It.: Valutazione; Fr.: Èvaluation; Germ.: Bewertung; Span.: Evaluación. The term comes from the Latin vàlitus, which in turn derives from valére. The Latin verb involves the sense of “putting a value upon”, “assigning value to”.
Evaluation refers to the prototype of a complex terminological and conceptual family, including also appraisal, assessment, and appreciation. To sketch the main properties of this prototype it is useful to move from the definition of evaluation largely shared in cognitive sciences (Tesser and Martin 1996; Miceli and Castelfranchi 2000). According to this definition, evaluation represents a particular mental process of cognitive agents. A cognitive agent is an agent whose behaviour is internally regulated by goals and beliefs. A goal is a representation of the world that regulates the agent’s behaviour, so as to adapt the world to that representation. A belief is a representation of the world “as it is”, a mental state that the agent tends to adapt to the world. When agents record a discrepancy between their goals and their beliefs about the world, they act to adjust the world to the goal. Evaluations are beliefs about “what is good for what” and “why”. Unlike mere neutral beliefs, they do not simply describe objects and states of the worlds but already put them into some means/end relationship with the agent’s goals. Evaluations assign a value to such objects and states relative to the agent’s interest: these objects and states are useful or useless with regard to the agent’s goals.
In the last decades, the neuropsychological research has highlighted the complex and multifaceted nature of evaluation. The ordinary activity of evaluation is not a single process: it varies in the degree to which it is automatic and implicit or more deliberate and controlled. In this perspective, it cannot be exclusively reduced to cognitive evaluation. This kind of evaluation is declarative and explicit; it contains a judgment of a means/end connection; it is supported by some reasons for this judgment. In contrast, a large body of our daily evaluations is constituted by mere affective appraisal, a kind of evaluation based on associative learning and memory (Cacioppo et al. 2004). Affective appraisal does not involve justifiable reasons: it engages an automated activation of a positive or negative affective response associated with a previously conditioned stimulus. It is unconscious, implicit, and intuitive. It can be primed by subliminal stimuli; it is very fast and anticipates high level processing of the stimulus (Zajonc 2000).
Affective appraisal can co-occur with cognitive evaluation and it can direct and motivate cognitive evaluation, especially when the control system is not activated (Slovic et al. 2002). From this point of view, ordinary information processing is dynamically based on iterative sequences of evaluative processes through which stimuli are interpreted and reinterpreted in light of an increasing set of contextually meaningful representations. Fast evaluations are likely to be determined by the most accessible properties of the stimulus (Kahneman 2003). Depending on motivation and opportunity (Fazio and Towles-Schwen 1999), additional iterations can transform these relatively automatic appraisals into more accurate evaluations (Cunningham and Zelazo 2007).
Recent research has also focused on evaluation as a crucial component of complex mental states such as attitudes and emotions. Attitudes represent relatively stable sets of evaluative representations of a stimulus. They can be activated by rapidly occurring appraisal processes or by more complex and consciously constructed judgments (Dovidio et al. 1997). A very interesting finding shows that fast and uncontrolled evaluative processes sometimes reveal an unintended activation of implicit biases, stereotypes, and prejudice in contrast to self-reported explicit attitudes (Rudman et al. 2001). According to the cognitive theories of emotions (Ekman and Davidson 1994), all emotions involve a valence component: the primary function of emotion is to provide information about how a situation has been appraised on the basis of a weighted assessment of an event’s relevance to central personal goals. This information is conveyed internally by felt experience: it serves as data for judgment and decision-making processes (Schwarz and Clore 2007). Since the interval between stimulus and emotional response is sometimes extraordinarily short, the appraisal mechanism must be capable of operating with great speed and it often happens without awareness. This automatic appraisal operates on what is given biologically and it is attributable to personality and culture. Complex emotions commonly arise from symbolic activity, reflective evaluation, and controlled regulation, created by the decoupling of stimulus and response (Ochsner and Gross 2005).
In recent years evaluation has been one of the most investigated issues both in philosophical and in scientific approaches to aesthetics. In the last two decades, analytical aesthetics has devoted much attention to aesthetic judgment and the related questions concerning aesthetic normativity, justification, and intersubjectivity (Zangwill 2003; Goldman 2006; Budd 2007). Moreover, the debate on aesthetic properties has focused on their crucial role of reasons supporting aesthetic evaluations (Matravers and Levinson 2005; De Clercq 2008). An important line of these researches has highlighted the difficulties of the “acquaintance principle”, created by the so-called “non-perceptual, conceptual art”, i.e. art that does not require a first-person experience (Shelley 2003).
Psychological and neurobiological approaches to aesthetics have mostly dealt with aesthetic evaluation (Silvia 2009). For many years, since the first studies of contemporary empirical research (Berlyne 1971), aesthetic experience has been essentially reduced to sensorial preferences and a simple feeling of liking and pleasure, usually measured with self-report items like pleasing/displeasing, like/dislike, positive/negative, pleasant/unpleasant. Accordingly, the first applications of neuroimaging to visual aesthetic experience involved a privileged position for perceptual beauty. However, in recent years the general focus of experimental studies has been deeply reoriented. At present these studies are centred on aesthetic experience conceived as an experience of knowledge (Sherman and Morrissey 2017; Pelowski et al. 2017). As a consequence of this renewed framework, new lines of investigations on aesthetic evaluation have been carried out.
The first line of research concerns the fast and automatic aesthetic appraisal. A set of experiments on images, based on eye movements and verbal reactions, suggests that people (both experts and non-experts) are able to evaluate the aesthetic quality of paintings immediately at first glance; that their evaluations remain fairly consistent across viewing time; that, unlike the typical mere exposure effect, in which familiarity tends to increase liking, the exposure itself, without intermediate reasoning, is sensitive to aesthetic value: mere exposure to bad paintings makes people like them less (Meskin et al. 2013).
The second line of research investigates the relationship between art-expertise and aesthetic evaluation. A growing body of evidence indicates that experts and non-experts differ in how they literally experience and understand art. Experts view art differently according to eye-tracking research, have more fine-grained emotional responses to art, evaluate art differently. Compared to non-experts, experts appreciate more negative, provocative, and disturbing artworks and have attenuated reactions to them, because they are less responsive to the artworks’ direct affective valence than are non-experts. Furthermore, their emotional response to valence is attenuated by their attention to other features, such as structural, stylistic, and formal properties. Art experts, more than non-experts, emphasize these properties more than content, craftsmanship, and level of realism (Leder et al. 2014).
The third line of research focuses on the role of metacognition in aesthetic evaluation (Silvia 2008; Graf and Landwehr 2017). The evaluative processes engaged by the so-called “knowledge emotions” (i.e. interest, confusion, and surprise) are metacognitive to the extent that they stem from people’s appraisals of what they know, what they expect to happen, and what they think they can understand. In particular, interest involves two appraisals: appraising a stimulus as new, complex, and unfamiliar (a high novelty-complexity appraisal) and as comprehensible (a high coping-potential appraisal). Like interest, confusion involves appraising something as new and complex; unlike interest, it involves appraising the stimulus as hard to understand. Surprise involves appraising something as novel and unexpected; the appraisal of coping potential that follows can lead to interest or confusion.
The fourth line of research stresses the deep connection between aesthetic evaluation and cognitive processing. Several experiments suggest that liking and appreciation significantly increase only after having an insight or a prolonged cognitive elaboration (Muth and Carbon 2016). Moreover, the intensity of insight shows direct influence on the degree of liking. This evidence supports a dis-fluent and dynamic conception of aesthetic evaluation. According to the classical theory of aesthetic pleasure based on fluency (Reber, Schwarz and Winkielman 2004), variables able to influence processing fluency, such as perceptual and semantic priming, stimulus repetition, and prototypicality, increase aesthetic appreciation. However, aesthetic appreciation often involves the success in establishing a new predictable pattern on a different level and so the transition from an initial state of uncertainty, associated with an unpleasant and negative affect, to a subsequent state of increased predictability, associated with an affective reward (Van de Cruys and Wageman 2011).