 280 individuals were asked about their attitudes in one specific domain (e.g. musical genres, life themes, eating habits, vocational preferences). Specifically, I examined
- 1. Global preferences for attitude objects (in the case of musical preferences): ‘How much do you like classical music, jazz, rock music, ...)?
- 2. Components: I presented a number of components (e.g. rhythm, improvisation, creativity, good performance, distraction, formal structure, etc...). Participants were asked to what degree they liked each of the components and to what degree they believed that the components were characteristic ofjazz music, classical music, etc. ... The hypothesis assumes a correspondence (similarity) between the global preferences towards the objects and a measure derived from the components. This implies that individuals regard those components which they evaluate very positively as being very characteristic of the ideal objects. Those components which are evaluated very negatively are not seen as characteristic of the ideal objects. I computed derived similarity measures of the degree to which individuals like the components and attribute them to a specific musical genre (strength of attribution). To prove the assumption, the correlation between general preferences for attitude objects and attributions of components to the ideal object were computed. The first part of Figure 4.1 depicts the values from two subjects to illustrate the correlation between component structure and global evaluation. Subject A’s global evaluation of musical genres can be predicted by the component structure. The preferences for those components attributed to rock and pop music are strong, whereas the components attributed to jazz and classical music are less preferred. In contrast to subject A, subject B’s global preferences for musical genres cannot be predicted by the component structure (see Figure 4.1a). Figure 4.1b displays the correlation between the valued component structure and global preferences for attitude objects for 280 subjects, each of whom had been presented with one of five attitude domains. The average correlation across subjects and attitude domains was r = 0.41***.
Components: Strength of attribution and valence
FIGURE 4.1 Global preferences of attitude objects as a function of component structure. Each line represents one individual. Strength of attribution: Preference for components attributed to ideal attitude objects.
A second assumption of the component model is that the attribution of the same components influences the global similarity judgements. The theory assumes a strong relationship. As expected, on average, the similarity profile significantly predicted the global judgements of similarity (r = 0.45***; see Figure 4.1c).
As would be expected from the component model, there is a strong relationship between the strength of attribution and evaluation. Dohmen et al. (1989) provided evidence that a change in the judged similarity of objects was accompanied by a change in the component-derived structure. Although individuals differ in what they value, similarity comparisons and valences interact. Attitude change is associated with both, judged similarities and component structure. The component model draws attention to the cognitive-perceptual structure of attitude regulation and demonstrates how discrepancies and valences are regulated, that is, that we make comparisons and change the attribution and valence of the associated components. As Dohmen and colleagues argue, some components contain an explicitly more or less intentional orientation. Since adults face many age-related losses which can threaten their goals and render them unattainable, processes of reorientation become important, but these presuppose that valences can be redirected towards alternatives. These accommodative shifts are sometimes not easy to achieve or execute, but would be useful in situations where we cannot attain what we want and need to change priorities (elective and loss-based selection; Freund & Baltes, 2002).
The component model illustrates the processes involved in dynamic shifts and difficulties that arise when we try to change goals and life projects intentionally. Normally there are many facets that we can acknowledge and that contribute to a goal or life project. Given that our attention is directed towards a diversity of components, the principle of congruity or the need for consistency becomes visible. Similar results have been used as an empirical basis to illustrate immunization tendencies of the self (Greve & Wentura, 2003). Individuals are able to protect their self-concept when faced with failure or loss. They can reduce self-esteem problems and identity threats and adapt action strategies to accommodate on the level of self-evaluation and self-representation:
- • On the level of self-evaluation by adjusting the valence of components associated with self-defining attitudes. One can change the preference for a political party or the value of components (e.g. downgrade the importance of parts of a manifesto that is opposed to one’s opinion).
- • On the level of self-representation by changing the semantic structure of attitudes. This can be done by attributing new components or by denying their critical relevance for specific attitude objects.
The present results are in line with studies showing that when information does not fit, individuals tend to select options which are consistent with their established beliefs. Despite the assumption of immunizing self-concept protection (Brandtstadter & Greve, 1994; Greve & Wentura, 2003), associated components might contribute to our understanding of how intentions develop. For instance, global preferences consisting of many attributed components probably remain stable if one component is negatively valued, compared with less complex attitudes. The multidimensionality of the component structure of attitude objects brings us to the issue of perception differentiation. Using research on human perception I will illustrate how similarity comparisons are influenced by situational factors.
Perception and context effects
During the last 50 years, several studies in experimental psychology have demonstrated that decisions, comparisons between components (aspects of attitude objects), or similarity judgements are dependent on situational changes and context effects (de-Wit & Wagemans, 2015; Estes et al., 2012). In my view, these studies help to understand some of the principles of intentional processes in that choices and decision behaviour also depend on changes in context. I focus in the following on processes of perception, with the possibility in mind that we possibly can use them to learn about ISD and to differentiate between schemas of perception.
Harry Helson investigated processes of adaptation to visual stimuli and provided evidence on how context affects influence judgements (Helson, 1971). According to the adaptation-level theory, subjective judgements are necessarily relative to the prevailing norm or adaptation level. Exposure to earlier stimuli serves as a frame of reference by which later stimuli are judged. Helson described a study in which participants were asked to estimate the number of dots. Prior to estimating the dots, participants were shown an anchor of four dots, an anchor of 13 dots, or an anchor of 32 dots. The estimates of the target dots were influenced by the number in the anchor. A small anchor resulted in an increase in the number of dots perceived, whereas the large anchor resulted in a decrease. Adaptation-level theory has been applied to phenomena of motivation, affect, and well-being regulation. There is also some research on adaptation to stress that emphasized that there is a large variation in satisfaction. For instance, studies on the impact of critical life events provided evidence that critical events do not affect the level of well-being for more than a few weeks or months (Luhmann et al., 2012). A widely known study by Brickman et al. (1978) demonstrated that the average well-being of recent lottery winners was comparable to the well-being level of a control group. Phenomena such as these can be explained by the ability to adjust personal standards such as norms and evaluation criteria.
Sjoberg and Thorslund (1979) have applied the idea of context adjustments to the cognitive dynamics of similarity judgements. If we compare nationalities (each nationality defines one class), the set of all nationalities defines the universe.
For example, similarity judgements of two nationalities are influenced by the following principles:
- 1. The larger the universe (the number of classes, e.g. nationalities) is, the larger is the judged similarity of two individuals from different nationalities.
- 2. The second principle holds that similarity judgements depend on the number of objects or components contained in the class (defined by two comparison stimuli): The larger the common class (e.g. the class of Northern Europeans), the smaller the similarity is (see Sjbberg and Thorslund, 1979 for further explanation). The common class is defined as the class of objects formed to contain both of the two objects to be judged in terms of similarity (e.g. similar languages, culture, etc.).
Suppose your task is to judge the similarity between two Chinese people in two different places. According to Sjoberg and Thorslund, on a street in Peking, a smaller similarity rating would be expected than on a street in Sweden. The authors provided evidence for their assumptions by manipulating context factors. The estimated similarity in homogeneous pairs (i.e. pairs of string instruments, e.g. banjo, violin, harp, electric guitar) was increased when a heterogeneous stimulus (clarinet) was introduced into the comparison list and decreased when a homogeneous stimulus (double bass) was added. Thus, if we are able to broaden or limit the associative context, then the similarity judgements (a basic mechanism of attitudes and related intentions) should change as a consequence.
From a short-term perspective on intentional development, it seems useful to highlight the role of perception, because we believe that these processes are relevant in concrete decision situations. Indeed, there is evidence that intention and perception of objects in the surrounding are basically interconnected (see concepts such as salience, valence, or importance). Wendell Garner’s studies (e.g. Garner, 1966, 1978) are not only important for perception of visual stimuli, but also provide useful guidelines for research on motivation and goal-related behaviour. In his article ‘To perceive is to know’, Garner (1966) describes visual and auditory perception of patterns as a cognitive process involving comprehending, organizing, and knowing - as ‘an active process in which the perceiver participates fully’ (p. 11). His study showed that individuals differ in preferences for stimuli attributes at a basic level of perception (i.e. liking the simplest patterns; stimuli were cards on which two dots had been placed).
How well we discriminate between stimuli attributes is related to the interface between interests and ability, and thus remains a challenging question (Meumann, 1908). Of course, we can learn to discriminate to a certain degree, but it is not necessarily exactness that determines this degree: Categories can be more or less unambiguous and can pertain to problems such as whether or not the cathedral of Milano actually consists of gothic elements from the north. Nonetheless, we can learn to differentiate between elements of gothic style and others. Given such an interest, perception is closely related to intentional striving and to sensory processes. Perceivers, however, do not only intentionally select the structure (e.g. values, specific goals, desired states) to which they will attend or react. The important point is that goals are products of automatic processes of classification as well, and it seems that we can learn about goals if we learn more about the processes of similarity perception.
Interindividual differences in processing contextual information have been found in perceptual styles (Milne & Szczerbinski, 2009). The tendency to group figures was identified as a basic law of perception by Gestalt psychology. Studies demonstrated that individuals have schemata which they use to organize their surroundings. Some of them tend to be highly influenced by the context of the visual scene, whereas others are more able to perceive elements independently from context (e.g. field dependence; Witkin & Goodenough, 1977). Such tendencies are not limited to visual patterns or stimuli. People differ in their tendencies to process self-relevant information autonomously or as related to other people. Independent self-definitions involve the tendency to process stimuli unaffected by the context, whereas interdependent self-definitions imply attention to the given context (Hannover & Kühnen, 2003).
What can we learn from this section about ISD? I have focused on the interplay between processes of classification and perception because I want to show their role as a prerequisite when individuals act, interpret, and give meaning to their lives. Although people can draw attention to attitudes and goals, the difficulty for intentional self-regulation is that processes of categorizing and similarity comparisons are basic and, in part, automatic processes with limited potential for control. Nonetheless, they are interconnected with preferences and, in this function, relevant for the development of intentions because they contribute to the regulation of valence. This touches upon our challenge to understand the rules with which individuals perceive and order aspects of their surroundings and how they categorize and build dimensions and regulate standards or achievement norms. Since individuals are not simply machines that all respond to stimuli in the same way, but instead react to stimuli on the basis of prior experiences, we turn to the mediating role of memory. Processes of learning and memory are important for the understanding of attitudes, preferences, and personally important goals.